WO2022206633A1 - Intelligent network selection method and related apparatus - Google Patents

Intelligent network selection method and related apparatus Download PDF

Info

Publication number
WO2022206633A1
WO2022206633A1 PCT/CN2022/083216 CN2022083216W WO2022206633A1 WO 2022206633 A1 WO2022206633 A1 WO 2022206633A1 CN 2022083216 W CN2022083216 W CN 2022083216W WO 2022206633 A1 WO2022206633 A1 WO 2022206633A1
Authority
WO
WIPO (PCT)
Prior art keywords
network
service
predicted
current
attribute information
Prior art date
Application number
PCT/CN2022/083216
Other languages
French (fr)
Chinese (zh)
Inventor
窦凤辉
杨春刚
Original Assignee
华为技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Publication of WO2022206633A1 publication Critical patent/WO2022206633A1/en

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W16/00—Network planning, e.g. coverage or traffic planning tools; Network deployment, e.g. resource partitioning or cells structures
    • H04W16/22—Traffic simulation tools or models
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W36/00—Hand-off or reselection arrangements
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W40/00—Communication routing or communication path finding
    • H04W40/02—Communication route or path selection, e.g. power-based or shortest path routing
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W48/00—Access restriction; Network selection; Access point selection
    • H04W48/18—Selecting a network or a communication service
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D30/00—Reducing energy consumption in communication networks
    • Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks

Definitions

  • the present invention relates to the technical field of wireless communication, and in particular, to a method and related device for intelligent network selection.
  • heterogeneous networks are becoming more and more popular. Different types of wireless networks have great differences in coverage, service support, transmission rate, and mobility support. These heterogeneous characteristics prompt the network as a whole to take full advantage of different networks and provide users absent service.
  • the primary problem is which wireless network the user should access, which can not only meet the user's different business service requirements, but also achieve optimal system performance.
  • the selection of the optimal network generally depends on multiple attributes such as bandwidth, throughput, delay, jitter, packet loss rate, power consumption, and cost. attribute decision problem.
  • the proposed heterogeneous network selection methods mostly focus on the use of intelligent optimization algorithms, that is, the access decision is made according to the current network attribute information of each network in a heterogeneous environment. This method does not consider the network attributes after selecting the target network for access. The influence of dynamic changes of information on mobile users, network reward function and handover decision conditions.
  • existing heterogeneous network selection methods whose goal is to maximize the expected total return per connection, do not differentiate the service types of mobile users.
  • the technical problem to be solved by the embodiments of the present invention is to provide a method and related device for intelligent network selection, which can realize the selection of an optimal network and improve user experience.
  • an implementation case of the present invention provides a method for intelligent network selection, which is characterized in that, when applied to a terminal device, the method includes: determining a comprehensive service weight of the terminal device; the comprehensive service weight includes the terminal device.
  • the prior art only selects the network based on the current service of the terminal device and the current network state. If the service or network state of the terminal device changes, the communication service provided by the target network will be unstable. As a result, the terminal device needs to switch the network frequently, and because the future state of the network is not considered in the prior art, the accuracy of the network switching cannot be guaranteed, which eventually leads to poor service experience of the terminal device.
  • the current service status of the terminal device and the service status in a future period of time, as well as the current status of different networks that are candidates for the terminal equipment and the status in a future period of time are comprehensively considered to determine the current service status of the terminal equipment.
  • the terminal device can select a target network from multiple candidate networks, which can ensure the stability of the communication service, thereby reducing the frequent switching of the network by the terminal device, improving the communication performance index, and realizing the user experience. Personalize the business experience.
  • the determining the comprehensive service weight of the terminal device includes: obtaining the current service weight according to the current service attribute information of the terminal device, and obtaining the current service weight according to the current service attribute information of the terminal device;
  • the predicted service attribute information is used to obtain the predicted service weight;
  • the current service attribute information includes one or more of the current service type, the current service start time, the current service duration, the service flow, the service characteristics and the user preference.
  • the predicted service attribute information includes one or more of predicted service type, predicted service start time, predicted service duration, predicted service flow, predicted service characteristics and the user preference.
  • the terminal equipment for each network selection can obtain the comprehensive service weight of the terminal equipment according to the currently recorded information, which improves the efficiency of intelligent network selection of the terminal equipment, and the selected target
  • the network can ensure the stability of communication services, thereby reducing the frequent switching of networks by terminal devices, improving communication performance indicators, and realizing personalized service experience for users.
  • the determining the comprehensive service weight of the terminal device includes: sending the current service attribute information and the predicted service attribute information of the terminal device to a network device; the network device is a network device associated with one or more candidate networks in the M candidate networks; and receives the comprehensive service weight sent by the network device.
  • the terminal device may send current service attribute information (such as current service type, current service duration, etc.) and predicted service attribute information (such as predicted service type, predicted service duration, etc.)
  • current service attribute information such as current service type, current service duration, etc.
  • predicted service attribute information such as predicted service type, predicted service duration, etc.
  • One device in one or more network devices of the network and then the network device calculates the comprehensive service weight, which reduces the data processing capability requirements of the terminal device in the process of intelligent network selection, and the selected target network can guarantee.
  • the stability of the communication service reduces the frequent network switching of terminal equipment, improves the communication performance index, and realizes the user's personalized service experience.
  • the determining the network utility values of the M candidate networks respectively includes: determining the current network utility value of each candidate network according to the current network attribute information corresponding to the M candidate networks network utility value, and determining the predicted network utility value of each candidate network according to the predicted network attribute information corresponding to the M candidate networks, to obtain the network utility value of the M candidate networks;
  • the current network attribute information includes one or more of the current network bandwidth, delay, jitter, and packet loss rate;
  • the predicted network attribute information includes one or more of the network's future bandwidth, delay, jitter, and packet loss rate. one or more.
  • the terminal device may determine the current network utility value based on the respective current network attribute information (such as the current network bandwidth, delay, jitter, packet loss rate, etc.) of the multiple candidate networks, and predict the network attribute information (such as the future network bandwidth, delay, jitter, packet loss rate, etc.) to determine the predicted network utility value, so as to obtain the network utility value of multiple candidate networks.
  • the network utility values of multiple candidate networks can be obtained on the terminal device, which improves the efficiency of the terminal device's intelligent network selection, and the selected target network can ensure the stability of communication services.
  • frequent network switching of terminal equipment is reduced, communication performance indicators are improved, and user personalized service experience is realized.
  • the determining the network utility values of the M candidate networks respectively includes: receiving the network utility values of the M candidate networks respectively sent by the one or more network devices associated with the M candidate networks. the network utility value.
  • the terminal device may receive the network utility values of the network calculated by the respective network devices of the multiple candidate networks, and then the terminal device may select the target access from the multiple candidate networks based on the network utility values.
  • the network reduces the data processing capability requirements of the terminal equipment, and the selected target network can ensure the stability of the communication service, thereby reducing the frequent switching of the network by the terminal equipment, improving the communication performance index, and realizing the personalized service experience of the user.
  • the obtaining M comprehensive utility values based on the service comprehensive weight and the network utility values of the M candidate networks includes: based on the current service weight and each of the M candidate networks. the current network utility value of the candidate network, and a comprehensive utility value for each of the candidate networks is calculated based on the predicted traffic weight and the predicted network utility value for each of the candidate networks.
  • the current network utility value and the predicted network utility value of different candidate networks are respectively weighted with the corresponding current service weight and predicted service weight to obtain the comprehensive utility value of multiple networks. It is based on the current service state of the terminal device and the service state in the future, as well as the current state of the network and the state in the future. Therefore, by implementing the method provided by the embodiment of the present invention, the terminal device is based on the comprehensive utility of multiple candidate networks.
  • the value of the target network selected from multiple candidate networks can ensure the stability of communication services, thereby reducing the frequent switching of networks by terminal equipment, improving communication performance indicators, and realizing user personalized service experience.
  • the method further includes: acquiring first history record information of the terminal device; the first history record information includes application access records, content request preferences, service start time, service traffic , one or more of service characteristics and usage duration; and predicting according to the first historical record information to obtain the predicted service attribute information.
  • the predicted service attribute information of the terminal device such as predicted service type, predicted service duration, etc.
  • the predicted service attribute information can be determined by the terminal device based on the previous history
  • the record information (such as application access records, content request preferences, etc.) is obtained through a preset algorithm (such as neural network algorithm, etc.) to realize the target network selected by the terminal device based on the predicted service attribute information when selecting a network, which can ensure communication services.
  • the stability of the network reduces the frequent network switching of terminal equipment, improves the communication performance index, and realizes the user's personalized service experience.
  • the current service attribute information includes current service type, current service duration and user preference; the current service weight based on the current service weight and the current network utility value of each candidate network , and based on the predicted service weight and the predicted network utility value of each candidate network, including: obtaining the service quality requirement of the current service according to the current service attribute information; the service quality requirement includes the current one or more of the bandwidth, delay, jitter and packet loss rate required by the terminal device; obtain a network utility function based on the quality of service and the user preference; according to the current The current network utility value is obtained from the network attribute information and the network utility function; the predicted network utility value is obtained according to the predicted network attribute information and the network utility function of each candidate network.
  • a network utility function is constructed based on the service quality requirements of the current user service of the terminal equipment and user preferences, and then a network utility value is obtained according to the network utility function, and the terminal equipment performs personalized network selection based on the network utility value,
  • the selected target network can more accurately reflect the user's service intention, realize the user's personalized service experience, ensure the stability of the communication service, reduce the frequent network switching of terminal equipment, and improve the communication performance index.
  • the obtaining the network utility function based on the quality of service and the user preference includes: obtaining L network decision attribute utility functions based on the quality of service requirement; According to the user preference, the network decision attribute weight corresponding to each of the network decision attribute utility functions in the L network decision attribute utility functions is obtained; according to the L network decision attribute utility functions and their corresponding network decision attributes Weights, which determine the network utility function.
  • the service quality requirements of the current service of the terminal device include L network decision attributes (such as bandwidth, delay, etc.), and different network decision attributes are represented by corresponding network decision attribute utility functions, and then Combined with user preference information, the weights of the utility functions of different network decision-making attributes are determined, and then the network utility functions are obtained.
  • the network utility function can reflect the relationship between different network decision attributes and user satisfaction, so that the target network selected based on the network utility function can more accurately reflect the user's business intention, realize the user's personalized business experience, and ensure communication services.
  • the stability of the network reduces the frequent switching of the network by the terminal equipment, and improves the communication performance index.
  • the method further includes: measuring the reference signal received power or reference signal received quality of the N candidate networks; judging whether the reference signal received power or reference signal received quality meets a pre-evaluation condition; M candidate networks that satisfy the pre-evaluation conditions are determined as the M candidate networks.
  • the terminal device selects M candidate networks that meet the pre-evaluation conditions from the N candidate networks, and then selects the target network to access from the M candidate networks, and the terminal device only calculates the M candidate networks.
  • the comprehensive utility value reduces the calculation amount of terminal equipment in the process of intelligent network selection.
  • the present application provides a method for intelligent network selection, which is characterized in that it is applied to a network device, and the network device provides S candidate networks; the S candidate networks are M candidates for the terminal device A network in a network; the method includes: acquiring second history record information of each candidate network in the S candidate networks; the second history record information includes the second history record information including each cell's One or more of historical bandwidth, time delay, jitter, and packet loss rate; and S pieces of predicted network attribute information are obtained by performing prediction according to each of the second historical record information.
  • the terminal device when the terminal device determines the network utility value of the candidate network, it needs to consider the predicted network attribute information of the candidate network (such as the future network bandwidth, delay, jitter, packet loss rate, etc.), the predicted network attribute information It can be obtained by the network equipment of the candidate network based on the historical information between the networks (such as the historical bandwidth, delay, jitter, and packet loss rate of each cell) through a preset algorithm (such as neural network algorithm, etc.)
  • the target network selected by NetTime based on the predicted network attribute information of the candidate network provided by the network device can ensure the stability of the communication service, thereby reducing the frequent network switching of the terminal device, improving the communication performance index, and realizing the personalized service experience of the user.
  • the method further includes: acquiring current service attribute information and predicted service attribute information of the terminal device; determining the current service weight according to the current service attribute information, and determining the current service weight according to the prediction
  • the service attribute information determines the predicted service weight, and obtains the comprehensive service weight of the terminal device;
  • the current service attribute information includes one or more of the current service type, current service duration and user preference, and the predicted service
  • the attribute information includes one or more of the predicted service type, the predicted service duration and the user preference; and the service comprehensive weight is sent to the terminal device.
  • the terminal device can send current service attribute information (such as current service type, current service duration, etc.) and predicted service attribute information (such as predicted service type, predicted service duration, etc.) to the network providing multiple candidate networks One of the one or more network devices, and then the network device calculates the integrated service weight. Next, the network device sends the obtained comprehensive service weight to the terminal device, so that the terminal device can select the target network based on the comprehensive service weight, which improves the communication performance index and realizes the user's personalized service experience.
  • current service attribute information such as current service type, current service duration, etc.
  • predicted service attribute information such as predicted service type, predicted service duration, etc.
  • the method further includes: determining the current network utility value according to the current network attribute information corresponding to each candidate network, and determining the current network utility value according to the prediction corresponding to each candidate network
  • the network attribute information determines the predicted network utility value; the current network attribute information includes one or more of the current network bandwidth, delay, jitter, and packet loss rate; the predicted network attribute information includes the future network bandwidth , one or more of delay, jitter, and packet loss rate; S network utility values are obtained based on each of the current network utility values and each of the predicted network utility values; the network utility values include the same The current network effect value and the predicted network utility value of the candidate network; and the S network utility values corresponding to the S candidate networks are sent to the terminal device.
  • the network device may obtain the current network utility value based on the current network attribute information (such as the current network bandwidth, delay, jitter, packet loss rate, etc.) of the S candidate networks, and predict the network attribute information (such as the future network bandwidth, delay, jitter, packet loss rate, etc.) to determine the predicted network utility value, so as to obtain the network utility value of the S candidate networks, so that the terminal device can select the target network based on the network utility value, which improves the communication performance. Metrics to realize user personalized service experience.
  • the current network attribute information such as the current network bandwidth, delay, jitter, packet loss rate, etc.
  • the present application provides an apparatus for intelligent network selection, which is characterized in that, when applied to terminal equipment, the apparatus includes: a first processing unit configured to determine a comprehensive service weight of the terminal equipment; the service The comprehensive weight includes the current service weight and the predicted service weight of the terminal equipment; the current service weight is a weight obtained based on the current service attribute information of the terminal equipment, and the predicted service weight is based on the predicted service attribute information of the terminal equipment.
  • a second processing unit configured to respectively determine the network utility values of the M candidate networks of the terminal device; each of the network utility values includes the current network utility value and the predicted network utility value of the corresponding network; the The current network utility value is a value obtained based on the current network attribute information, and the predicted network utility value is a value obtained based on the predicted network attribute information; a third processing unit is configured to be based on the service comprehensive weight and the M candidate networks The network utility values obtained are obtained, and M comprehensive utility values are obtained; and the fourth processing unit is configured to select the target network of the terminal device from the M candidate networks based on the M comprehensive utility values.
  • the first processing unit is configured to: obtain the current service weight according to the current service attribute information of the terminal device, and obtain the current service weight according to the predicted service of the terminal device
  • the attribute information obtains the weight of the predicted service
  • the current service attribute information includes one or more of the current service type, the current service start time, the current service duration, the service flow, the service feature and the user preference
  • the predicted service includes one or more of predicted service type, predicted service start time, predicted service duration, predicted service flow, predicted service characteristics and the user preference.
  • the first processing unit is specifically configured to: send the current service attribute information and the predicted service attribute information of the terminal device to a network device; the network device is the same as the M A network device associated with one or more candidate networks in the candidate networks; and receiving the service comprehensive weight sent by the network device.
  • the second processing unit is specifically configured to: determine the current network utility value of each of the candidate networks according to the current network attribute information corresponding to the M candidate networks, and determining the predicted network utility value of each of the candidate networks according to the predicted network attribute information corresponding to the M candidate networks, to obtain the network utility value of the M candidate networks;
  • the current network attribute The information includes one or more of the current network bandwidth, delay, jitter, and packet loss rate;
  • the predicted network attribute information includes one or more of the future network bandwidth, delay, jitter, and packet loss rate.
  • the second processing unit is specifically configured to: receive the network utility of the M candidate networks respectively sent by the one or more network devices that provide M candidate network associations value.
  • the third processing unit is specifically configured to: based on the current service weight and the current network utility value of each candidate network, and based on the predicted service weight and each candidate network The predicted network utility values of each of the candidate networks, and a comprehensive utility value for each of the networks is calculated.
  • the apparatus further includes: a first acquiring unit, configured to acquire first history record information of the terminal device; the first history record information includes application access records, content request preferences , one or more of service start time, service flow, service characteristics and use duration; and a fifth processing unit, configured to perform prediction according to the first historical record information to obtain the predicted service attribute information.
  • the second processing unit is specifically configured to: obtain the service quality requirement of the current service according to the current service attribute information; the service quality requirement includes the current requirements of the terminal device one or more of the bandwidth, delay, jitter and packet loss rate; obtain a network utility function based on the quality of service and the user preference; according to the current network attribute information of each candidate network and all The current network utility value is obtained from the network utility function; the predicted network utility value is obtained according to the predicted network attribute information of each candidate network and the network utility function.
  • the second processing unit is specifically configured to: obtain L network decision attribute utility functions based on the service quality requirement; obtain the L network decision attribute utility functions based on the service quality and the user preference The network decision attribute weight corresponding to each of the network decision attribute utility functions in the network decision attribute utility function; according to the L network decision attribute utility functions and their corresponding network decision attribute weights, the network utility function is determined. .
  • the apparatus further includes: a sixth processing unit, configured to measure the reference signal received power or reference signal received quality of the N candidate networks; judge the reference signal received power or reference signal received quality Whether the pre-evaluation condition is satisfied; M candidate networks that satisfy the pre-evaluation condition are determined as the M candidate networks.
  • a sixth processing unit configured to measure the reference signal received power or reference signal received quality of the N candidate networks; judge the reference signal received power or reference signal received quality Whether the pre-evaluation condition is satisfied; M candidate networks that satisfy the pre-evaluation condition are determined as the M candidate networks.
  • the present application provides a device for intelligent network selection, which is characterized in that, when applied to a network device, the network device provides S candidate networks; the S candidate networks are M candidates for the terminal device A network in a network; the apparatus includes: a first acquisition unit configured to acquire second history record information of each candidate network in the S candidate networks; the second history record information includes the history of each cell One or more of bandwidth, delay, jitter, and packet loss rate; a first processing unit, configured to perform prediction according to each of the second historical record information to obtain S pieces of predicted network attribute information.
  • the apparatus further includes: a second processing unit, configured to acquire current service attribute information and predicted service attribute information of the terminal device; determine the current service attribute information according to the current service attribute information weight, and determine the predicted service weight according to the predicted service attribute information to obtain the comprehensive service weight of the terminal device; the current service attribute information includes one of the current service type, the current service duration and the user preference or The predicted service attribute information includes one or more of predicted service type, predicted service duration and the user preference; a first sending unit is configured to send the comprehensive weight of the service to the terminal device.
  • the apparatus further includes: a third processing unit, configured to determine the current network utility value according to the current network attribute information corresponding to each candidate network, and according to each of the The predicted network attribute information corresponding to the candidate network determines the predicted network utility value;
  • the current network attribute information includes one or more of the current network bandwidth, delay, jitter, and packet loss rate;
  • the predicted network The attribute information includes one or more of the future bandwidth, delay, jitter, and packet loss rate of the network;
  • S network utility values are obtained based on each of the current network utility values and each of the predicted network utility values;
  • the network utility value includes the current network effect value and the predicted network utility value of the same candidate network;
  • the second sending unit is configured to send the S network utility values corresponding to the S candidate networks to the terminal device .
  • the present application provides a method for intelligent network selection, which is characterized in that, when applied to a network device, the method includes: determining a comprehensive service weight of a terminal device; the comprehensive service weight includes the current value of the terminal device. Service weight and predicted service weight; the current service weight is the weight obtained based on the current service attribute information of the terminal equipment, and the predicted service weight is the weight obtained based on the predicted service attribute information of the terminal equipment;
  • the network utility values of the M candidate networks of the device each of the network utility values includes the current network utility value and the predicted network utility value of the corresponding network; the current network utility value is a value obtained based on the current network attribute information, and the The predicted network utility value is a value obtained based on the predicted network attribute information; based on the service comprehensive weight and the network utility value of the M candidate networks, M comprehensive utility values are obtained; based on the M comprehensive utility values,
  • the target network of the terminal device is selected from the M candidate networks.
  • the present application provides a computer storage medium, characterized in that, the computer storage medium stores a computer program, and when the computer program is executed by a processor, implements any one of the methods described in the first aspect above.
  • the present application provides a computer storage medium, characterized in that, the computer storage medium stores a computer program, and when the computer program is executed by a processor, implements any one of the methods described in the second aspect above.
  • the present application provides a computer storage medium, wherein the computer storage medium stores a computer program, and when the computer program is executed by a processor, implements any one of the methods described in the first aspect.
  • the present application provides a computer storage medium, characterized in that the computer storage medium stores a computer program, and when the computer program is executed by a processor, implements any one of the methods described in the second aspect above.
  • an embodiment of the present invention provides an electronic device, the electronic device includes a processor, and the processor is configured to support the electronic device to implement corresponding functions in the method for intelligent network selection provided in the first aspect.
  • the electronic device may also include a memory for coupling with the processor that holds program instructions and data necessary for the electronic device.
  • the electronic device may also include a communication interface for the electronic device to communicate with other devices or a communication network.
  • an embodiment of the present invention provides an electronic device, the electronic device includes a processor, and the processor is configured to support the electronic device to implement corresponding functions in the method for intelligent network selection provided in the second aspect.
  • the electronic device may also include a memory for coupling with the processor that holds program instructions and data necessary for the electronic device.
  • the electronic device may also include a communication interface for the electronic device to communicate with other devices or a communication network.
  • the present application provides a chip system
  • the chip system includes a processor for supporting an electronic device to implement the functions involved in the above-mentioned first aspect, for example, generating or processing the above-mentioned method for intelligent network selection. information involved.
  • the chip system further includes a memory for storing necessary program instructions and data of the electronic device.
  • the chip system may be composed of chips, or may include chips and other discrete devices.
  • the present application provides a chip system
  • the chip system includes a processor for supporting an electronic device to implement the functions involved in the above second aspect, for example, generating or processing the above method for intelligent network selection. information involved.
  • the chip system further includes a memory for storing necessary program instructions and data of the electronic device.
  • the chip system may be composed of chips, or may include chips and other discrete devices.
  • FIG. 1A is a schematic diagram of an architecture of a wireless communication system according to an embodiment of the present invention.
  • FIG. 1B is a schematic structural diagram of an intelligent network selection provided by an embodiment of the present invention.
  • FIG. 1C is a schematic diagram of service quality requirements for different services in the prior art.
  • FIG. 1D is a schematic diagram of a predicted user service model according to an embodiment of the present invention.
  • FIG. 2A is a schematic flowchart of an intelligent network selection method in an embodiment of the present application.
  • FIG. 2B is a schematic flowchart of another method for intelligent network selection in an embodiment of the present application.
  • FIG. 3A is a flowchart illustrating an example of an intelligent network selection method provided by an embodiment of the present invention.
  • FIG. 3B is an exemplary flowchart of a method for determining a network utility function according to an embodiment of the present invention.
  • FIG. 4A is a schematic diagram of an apparatus for intelligent network selection provided by the present application according to an embodiment of the present invention.
  • FIG. 4B is a schematic diagram of an apparatus for intelligent network selection provided by the present application according to an embodiment of the present invention.
  • FIG. 1A is a schematic diagram of the architecture of a wireless communication system according to an embodiment of the present invention.
  • the wireless communication system 100 is used for intelligently selecting an optimal network for a terminal device to improve user experience.
  • the system architecture may include a terminal device 101 and a network device 102 . in,
  • the terminal devices 101 may be distributed in the entire wireless communication system 100, and may be stationary or mobile.
  • the terminal device 101 may be a mobile device, a mobile station, a mobile unit, an M2M terminal, a wireless unit, a remote unit, a terminal agent, a mobile client, and the like.
  • Common mobile terminals include devices such as smartphones, smart watches, and tablet computers.
  • the terminal device 101 may also be the intelligent network selection device in this application, or a terminal device configured to include the intelligent network selection device in this application.
  • the terminal device 101 accesses the network provided by the network device 102, the terminal device 101 can use different services through the network.
  • the smartphone when the smartphone is in the overlapping area covered by cell A of base station 1, cell B of base station 3, and cell C of base station 2, the smartphone can select one of the three cells to access, and then use the The network performs corresponding services (such as voice services, video calling services, etc.).
  • corresponding services such as voice services, video calling services, etc.
  • the network device 102 can be distributed in the entire communication system 100, and can be a base station, such as base station 1, base station 2, and base station 3, and the base station can be a time division synchronous code division multiple access (Time Division Synchronous Code Division Multiple Access, TD-SCDMA) system
  • the base transceiver station (Base Transceiver Station, BTS) in the LTE system can also be an evolved base station (Evolutional Node B, eNB) in the LTE system, and the base station in the 5G system and the New Radio (NR) system.
  • BTS Base Transceiver Station
  • eNB evolved base station
  • 5G system and the New Radio (NR) system New Radio
  • the base station may also be an access point (Access Point, AP), a transmission node (Trans TRP), a central unit (Central Unit, CU) or other network entities, and may include some or all of the functions of the above network entities.
  • each base station can be divided into one or more cells, and each cell can provide a network for the terminal device 101 .
  • a target cell can be selected for access, so that the network provided by the cell can be used to perform corresponding services.
  • the smartphone when the smartphone is in the overlapping area covered by cell A of base station 1, cell B of base station 3, and cell C of base station 2, the smartphone can select among the three cells by using the intelligent network selection method provided in the embodiment of the present invention.
  • the cell with the best performance is accessed to select the optimal network.
  • system architecture of a kind of intelligent network selection in FIG. 1A is only an exemplary implementation in the embodiment of the application, and the intelligent network selection system architecture in the embodiment of the application includes but is not limited to the above system architecture. .
  • FIG. 1B is a schematic diagram of the architecture of an intelligent network selection provided by an embodiment of the present invention.
  • the intelligent network selection architecture is used to intelligently select an optimal network for a terminal device to improve user experience.
  • the architecture may include a service identification module 201 , a service prediction module 202 , a policy mapping module 203 , an intention mapping module 204 and an analysis decision module 205 . in,
  • the service identification module 201 is used to identify the type of service according to the identification of the service initiated by the user on the terminal device, such as the ID of the APP, the service type ID, or according to the length of the service package, the arrival interval of the service package, etc., thereby outputting the current service.
  • QoS requirements taking 5G as an example, use a set of integer values to represent different service levels of a business, including parameters such as priority, delay, and packet error rate.
  • FIG. 1C is a schematic diagram of service quality requirements of different services in the prior art.
  • the QoS requirements of the conversational voice service include: the priority is 20, the packet delay budget is 100ms, the packet error rate is 10-2 ; the QoS requirements of the session video service include: the priority is 40, the packet delay budget is 150ms, and the packet error rate is 10-3 ; the QoS requirements of the real-time game service include: the priority is 30, the packet delay budget is is 50ms, and the packet error rate is 10 -3 .
  • the service prediction module 202 is used for predicting the service type and usage time of the user in a certain period of time in the future.
  • FIG. 1D is a schematic diagram of a predictive user service model provided by an embodiment of the present invention.
  • the input of the service prediction module in the figure includes information such as the user's access record to the application, content request preference, and usage duration statistics, and the output is Predicted business type and usage duration.
  • the terminal device can train the neural network algorithm through offline supervised learning according to the user's movement in the space-time dimension (such as location information at different times) and historical data of business habits (such as APP or business information used at different times). , predict the user's business needs for a period of time in the future.
  • the predicted business information can include when and where to carry out what kind of business, business duration, etc., and can be combined with the local information of the terminal device (such as electricity, user preferences, etc.) to obtain user Predict business attribute information.
  • the user-predicted service attribute information may include the user's service requirements (such as service type, service duration, etc.) and device status (such as power level, user preference, etc.) in a future period of time.
  • the policy mapping module 203 is used to determine the network decision attribute utility function required by the service; and to determine the comprehensive weight of the current and future services.
  • the terminal device can determine the network performance index utility function required by the service according to the QoS requirements of the user, that is, the network decision attribute utility function; according to the user's current service attribute information and the user's predicted service attribute information, determine the current and future services.
  • Comprehensive weight For example, a user is watching a video, and it is predicted that the user will play cloud AR/VR games in the next hour. Since cloud AR/VR games have higher QoS requirements than videos, the comprehensive weight of future services needs to be higher than that of current services.
  • the weight is heavy, for example, it can be determined that the comprehensive weight of the service is [0.3, 0.7], which means that compared with the future network performance and user status, the user's demand for the services currently provided by the network is lower, and the current resource demand of terminal equipment, such as energy demand, is higher. Low.
  • the comprehensive weight of future services may be smaller than the current weight of services.
  • the comprehensive weight of services can be determined to be [0.6, 0.4].
  • the comprehensive weight of services changes dynamically, and is used to perform weighted comprehensive calculation on the current and future utility values of each cell in the analysis and decision module.
  • the policy mapping module adds the prediction of user service requirements and user status, comprehensively considers the user's current and future service requirements, and obtains a dynamically changing service comprehensive weight, which is beneficial to meet the different needs of users and ensure that Continuity of user service.
  • the intent mapping module 204 is configured to determine the corresponding network decision attribute weight.
  • the terminal device may determine the corresponding network decision attribute weight according to the user's QoS requirement and user preference.
  • QoS requirement For the one hand, users have different requirements for network performance when conducting different services such as voice calls, games, and watching live broadcasts.
  • different users have different communication service preferences for the same service. Some users prefer the ultimate service experience, ignore the cost of network traffic, and pursue the service experience with large bandwidth and high speed, while some users prefer low service experience. Available services at a low price, the pursuit of bandwidth speed is not large. Therefore, the terminal device can perform fast and active intelligent network access selection according to the service intention, and it is necessary to assign weights to each network decision attribute according to the QoS requirements and user preferences corresponding to different service types.
  • the analysis and decision module 205 can combine the utility functions of each unit according to the utility function of each network decision attribute provided by the strategy mapping module and the weight distribution of the network decision attribute provided by the intention mapping module to obtain a multivariate utility function (for example, for the unit utility function).
  • the function calculates the weight to the power, and then multiplies to calculate the product); then substitute the obtained network decision attribute information into the calculation to obtain the current overall utility value and the future overall utility value of each cell in the network; and then according to the policy mapping module.
  • the comprehensive service weight combines the current and future utility values of each cell (eg, weighted summation), and selects the cell with the largest current and future comprehensive utility value after calculation, that is, the cell that best matches the user's business intention.
  • the various network attribute information obtained by the terminal equipment are delivered by the base station to the terminal equipment according to the request of the terminal equipment.
  • the network attribute information stored in the Neighbor Cell List (NCL) delivered by the base station through a broadcast message (for example, including the current network attribute information and predicted future network attribute information of each cell in the network environment) .
  • the network attribute information is the current value and predicted value of each network attribute of each cell, as shown in the following table:
  • Table 1 Schematic representation of current and predicted values of network attribute information
  • Network attribute information service area Neighborhood 1 ... Neighborhood N current bandwidth BW 1 BW 2 BW N-1 forecast bandwidth BW 1 ' BW 2 ' BW N-1 ' ... ... current delay T1 T 2 T N-1 prediction delay T 1 ' T 2 ' T N-1 '
  • the analysis and decision module can obtain the values of the corresponding network attributes (such as current bandwidth, predicted bandwidth, etc.) of each cell through the network attribute information, so as to calculate the comprehensive utility value of each cell.
  • FIG. 1B the architecture of an intelligent network selection in FIG. 1B is only an exemplary implementation in the embodiments of the present application, and the intelligent network selection architecture in the embodiments of the present application includes but is not limited to the above system architecture.
  • Fig. 2A is a schematic flowchart of an intelligent network selection method in the embodiment of the present application, below will be combined with Fig. 2A and based on the intelligent network selection system architecture in the above-mentioned Fig. 1A from the interactive side of terminal equipment and network equipment
  • the intelligent network selection method in the embodiment of the present application is described.
  • the present application describes in each process step that the corresponding execution subject is a terminal device or a network device, respectively, but this does not mean that the embodiments of the present application only The corresponding method flow can be performed through the described execution body.
  • Step S301 The terminal device determines the comprehensive service weight of the terminal device.
  • the service comprehensive weight includes the current service weight and the predicted service weight of the terminal equipment; the current service weight is a weight obtained based on the current service attribute information of the terminal equipment, and the predicted service weight is based on the terminal equipment prediction.
  • terminal devices can be understood as mobile devices such as smart phones, smart watches, and tablet computers; the comprehensive service weight can be understood as representing the network resource requirements of the current service and the predicted service respectively; the current service weight can be understood as the current service weight of the terminal device.
  • the demand for network resources of the running service; the predicted service weight can be understood as the demand for network resources of the service that the terminal device may run in a period of time in the future; the current service attribute information can be understood as the current service type and current service duration of the terminal device.
  • predicted service attribute information can be understood as information such as the type of service that the terminal device may run in the future and the duration of the predicted service. For example, a user is currently using video software to watch videos, and it is predicted that the user will play AR/VR games in the next hour. Since AR/VR games have higher requirements for network service quality than videos, future The comprehensive weight of the business needs to be larger than the current business weight. For example, it can be determined that the comprehensive weight of the business is [0.3, 0.7], which means that compared with the future network performance and user status, the user has a lower demand for the service quality currently provided by the network, which is understandable.
  • the network resources (eg energy requirements) currently required for the terminal device are lower than the network resources required in the future.
  • the method further includes: the terminal device obtains first history record information of the terminal device; the first history record information includes application access records, content request preferences, service start time, One or more of service flow, service characteristics, and usage duration; the predicted service attribute information is obtained by predicting according to the first historical record information.
  • the first historical record information can be understood as historical information such as historical application access records, user content request preferences, and application usage duration before the terminal device; application access records can be understood as records of users using applications at different times; content requests Preference can be understood as the user's request preference for content at different times; usage duration can be understood as service duration; service features can include network slices corresponding to corresponding services, data network name DNN or access point name APN, etc.; predicted service attributes
  • the information can be understood as the possible operation of the terminal device for a period of time in the future and the possible duration of the service obtained through a prediction algorithm (such as a neural network algorithm) based on the terminal device's historical record information.
  • the terminal device can train a neural network algorithm through offline supervised learning based on the user's movement in time and space (such as location information at different times) and historical data of business habits (such as APP or business information used at different times) to predict The user's business needs for a period of time in the future, the predicted business information can include when and where to carry out what kind of business, business duration, etc., and can be combined with the local information of the terminal device (such as electricity, user preferences, etc.) to obtain the user's predicted business property information.
  • the user-predicted service attribute information may include the user's service requirements (such as service type, service duration, etc.) and device status (such as power level, user preference, etc.) in a future period of time.
  • determining the comprehensive service weight of the terminal device involved in the above step S301 may include the following two implementations:
  • Embodiment 1 The terminal device generates the comprehensive weight of the service.
  • the terminal device determining the comprehensive service weight of the terminal device includes: obtaining the current service weight by the terminal device according to the current service attribute information of the terminal device, and obtaining the current service weight according to the current service attribute information of the terminal device;
  • the predicted service attribute information of the terminal device obtains the predicted service weight;
  • the current service attribute information includes one or more of current service type, current service duration and user preference, and the predicted service attribute information includes prediction One or more of service type, predicted service duration, and said user preference.
  • the current service attribute information can be understood as information such as the service currently running on the terminal device, the duration of the current service, and user preferences;
  • the predicted service attribute information can be understood as a prediction algorithm (such as a neural network algorithm) based on the historical record information of the terminal device.
  • the obtained information such as the services that the terminal equipment may run in the future, and the possible duration of the services.
  • the terminal device can obtain the current service weight based on the current service attribute information, and obtain the predicted service weight based on the predicted service attribute information.
  • the user's preference for the future game service makes the future resource demand (such as energy demand) of the terminal device lower. Therefore, the comprehensive weight of the future service can be smaller than the current service weight. For example, it can be determined that the comprehensive weight of the service The weights are [0.6, 0.4].
  • Embodiment 2 The comprehensive weight of the service is generated by the network device.
  • the method further includes: acquiring the current service attribute information and predicted service attribute information of the terminal device by the network device; determining the current service weight according to the current service attribute information, and The predicted service attribute information determines the predicted service weight, and obtains the comprehensive service weight of the terminal device; the current service attribute information includes one or more of the current service type, current service duration and user preference, the The predicted service attribute information includes one or more of the predicted service type, the predicted service duration and the user preference; and the service comprehensive weight is sent to the terminal device.
  • the terminal device can send current service attribute information (such as current service type, current service duration, etc.) and predicted service attribute information (such as predicted service type, predicted service duration, etc.) to the network providing multiple candidate networks One of the one or more network devices, and then the network device calculates the integrated service weight. Next, the network device sends the obtained comprehensive service weight to the terminal device, so that the terminal device can select the target network based on the comprehensive service weight, which improves the communication performance index and realizes the user's personalized service experience.
  • current service attribute information such as current service type, current service duration, etc.
  • predicted service attribute information such as predicted service type, predicted service duration, etc.
  • the terminal device determining the comprehensive service weight of the terminal device includes: the terminal device sending the current service attribute information and the predicted service attribute of the terminal device to a network device information; the network device is a network device associated with one or more candidate networks in the M candidate networks; and the service comprehensive weight sent by the network device is received.
  • the M candidate networks can be understood as the network currently accessed by the terminal device and its adjacent networks, such as 4G network, 5G network, cellular network, WiFi network, etc.;
  • the M candidate networks can be understood as the networks that can be connected to the area where the terminal device is located.
  • the terminal device when the terminal device is in the network connection state, it can be understood that there is a signaling connection (control plane connection) between the terminal and the core network, or there is a signaling connection and data transmission connection (user plane connection), including the UE and the base station. Between the air interface connection and the connection between the base station and the core network, the terminal in the network connection state can directly interact with the network to establish a connection for data transmission or directly perform data transmission on the existing data transmission connection.
  • the terminal device if the terminal device is not connected to the network, it can be understood that there is no signaling connection between the terminal and the network, or there is a signaling connection but is in an inactive state, and the terminal cannot directly perform signaling interaction or data transmission with the core network.
  • the network device associated with the candidate network can be understood as the network device involved in the network architecture associated with the adjacent network currently accessing the network, or as the network device that provides the candidate network, that is, the device that provides network services for the candidate network, such as Provide access network equipment of candidate networks (base station equipment of 4G, 5G and future networks, or ePDG, N3IWF and some access network equipment of future networks), core network equipment (MME, Serving-GW, P-GW in 4G etc., or AMF, SMF, UPF, etc.
  • the network equipment associated with the current access network can be understood as the network equipment involved in the network architecture of the current access network , can also be understood as the network equipment that provides the current access network, that is, the equipment that provides network services for the currently accessed network, such as the access network equipment that provides the current access (4G, 5G and future network base station equipment, or ePDG , N3IWF and some access network equipment in future networks), core network equipment (MME, Serving-GW, P-GW, etc. in 4G, or AMF, SMF, UPF, etc. in 5G, and core network equipment in future networks) ).
  • the terminal device may receive the network utility values of the network calculated by the respective network devices of the multiple candidate networks, and then the terminal device may select the target access from the multiple candidate networks based on the network utility values.
  • the network reduces the data processing capability requirements of the terminal equipment, and the selected target network can ensure the stability of the communication service, thereby reducing the frequent switching of the network by the terminal equipment, improving the communication performance index, and realizing the personalized service experience of the user.
  • the network device obtains the first history record information of the terminal device; the first history record information includes the access record of the application of the terminal device, the content request preference, the service start time, the service flow, the service feature, and the usage duration.
  • the network device performs prediction according to the first historical record information to obtain the predicted service attribute information, and further, the network device determines the comprehensive service weight based on the predicted service attribute information and the current service attribute information.
  • Step S302 The terminal device respectively determines the network utility values of the M candidate networks of the terminal device.
  • Each of the network utility values includes a current network utility value and a predicted network utility value of the corresponding network; the current network utility value is a value obtained based on the current network attribute information, and the predicted network utility value is obtained based on the predicted network attribute information value of .
  • the M candidate networks can be understood as the network currently accessed by the terminal device and its adjacent networks, such as 4G network, 5G network, cellular network, WiFi network, etc.;
  • the M candidate networks can be understood as the networks that can be connected in the area where the terminal device is located;
  • the network utility value of the candidate network can be understood as the user's service experience satisfaction under the network, such as stuck, smooth etc.;
  • the current network utility value of the candidate network can be understood as the user's satisfaction with the current service experience under this network;
  • the predicted network utility value of the candidate network can be understood as the user's satisfaction with the predicted service experience under this network.
  • the network device provides S candidate networks; the S candidate networks are networks among M candidate networks of the terminal device; the method includes: a network The device obtains the second history record information of each candidate network in the S candidate networks; the second history record information includes one or more of the historical bandwidth, delay, jitter, and packet loss rate of each cell Prediction is performed according to each of the second historical record information to obtain S pieces of predicted network attribute information. It is understandable that there may be some restrictions when predicting the network attribute information based on the historical bandwidth, delay, jitter, and packet loss rate of each cell, such as the same time period, a specific location area, and so on.
  • the terminal device when the terminal device determines the network utility value of the candidate network, it needs to consider the predicted network attribute information of the candidate network (such as the future network bandwidth, delay, jitter, packet loss rate, etc.), the predicted network attribute information It can be obtained by the network equipment of the candidate network based on the historical information between the networks (such as the historical bandwidth, delay, jitter, and packet loss rate of each cell) through a preset algorithm (such as neural network algorithm, etc.)
  • the target network selected by NetTime based on the predicted network attribute information of the candidate network provided by the network device can ensure the stability of the communication service, thereby reducing the frequent network switching of the terminal device, improving the communication performance index, and realizing the personalized service experience of the user.
  • the method further includes: the terminal device measures the reference signal received power or the reference signal received quality of the N candidate networks; and judges whether the reference signal received power or the reference signal received quality meets a pre-evaluation condition ; Determine the M candidate networks that satisfy the pre-evaluation conditions as the M candidate networks.
  • the N candidate networks can be understood as at least one cell among the cell currently accessed by the terminal equipment UE and the adjacent cells, for example, the UE currently accesses the cell A, and B, C, and D are the neighbors of the cell A.
  • the N candidate networks can be understood as at least one cell among the four cells A, B, C, and D, or the N candidate networks can also be understood as the network standard currently accessed by the UE and the network that can be accessed.
  • At least one network in other network standards or a cell under at least one network for example, if the UE currently accesses a 4G network and can access a 5G network, a WiFi network, etc., the N candidate networks can be understood as 4G, At least one network or at least one cell in one of the three networks of 5G and WiFi; the reference signal reception power reference and signal reception quality can be understood as the key parameters representing the wireless signal strength and wireless signal quality; the pre-evaluation conditions can be understood as The reference signal received power reference and/or the received signal quality meet a certain threshold value, which is usually a parameter configured by the base station to the terminal for performing cell reselection in an idle state and an event for initiating a cell handover in a connected state.
  • a certain threshold value which is usually a parameter configured by the base station to the terminal for performing cell reselection in an idle state and an event for initiating a cell handover in a connected state.
  • the base station configures the A3 event for the UE (in the same wireless access technology, the signal strength measurement value of the neighboring cell is higher than that of the serving cell) A certain threshold value and lasts for a certain period of time.
  • the measurement value of the neighboring cell is 65dbm
  • the measurement value of the serving cell is 70dbm
  • the threshold value is 3dB and the duration is 3s
  • the A3 event is satisfied
  • the B2 event The signal strength measurement value of the serving cell is lower than the first threshold, and the signal strength measurement value of the neighboring cell is higher than the second threshold value, wherein the serving cell and the neighboring cell belong to different radio access technology cells
  • M candidate networks can be It is understood as the network currently accessed and the network that meets the pre-evaluation conditions in its neighboring networks.
  • implementation of determining the network utility values of the M candidate networks of the terminal device involved in the foregoing step S302 may include the following two implementations:
  • Embodiment 1 The network utility value is generated by the terminal device.
  • the terminal device determining the network utility values of the M candidate networks respectively includes: the terminal device determining each candidate network according to the current network attribute information corresponding to the M candidate networks and the predicted network utility value of each candidate network is determined according to the predicted network attribute information corresponding to the M candidate networks, and the network of the M candidate networks is obtained.
  • Utility value the current network attribute information includes one or more of the current network bandwidth, delay, jitter, and packet loss rate
  • the predicted network attribute information includes future network bandwidth, delay, jitter, packet loss one or more of the rates.
  • the current network attribute information of the candidate network can be understood as the current network bandwidth, delay, jitter, packet loss rate and other information;
  • the current network utility value of the candidate network can be understood as the user's satisfaction with the current service experience under the network degree;
  • the predicted network utility value of the candidate network can be understood as the user's satisfaction with the predicted service experience under this network.
  • the terminal device may determine the current network utility value based on the respective current network attribute information (such as the current network bandwidth, delay, jitter, packet loss rate, etc.) of the multiple candidate networks, and predict the network attribute information (such as the future network bandwidth, delay, jitter, packet loss rate, etc.) to determine the predicted network utility value, so as to obtain the network utility value of multiple candidate networks.
  • the network utility values of multiple candidate networks can be obtained on the terminal device, which improves the efficiency of the terminal device's intelligent network selection, and the selected target network can ensure the stability of communication services.
  • frequent network switching of terminal devices is reduced, communication performance indicators are improved, and user personalized service experience is realized.
  • the terminal device requests the network device (such as the base station) for network decision attribute information through designated signaling, and the network delivers the network decision attribute information of the neighboring cells that meet the handover pre-evaluation conditions, including the current and Predicted network decision attribute information.
  • the network device such as the base station
  • the network delivers the network decision attribute information of the neighboring cells that meet the handover pre-evaluation conditions, including the current and Predicted network decision attribute information.
  • the current service attribute information includes current service type, current service duration and user preference; the terminal device is based on the current service weight and the current network of each candidate network utility value, and the predicted network utility value based on the predicted service weight and each candidate network, including: the terminal device obtains the service quality requirement of the current service according to the current service attribute information; the service quality The requirements include one or more of the bandwidth, delay, jitter and packet loss rate currently required by the terminal device; a network utility function is obtained based on the quality of service and the user preference; according to each candidate network The current network utility value is obtained from the current network attribute information and the network utility function; the predicted network utility value is obtained according to the predicted network attribute information and the network utility function of each candidate network.
  • service quality requirements can be understood as information such as bandwidth, delay, jitter, and packet loss rate required by the terminal device to run the current service
  • user preferences can be understood as the user's communication service preferences for services, for example, when a user is making a voice call
  • Different services such as games, live streaming, etc., have different requirements for network performance.
  • different users have different communication service preferences for the same service. Some users prefer the ultimate service experience, and the cost of network traffic can be ignored.
  • the pursuit of high-bandwidth and high-speed service experience while some users prefer low-cost available services, and have little pursuit of bandwidth speed
  • network utility function can be understood as a utility function between network services and user satisfaction.
  • a network utility function is constructed based on the service quality requirements of the current user service of the terminal equipment and user preferences, and then a network utility value is obtained according to the network utility function, and the terminal equipment performs personalized network selection based on the network utility value,
  • the selected target network can more accurately reflect the user's service intention, realize the user's personalized service experience, ensure the stability of the communication service, reduce the frequent network switching of terminal equipment, and improve the communication performance index.
  • the terminal device obtains a network utility function based on the service quality and the user preference, including: the terminal device obtains L network decision attribute utility functions based on the service quality requirement; The quality of service and the user preference obtain the network decision attribute weight corresponding to each of the L network decision attribute utility functions; according to the L network decision attribute utility functions and their corresponding The network decision attribute weight is used to determine the network utility function.
  • the service quality requirements of the current service of the terminal device include L network decision attributes (such as bandwidth, delay, etc.), and different network decision attributes are represented by the corresponding network decision attribute utility functions, and then combined with user preference information. Determine the weight of the utility function of different network decision attributes, and then obtain the network utility function.
  • the network utility function can reflect the relationship between different network decision attributes and user satisfaction, so that the target network selected based on the network utility function can more accurately reflect the user's business intention, realize the user's personalized business experience, and ensure communication services.
  • the stability of the network reduces the frequent switching of the network by the terminal equipment, and improves the communication performance index.
  • a specific method for determining a network utility function is exemplarily given in the embodiment of the present application, but the specific method for determining the network utility function in the embodiment of the present application includes but is not limited to the following description.
  • Different network decision attributes need to be represented by corresponding utility functions, so as to reflect the relationship between different network decision attributes and user satisfaction. For example, there is an interval [x min , x max ] in the user's requirement for the available bandwidth. In this interval, the user satisfaction increases with the increase of the bandwidth decision attribute value. If the available bandwidth is lower than x min , it means that the user does not Satisfied with the service, because it can no longer meet the basic requirements of the current business. On the other hand, if the available bandwidth is higher than x max , the user's satisfaction will not be significantly improved, because the user has been satisfied with the highest service requirements, and more bandwidth can no longer bring about an improvement in the user experience.
  • the sigmoid function conforms to the above characteristics, therefore, the sigmoid function can be used as the utility function of the available bandwidth network decision properties.
  • a basic sigmoid function representation is as follows:
  • the utility function u(x 1 ) of the decision attribute of the bandwidth network can be expressed as:
  • x m represents the equivalent spectral bandwidth of the user's service requirements in the current communication environment, that is, the user's desired bandwidth
  • x 1 represents the available bandwidth of the current network
  • BW represents the upper limit of the network bandwidth
  • ⁇ represents the user's sensitivity to the bandwidth attribute, and the higher the sensitivity, the larger the value of ⁇ .
  • the user's requirement is that the lower the value, the better, and the user expects to access a network with low delay.
  • the utility function u(x 2 ) of the decision attribute of the delay network can be expressed as:
  • x max represents the upper limit of the delay decision attribute in the wireless network (indicating timeout)
  • x 2 represents the current network delay decision attribute value
  • ⁇ represents the sensitivity of the user service to the delay attribute. The higher the sensitivity, the ⁇ The larger the value is.
  • the user preference of the terminal device is [4, 2, 1, 3]. The larger the user preference value corresponding to a certain service quality requirement, the more attention the user pays.
  • the available bandwidth is the fundamental guarantee of the service quality
  • the packet loss rate is the root cause of the deterioration of the image quality. It is not important, because the software running the online game service on the user equipment has the ability to reduce the impact of delay and jitter on the service experience through appropriate caching. Therefore, in the weight distribution of each network decision attribute, the weight of the available bandwidth is the largest, the packet loss rate this time, and the delay and jitter are the smallest.
  • the weight of each network decision attribute can be determined by the following formula:
  • the multi-unit utility function is multiplied or added according to the weight, or the multiplication and addition are combined to obtain the multi-functional utility function, that is, the network utility function.
  • the multiplicative utility function expression for the multiplicative combination is:
  • ui (x) and wi represent the utility function and weight of the i-th network decision attribute, respectively.
  • Embodiment 2 The network utility value is generated by the network device.
  • the network device determines the current network utility value according to the current network attribute information corresponding to each candidate network, and determines the current network utility value according to the predicted network attribute corresponding to each candidate network information to determine the predicted network utility value;
  • the current network attribute information includes one or more of the current network bandwidth, delay, jitter, and packet loss rate;
  • the predicted network attribute information includes future network bandwidth, time one or more of delay, jitter, and packet loss rate;
  • S network utility values are obtained based on each of the current network utility values and each of the predicted network utility values; the network utility values include the same The current network effect value and the predicted network utility value of the candidate network; the S network utility values corresponding to the S candidate networks are sent to the terminal device.
  • the current network attribute information of the candidate network can be understood as the current network bandwidth, delay, jitter, packet loss rate and other information;
  • the current network utility value of the candidate network can be understood as the user's satisfaction with the current service experience under the network degree;
  • the predicted network utility value of the candidate network can be understood as the user's satisfaction with the predicted service experience under this network.
  • the network device may obtain the current network utility value based on the current network attribute information (such as the current network bandwidth, delay, jitter, packet loss rate, etc.) of the S candidate networks, and predict the network attribute information (such as the future network bandwidth, delay, jitter, packet loss rate, etc.) to determine the predicted network utility value, so as to obtain the network utility value of the S candidate networks, so that the terminal device can select the target network based on the network utility value, which improves the communication performance. Metrics to realize user personalized service experience.
  • the current network attribute information such as the current network bandwidth, delay, jitter, packet loss rate, etc.
  • the network device obtains the current service attribute information of the terminal device, and then the network device obtains the service quality requirement of the current service according to the current service attribute information of the terminal device; one or more of delay, jitter and packet loss rate; network equipment obtains a network utility function based on the service quality and user preference; input the current network attribute information of the candidate network into the network utility function to obtain the current network utility value; input the predicted network attribute information of each candidate network into the network utility function to obtain the predicted network utility value.
  • the current service attribute information may include current service type, current service duration and user preference.
  • the network device obtains a network utility function based on the service quality and the user preference, including: the network device obtains L network decision attribute utility functions based on the service quality requirement; the network device obtains L network decision attribute utility functions based on the service quality and the user preference; The user preference obtains the network decision attribute weight corresponding to each of the network decision attribute utility functions in the L network decision attribute utility functions; according to the L network decision attribute utility functions and the corresponding network decision Attribute weights, which determine the network utility function.
  • the terminal device determining the network utility values of the M candidate networks respectively includes: the terminal device receiving the M messages respectively sent by the one or more network devices that provide associations of the M candidate networks. the network utility values for each candidate network.
  • the terminal device may receive the network utility values of the network calculated by the respective network devices of the multiple candidate networks, and then the terminal device may select the target access from the multiple candidate networks based on the network utility values.
  • the network reduces the data processing capability requirements of the terminal equipment, and the selected target network can ensure the stability of the communication service, thereby reducing the frequent switching of the network by the terminal equipment, improving the communication performance index, and realizing the personalized service experience of the user.
  • Step S301 may be executed first and then step S302 may be executed, or step S302 may be executed first and then step S301 may be executed, or steps S301 and S301 may be executed simultaneously.
  • Step S303 The terminal device obtains M comprehensive utility values based on the service comprehensive weight and the network utility values of the M candidate networks.
  • the terminal device obtains M comprehensive utility values based on the service comprehensive weight and the network utility values of the M candidate networks, including: the terminal device is based on the current service weight and the current network utility value for each of the candidate networks, and a composite utility value for each of the candidate networks is calculated based on the predicted traffic weight and the predicted network utility value for each of the candidate networks.
  • the current network utility value and the predicted network utility value of different candidate networks are respectively weighted with the corresponding current service weight and predicted service weight to obtain the comprehensive utility value of multiple networks. It is based on the current service state of the terminal device and the service state in the future, as well as the current state of the network and the state in the future.
  • the terminal device is based on the comprehensive utility of multiple candidate networks.
  • the value of the target network selected from multiple candidate networks can ensure the stability of communication services, thereby reducing the frequent switching of networks by terminal equipment, improving communication performance indicators, and realizing user personalized service experience.
  • the current and future utility values of each cell are comprehensively weighted according to the service comprehensive weight, that is, the current and future utility values of each cell are comprehensively considered, and the target cell with the largest comprehensive utility value is selected.
  • the comprehensive utility value calculation expression is:
  • w cur and are the current business weight and the current network decision attribute utility value
  • w fut and are the future business weight and the future network decision attribute utility value, respectively.
  • Step S304 The terminal device selects a target network of the terminal device from the M candidate networks based on the M comprehensive utility values.
  • the comprehensive utility values of the M candidate networks can be obtained respectively, and then the candidate network with the highest comprehensive utility value score can be used as the target network. Further, the terminal equipment can report the target cell and enter the handover execution process. For example, assuming that there are 4 cells ⁇ 0, 1, 2, 3 ⁇ , according to their current and future network decision attributes, the current and future utility values of each cell can be obtained. like:
  • the terminal device may also perform target cell selection. For example, when the terminal device is in an idle state (such as not accessing the network), the terminal device autonomously selects the target cell; when the terminal device is in a connected state, the terminal sends a measurement report message to the network, which indicates the UE's preferred target cell. , or priority target cell ranking, etc., the UE selects a target cell to perform the handover process to the target cell in combination with the information sent by the UE.
  • target cell selection For example, when the terminal device is in an idle state (such as not accessing the network), the terminal device autonomously selects the target cell; when the terminal device is in a connected state, the terminal sends a measurement report message to the network, which indicates the UE's preferred target cell. , or priority target cell ranking, etc., the UE selects a target cell to perform the handover process to the target cell in combination with the information sent by the UE.
  • the current service status of the terminal device and the service status in a future period of time are comprehensively considered to determine the current service of the terminal equipment and the predicted service for a future period of time.
  • the comprehensive service weight of the network, as well as the current utility value of different networks and the predicted utility value for a period of time in the future so as to select the network that best matches the user's service intention, avoiding the need to use multiple network utility functions only based on the user's current service and current network status.
  • the target network selected by the terminal device from multiple candidate networks can ensure the stability of the communication service, thereby reducing the frequent switching of the network by the terminal device, improving the communication performance index, and realizing the user's individuality. business experience.
  • Fig. 2B is the schematic flow chart of another kind of intelligent network selection method in the embodiment of the present application, below will be combined with accompanying drawing 2B and based on the intelligent network selection system architecture in above-mentioned Fig. 1A from terminal equipment and network equipment
  • the interactive side describes the intelligent network selection method in the embodiment of the present application.
  • the present application describes in each process step that the corresponding execution subjects are network devices, respectively, but this does not mean that the embodiments of the present application can only be implemented through all The described execution body performs the corresponding method flow.
  • Step S3001 The network device determines the comprehensive service weight of the terminal device.
  • the service comprehensive weight includes the current service weight and the predicted service weight of the terminal equipment; the current service weight is a weight obtained based on the current service attribute information of the terminal equipment, and the predicted service weight is based on the terminal equipment prediction.
  • a network device can be understood as a base station, for example, a base station can be a base transceiver station (Base Transceiver Station, BTS) in a Time Division Synchronous Code Division Multiple Access (TD-SCDMA) system, or a Evolved base station (Evolutional Node B, eNB) in LTE system, as well as base station in 5G system, new air interface (NR) system, etc.; terminal equipment can be understood as mobile devices such as smartphones, smart watches, tablet computers, etc.; business comprehensive weight It can be understood as representing the requirements of the current service and the predicted service for network resources respectively; the current service weight can be understood as the demand for network resources of the service currently running on the terminal device; the predicted service weight can be understood as the terminal equipment may run in the future Service requirements for network resources; current service attribute information can be understood as information such as the current service type and current service duration of the terminal equipment; predicted service attribute information can be understood as the terminal equipment may run in the future.
  • BTS Base Transceiver Station
  • Service type and predicted service duration, etc. information For example, a user is currently using video software to watch videos, and it is predicted that the user will play AR/VR games in the next hour. Since AR/VR games have higher requirements for network service quality than videos, future The comprehensive weight of the business needs to be larger than the current business weight. For example, it can be determined that the comprehensive weight of the business is [0.3, 0.7], which means that compared with the future network performance and user status, the user has a lower demand for the service quality currently provided by the network, which is understandable.
  • the network resources (eg energy requirements) currently required for the terminal device are lower than the network resources required in the future.
  • the terminal device acquires first history record information of the terminal device; the first history record information includes application access records, content request preferences, service start time, service flow, service characteristics and Use one or more of the durations; perform prediction according to the first historical record information to obtain the predicted service attribute information.
  • the first historical record information can be understood as historical information such as historical application access records, user content request preferences, and application usage duration before the terminal device; application access records can be understood as records of users using applications at different times; content requests Preference can be understood as the user's request preference for content at different times; usage duration can be understood as service duration; service features can include the network slice corresponding to this service, data network name DNN or access point name APN, etc.; prediction The service attribute information can be understood as the service that the terminal device may run for a period of time in the future and the possible duration of the service obtained by a prediction algorithm (such as a neural network algorithm) based on the historical record information of the terminal device.
  • a prediction algorithm such as a neural network algorithm
  • the network device obtains the first history record information of the terminal device; the first history record information includes the access record of the application of the terminal device, the content request preference, the service start time, the service flow, the service feature, and the usage duration.
  • the network device performs prediction according to the first historical record information to obtain the predicted service attribute information, and further, the network device determines the comprehensive service weight based on the predicted service attribute information and the current service attribute information.
  • determining the comprehensive service weight of the terminal device involved in the above step S3001 may include the following two implementations:
  • Embodiment 1 The terminal device generates the comprehensive service weight, and then sends it to the network device.
  • the terminal device obtains the current service weight according to the current service attribute information of the terminal device, and obtains the predicted service weight according to the predicted service attribute information of the terminal device; Send the comprehensive service weight to the network device;
  • the current service attribute information includes one or more of current service type, current service duration and user preference, and the predicted service attribute information includes predicted service type, prediction One or more of service duration and said user preference.
  • the current service attribute information can be understood as information such as the service currently running on the terminal device, the duration of the current service, and user preferences;
  • the predicted service attribute information can be understood as a prediction algorithm (such as a neural network algorithm) based on the historical record information of the terminal device.
  • the obtained information such as the services that the terminal equipment may run in the future, and the possible duration of the services.
  • the network device determining the comprehensive service weight of the terminal device includes: the network device receiving the comprehensive service weight sent by the terminal device.
  • Embodiment 2 The comprehensive weight of the service is generated by the network device.
  • the method further includes: acquiring the current service attribute information and predicted service attribute information of the terminal device by the network device; determining the current service weight according to the current service attribute information, and The predicted service attribute information determines the predicted service weight, and obtains the comprehensive service weight of the terminal device; the current service attribute information includes one or more of the current service type, current service duration and user preference, the The predicted service attribute information includes one or more of predicted service type, predicted service duration and the user preference.
  • Step S3002 The network device respectively determines the network utility values of the M candidate networks of the terminal device.
  • Each of the network utility values includes a current network utility value and a predicted network utility value of the corresponding network; the current network utility value is a value obtained based on the current network attribute information, and the predicted network utility value is obtained based on the predicted network attribute information value of .
  • the M candidate networks can be understood as the network currently accessed by the terminal device and its adjacent networks, such as 4G network, 5G network, cellular network, WiFi network, etc.;
  • the M candidate networks can be understood as the networks that can be connected in the area where the terminal device is located;
  • the network utility value of the candidate network can be understood as the user's service experience satisfaction under the network, such as stuck, smooth etc.;
  • the current network utility value of the candidate network can be understood as the user's satisfaction with the current service experience under this network;
  • the predicted network utility value of the candidate network can be understood as the user's satisfaction with the predicted service experience under this network.
  • the method further includes: the network device obtains current network attribute information and predicted network attribute information of the M candidate networks respectively; the current service attribute information includes the current service type, current service duration and One or more of user preferences, the predicted service attribute information includes one or more of predicted service type, predicted service duration and the user preference; current network attribute information based on each of the candidate networks obtaining the current network utility value; obtaining the predicted network utility value based on the predicted network attribute information of each candidate network.
  • the method further includes: the terminal device measures the reference signal received power or the reference signal received quality of the N candidate networks; and judges whether the reference signal received power or the reference signal received quality meets a pre-evaluation condition ; Determine the M candidate networks that satisfy the pre-evaluation conditions as the M candidate networks.
  • the N candidate networks can be understood as being able to be at least one of the cell that the UE is currently accessing and the neighboring cells, for example, the UE is currently accessing the A cell, B, C, and D are the neighboring cells of the A cell, Then the N candidate networks can be understood as at least one cell among the four cells A, B, C, and D, or the N candidate networks can also be understood as the network standard currently accessed by the UE and the accessible network.
  • At least one network in other network standards or a cell under at least one network for example, if the UE currently accesses a 4G network and can access a 5G network, a WiFi network, etc., the N candidate networks can be understood as 4G, 5G, At least one of the three WiFi networks or a cell in at least one network; the reference signal received power reference and signal received quality can be understood as the key parameters representing the wireless signal strength and wireless signal quality; the pre-evaluation condition can be understood as the reference signal The received power reference and/or the received signal quality meet a certain threshold value, which are usually parameters configured by the base station to the terminal for performing cell reselection in the idle state and events for initiating cell handover in the connected state.
  • a certain threshold value which are usually parameters configured by the base station to the terminal for performing cell reselection in the idle state and events for initiating cell handover in the connected state.
  • the base station configures the A3 event for the UE (in the same wireless access technology, the signal strength measurement value of the neighboring cell is higher than that of the serving cell) A certain threshold value and lasts for a certain period of time.
  • the measurement value of the neighboring cell is 65dbm
  • the measurement value of the serving cell is 70dbm
  • the threshold value is 3dB and the duration is 3s
  • the A3 event is satisfied
  • the B2 event The signal strength measurement value of the serving cell is lower than the first threshold, and the signal strength measurement value of the neighboring cell is higher than the second threshold value, wherein the serving cell and the neighboring cell belong to different radio access technology cells
  • M candidate networks can be It is understood as the network currently accessed and the network that meets the pre-evaluation conditions in its neighboring networks.
  • the network device involved in the above step S3002 determines the network utility value of the M candidate networks of the terminal device.
  • the implementation may include the following two implementations:
  • Embodiment 1 The network utility value is generated by the terminal device.
  • the terminal device determining the network utility values of the M candidate networks respectively includes: the terminal device determining each candidate network according to the current network attribute information corresponding to the M candidate networks and the predicted network utility value of each candidate network is determined according to the predicted network attribute information corresponding to the M candidate networks, and the network of the M candidate networks is obtained.
  • utility value sending the network utility value of the M candidate networks to the network device;
  • the current network attribute information includes one or more of the current network bandwidth, delay, jitter, and packet loss rate ;
  • the predicted network attribute information includes one or more of the future network bandwidth, time delay, jitter, and packet loss rate.
  • the current network attribute information of the candidate network can be understood as the current network bandwidth, delay, jitter, packet loss rate and other information;
  • the current network utility value of the candidate network can be understood as the user's satisfaction with the current service experience under the network degree;
  • the predicted network utility value of the candidate network can be understood as the user's satisfaction with the predicted service experience under this network.
  • the terminal device may determine the current network utility value based on the respective current network attribute information (such as the current network bandwidth, delay, jitter, packet loss rate, etc.) of the multiple candidate networks, and predict the network attribute information (such as the future network bandwidth, delay, jitter, packet loss rate, etc.) to determine the predicted network utility value, so as to obtain the network utility value of multiple candidate networks.
  • the network utility values of multiple candidate networks can be obtained on the terminal device, so that the target network selected by the network device for the terminal device can ensure the stability of communication services, thereby reducing the frequency of terminal devices. Switching the network improves the communication performance index and realizes the user's personalized service experience.
  • the current service attribute information includes current service type, current service duration and user preference; the terminal device is based on the current service weight and the current network of each candidate network utility value, and the predicted network utility value based on the predicted service weight and each candidate network, including: the terminal device obtains the service quality requirement of the current service according to the current service attribute information; the service quality The requirements include one or more of the bandwidth, delay, jitter and packet loss rate currently required by the terminal device; a network utility function is obtained based on the quality of service and the user preference; according to each candidate network The current network utility value is obtained from the current network attribute information and the network utility function; the predicted network utility value is obtained according to the predicted network attribute information and the network utility function of each candidate network.
  • service quality requirements can be understood as information such as bandwidth, delay, jitter, and packet loss rate required by the terminal device to run the current service
  • user preferences can be understood as the user's communication service preferences for services, for example, when a user is making a voice call
  • Different services such as games, live streaming, etc., have different requirements for network performance.
  • different users have different communication service preferences for the same service. Some users prefer the ultimate service experience, and the cost of network traffic can be ignored.
  • the pursuit of high-bandwidth and high-speed service experience while some users prefer low-cost available services, and have little pursuit of bandwidth speed
  • network utility function can be understood as a utility function between network services and user satisfaction.
  • the terminal device obtains a network utility function based on the service quality and the user preference, including: the terminal device obtains L network decision attribute utility functions based on the service quality requirement; The quality of service and the user preference obtain the network decision attribute weight corresponding to each of the L network decision attribute utility functions; according to the L network decision attribute utility functions and their corresponding The network decision attribute weight is used to determine the network utility function.
  • the service quality requirements of the current service of the terminal device include L network decision attributes (such as bandwidth, delay, etc.), and different network decision attributes are represented by the corresponding network decision attribute utility functions, and then combined with user preference information. Determine the weight of the utility function of different network decision attributes, and then obtain the network utility function.
  • the network utility function can reflect the relationship between different network decision attributes and user satisfaction, so that the target network selected based on the network utility function can more accurately reflect the user's business intention, realize the user's personalized business experience, and ensure communication services.
  • the stability of the network reduces the frequent switching of the network by the terminal equipment, and improves the communication performance index.
  • the network device respectively determining the network utility values of the M candidate networks includes: receiving the network utility values of the M candidate networks sent by the terminal device.
  • Embodiment 2 The network utility value is generated by the network device.
  • the method further includes: the network device obtains current network attribute information and predicted network attribute information of the M candidate networks respectively; the current network attribute information includes the current network bandwidth, delay, jitter , one or more of packet loss rate; the predicted network attribute information includes one or more of future network bandwidth, delay, jitter, and packet loss rate.
  • the network device determines the current network utility value according to the current network attribute information corresponding to each candidate network, and determines the current network utility value according to the predicted network attribute corresponding to each candidate network information to determine the predicted network utility value;
  • the current network attribute information includes one or more of the current network bandwidth, delay, jitter, and packet loss rate;
  • the predicted network attribute information includes future network bandwidth, time one or more of delay, jitter, and packet loss rate;
  • each network utility value is obtained based on each of the current network utility values and each of the predicted network utility values; the network utility value includes the same The current network effect value and predicted network effect value of the candidate network.
  • the network device obtains the current service attribute information of the terminal device, and then the network device obtains the service quality requirement of the current service according to the current service attribute information of the terminal device; one or more of delay, jitter and packet loss rate; obtain a network utility function based on the service quality and user preference; input the current network attribute information of the candidate network into the network utility function to obtain the current network utility value; The predicted network attribute information of each candidate network is input into the network utility function to obtain the predicted network utility value.
  • service quality requirements can be understood as information such as bandwidth, delay, jitter, and packet loss rate required by the terminal device to run the current service
  • user preferences can be understood as the user's communication service preferences for services, for example, when a user is making a voice call
  • Different services such as games, live broadcasts, etc. have different requirements for network performance.
  • different users have different preferences for communication services of the same service. Some users prefer the ultimate service experience, and the cost of network traffic can be ignored.
  • the pursuit of high-bandwidth and high-speed service experience while some users prefer low-cost available services, and have little pursuit of bandwidth speed
  • network utility function can be understood as a utility function between network services and user satisfaction.
  • the network device obtains a network utility function based on the service quality and the user preference, including: the network device obtains L network decision attribute utility functions based on the service quality requirement; the network device obtains L network decision attribute utility functions based on the service quality and the user preference; The user preference obtains the network decision attribute weight corresponding to each of the network decision attribute utility functions in the L network decision attribute utility functions; according to the L network decision attribute utility functions and the corresponding network decision Attribute weights, which determine the network utility function.
  • Step S3001 may be executed first and then step S3002 may be executed, or step S3002 may be executed first and then step S3001 may be executed, or steps S3001 and S3001 may be executed simultaneously.
  • Step S3003 The network device obtains M comprehensive utility values based on the service comprehensive weight and the network utility values of the M candidate networks.
  • the network device obtains M comprehensive utility values based on the service comprehensive weight and the network utility values of the M candidate networks, including: the network device is based on the current service weight and the current network utility value for each of the candidate networks, and a composite utility value for each of the candidate networks is calculated based on the predicted traffic weight and the predicted network utility value for each of the candidate networks.
  • Step S3004 The network device selects the target network of the terminal device from the M candidate networks based on the M comprehensive utility values.
  • the comprehensive utility values of the M candidate networks can be obtained respectively, and then the candidate network with the highest comprehensive utility value score can be used as the target network.
  • the current service status of the terminal device and the service status in a future period of time, as well as the current status of different networks candidate for the terminal equipment and the status of a future period of time are comprehensively considered, so as to determine the current service status of the terminal equipment and the status of the terminal equipment.
  • the terminal device can select a target network from multiple candidate networks, which can ensure the stability of the communication service, thereby reducing the frequent switching of the network by the terminal device, improving the communication performance index, and realizing the user experience. Personalize the business experience.
  • FIG. 3A is an intelligent network selection method provided by an embodiment of the present invention.
  • FIG. 3B is an example flow chart of a method for determining a network utility function provided by an embodiment of the present invention.
  • the intelligent network selection process of terminal equipment can be as follows:
  • Step 401 The terminal device measures RSRP or RSRQ.
  • Step 402 The terminal device determines whether the network satisfies the pre-evaluation condition.
  • Step 403 The terminal device acquires the network attribute information of neighboring cells.
  • Step 404 Based on the user service attribute information and the network attribute information, the terminal device selects the neighbor cell with the largest comprehensive utility value.
  • Step 405 The terminal equipment performs target cell selection. For example, when the terminal device is in an idle state (such as not accessing the network), the terminal device autonomously selects the target cell; when the terminal device is in a connected state, the terminal sends a measurement report message to the network, which indicates the UE's preferred target cell. , or priority target cell ranking, etc., the UE selects a target cell to perform the handover process to the target cell in combination with the information sent by the UE.
  • Step 1 Determine the weights of the decision attributes.
  • Step 2 Match the utility function to determine the comprehensive weight of the business.
  • Step 3 Combine multivariate utility functions and comprehensively calculate.
  • Step 4 Network access selection.
  • the current service status of the terminal device and the service status in a future period of time are comprehensively considered to determine the current service of the terminal equipment and the predicted service for a future period of time.
  • FIG. 4A is a schematic diagram of an apparatus for intelligent network selection provided by the present application according to an embodiment of the present invention, which is applied to terminal equipment.
  • the intelligent network selection apparatus 50 may include a first processing unit 501 and a second processing unit 501 .
  • the first processing unit 501 is configured to determine the comprehensive service weight of the terminal equipment; the comprehensive service weight includes the current service weight and the predicted service weight of the terminal equipment; the current service weight is based on the current service of the terminal equipment.
  • the weight obtained by the attribute information, the predicted service weight is the weight obtained based on the predicted service attribute information of the terminal device;
  • the second processing unit 502 is configured to respectively determine the network utility values of the M candidate networks of the terminal device; each of the network utility values includes the current network utility value and the predicted network utility value of the corresponding network; the current network utility value The value is a value obtained based on the current network attribute information, and the predicted network utility value is a value obtained based on the predicted network attribute information;
  • a third processing unit 503, configured to obtain M comprehensive utility values based on the service comprehensive weight and the network utility values of the M candidate networks;
  • the fourth processing unit 504 is configured to select a target network of the terminal device from the M candidate networks based on the M comprehensive utility values.
  • the first processing unit 501 is configured to: obtain the current service weight according to the current service attribute information of the terminal device, and obtain the current service weight according to the prediction of the terminal device
  • the predicted service weight is obtained from the service attribute information;
  • the current service attribute information includes one or more of the current service type, the current service start time, the current service duration, the service flow, the service characteristics and the user preference.
  • the service attribute information includes one or more of predicted service type, predicted service start time, predicted service duration, predicted service flow, predicted service characteristics and the user preference.
  • the first processing unit 501 is specifically configured to: send the current service attribute information and the predicted service attribute information of the terminal device to a network device; the network device is a A network device associated with one or more candidate networks in the M candidate networks; and receiving the service comprehensive weight sent by the network device.
  • the second processing unit 502 is specifically configured to: determine the current network utility value of each candidate network according to the current network attribute information corresponding to the M candidate networks , and determine the predicted network utility value of each candidate network according to the predicted network attribute information corresponding to the M candidate networks to obtain the network utility value of the M candidate networks; the current network The attribute information includes one or more of the current network bandwidth, delay, jitter, and packet loss rate; the predicted network attribute information includes one or more of the future network bandwidth, delay, jitter, and packet loss rate. kind.
  • the second processing unit 502 is specifically configured to: receive the network of the M candidate networks respectively sent by the one or more network devices that provide M candidate network associations utility value.
  • the third processing unit 503 is specifically configured to: based on the current service weight and the current network utility value of each candidate network, and based on the predicted service weight and The predicted network utility value of each of the candidate networks, and a comprehensive utility value of each of the networks is calculated.
  • the apparatus further includes: a first obtaining unit 505, configured to obtain first history record information of the terminal device; the first history record information includes application access records and content requests One or more of preferences, service start time, service flow, service characteristics and usage duration; the fifth processing unit 506 is configured to perform prediction according to the first historical record information to obtain the predicted service attribute information.
  • the second processing unit 502 is specifically configured to: obtain the service quality requirement of the current service according to the current service attribute information; the service quality requirement includes the current status of the terminal device. One or more of the required bandwidth, delay, jitter and packet loss rate; obtain a network utility function based on the quality of service and the user preference; according to the current network attribute information of each of the candidate networks and The network utility function obtains the current network utility value; and the predicted network utility value is obtained according to the predicted network attribute information of each candidate network and the network utility function.
  • the second processing unit 502 is specifically configured to: obtain L network decision attribute utility functions based on the service quality requirement; obtain the L network decision attribute utility functions based on the service quality and the user preference The network decision attribute weight corresponding to each of the network decision attribute utility functions in the network decision attribute utility functions; according to the L network decision attribute utility functions and their corresponding network decision attribute weights, the network utility is determined function.
  • the apparatus further includes:
  • the sixth processing unit 507 is configured to measure the reference signal received power or reference signal received quality of the N candidate networks; determine whether the reference signal received power or reference signal received quality satisfies the pre-evaluation condition; M candidate networks are determined as the M candidate networks.
  • FIG. 4B is a schematic diagram of an apparatus for intelligent network selection provided by the present application according to an embodiment of the present invention, which is applied to network equipment.
  • the intelligent network selection apparatus 60 may include a first obtaining unit 601 , a first processing The unit 602, the second processing unit 603, the first sending unit 604, the third processing unit 605, and the second sending unit 606, wherein the detailed description of each module is as follows.
  • the first obtaining unit 601 is configured to obtain the second historical record information of each candidate network in the S candidate networks; the second historical record information includes the historical bandwidth, time delay, jitter, and packet loss of each cell one or more of one or more of the rates;
  • the first processing unit 602 is configured to perform prediction according to each of the second historical record information to obtain S pieces of predicted network attribute information.
  • the apparatus further includes: a second processing unit 603, configured to acquire current service attribute information and predicted service attribute information of the terminal device; determine the current service attribute information according to the current service attribute information service weight, and determining the predicted service weight according to the predicted service attribute information to obtain the comprehensive service weight of the terminal device; the current service attribute information includes one of the current service type, current service duration and user preference or more, the predicted service attribute information includes one or more of the predicted service type, the predicted service duration and the user preference; the first sending unit 604 is configured to send the comprehensive service to the terminal device Weights.
  • the apparatus further includes: a third processing unit 605, configured to determine the current network utility value according to the current network attribute information corresponding to each candidate network, and according to each candidate network
  • the predicted network attribute information corresponding to the candidate network determines the predicted network utility value;
  • the current network attribute information includes one or more of the current network bandwidth, delay, jitter, and packet loss rate;
  • the predicted network The network attribute information includes one or more of future network bandwidth, delay, jitter, and packet loss rate;
  • S network utility values are obtained based on each of the current network utility values and each of the predicted network utility values;
  • the network utility value includes the current network effect value and the predicted network utility value of the same candidate network;
  • the second sending unit 606 is configured to send the S networks corresponding to the S candidate networks to the terminal device utility value.
  • the present application provides a computer storage medium, characterized in that the computer storage medium stores a computer program, and when the computer program is executed by a processor, any one of the above-mentioned methods for intelligent network selection is implemented.
  • the present application provides a computer storage medium, characterized in that, the computer storage medium stores a computer program, and when the computer program is executed by a processor, any of the above-mentioned methods for intelligent network selection is implemented.
  • the present application provides a computer storage medium, characterized in that the computer storage medium stores a computer program, and when the computer program is executed by a processor, any one of the above-mentioned methods for intelligent network selection is implemented.
  • the present application provides a computer storage medium, characterized in that, the computer storage medium stores a computer program, and when the computer program is executed by a processor, any of the above-mentioned methods for intelligent network selection is implemented.
  • An embodiment of the present invention provides an electronic device, the electronic device includes a processor, and the processor is configured to support the electronic device to implement corresponding functions in any of the foregoing methods for intelligent network selection.
  • the electronic device may also include a memory for coupling with the processor that holds program instructions and data necessary for the electronic device.
  • the electronic device may also include a communication interface for the electronic device to communicate with other devices or a communication network.
  • An embodiment of the present invention provides an electronic device, the electronic device includes a processor, and the processor is configured to support the electronic device to implement corresponding functions in any of the above-mentioned methods for intelligent network selection.
  • the electronic device may also include a memory for coupling with the processor that holds program instructions and data necessary for the electronic device.
  • the electronic device may also include a communication interface for the electronic device to communicate with other devices or a communication network.
  • the present application provides a chip system
  • the chip system includes a processor for supporting an electronic device to implement the above-mentioned functions, for example, generating or processing the information involved in the above-mentioned method for intelligent network selection.
  • the chip system further includes a memory for storing necessary program instructions and data of the electronic device.
  • the chip system may be composed of chips, or may include chips and other discrete devices.
  • the present application provides a chip system
  • the chip system includes a processor, which is used for supporting the electronic device to realize the above-mentioned functions, for example, generating or processing the information involved in the above-mentioned another method for intelligent network selection.
  • the chip system further includes a memory for storing necessary program instructions and data of the electronic device.
  • the chip system may be composed of chips, or may include chips and other discrete devices.
  • the disclosed apparatus may be implemented in other manners.
  • the device embodiments described above are only illustrative.
  • the division of the above-mentioned units is only a logical function division.
  • multiple units or components may be combined or integrated. to another system, or some features can be ignored, or not implemented.
  • the shown or discussed mutual coupling or direct coupling or communication connection may be through some interfaces, indirect coupling or communication connection of devices or units, and may be in electrical or other forms.
  • the units described above as separate components may or may not be physically separated, and components shown as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution in this embodiment.
  • each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically alone, or two or more units may be integrated into one unit.
  • the above-mentioned integrated units may be implemented in the form of hardware, or may be implemented in the form of software functional units.
  • the integrated units are implemented in the form of software functional units and sold or used as independent products, they may be stored in a computer-readable storage medium.
  • the technical solutions of the present application can be embodied in the form of software products in essence, or the parts that contribute to the prior art, or all or part of the technical solutions, and the computer software products are stored in a storage medium , including several instructions to cause a computer device (which may be a personal computer, a server, or a network device, etc., specifically a processor in the computer device) to execute all or part of the steps of the foregoing methods in the various embodiments of the present application.
  • a computer device which may be a personal computer, a server, or a network device, etc., specifically a processor in the computer device
  • the aforementioned storage medium may include: U disk, mobile hard disk, magnetic disk, optical disk, Read-Only Memory (Read-Only Memory, abbreviation: ROM) or Random Access Memory (Random Access Memory, abbreviation: RAM), etc.
  • a medium that can store program code may include: U disk, mobile hard disk, magnetic disk, optical disk, Read-Only Memory (Read-Only Memory, abbreviation: ROM) or Random Access Memory (Random Access Memory, abbreviation: RAM), etc.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Computer Security & Cryptography (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

Disclosed in the embodiments of the present invention is an intelligent network selection method, which is characterized in being applied to a terminal device. The method comprises: determining a service comprehensive weight of a terminal device, wherein the service comprehensive weight comprises a current service weight and a predicted service weight of the terminal device; respectively determining network utility values of M candidate networks, wherein each of the network utility values comprises a current network utility value and a predicted network utility value of a corresponding network; obtaining M comprehensive utility values on the basis of the service comprehensive weight and the network utility values of the M candidate networks; and selecting a target network of the terminal device from among the M candidate networks on the basis of the M comprehensive utility values. By using the embodiments of the present invention, an optimal network can be selected from among heterogeneous networks, thereby improving the user experience.

Description

一种智能选网的方法及相关装置A kind of intelligent network selection method and related device
本申请要求于2021年04月02日提交中国专利局、申请号为202110365608.4、申请名称为“一种智能选网的方法及相关装置”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。This application claims the priority of the Chinese patent application with the application number 202110365608.4 and the application title "A method and related device for intelligent network selection", which was filed with the China Patent Office on April 2, 2021, the entire contents of which are incorporated by reference in in this application.
技术领域technical field
本发明涉及无线通信技术领域，尤其涉及一种智能选网的方法及相关装置。The present invention relates to the technical field of wireless communication, and in particular, to a method and related device for intelligent network selection.
背景技术Background technique
随着用户对服务质量要求的不断提高，异构网络越来越受到欢迎。不同类型的无线网络在覆盖范围、服务支持、传输速率和移动性支持等方面存在着很大的差异，这些异构特性促使将网络作为一个整体来充分利用不同网络的优势，为用户提供无处不在的服务。在各种无线网络共存的异构网络中，首要的问题就是用户应该接入哪种无线网络，既能满足用户的不同业务服务需求，又能实现***性能最优。With the continuous improvement of users' requirements for service quality, heterogeneous networks are becoming more and more popular. Different types of wireless networks have great differences in coverage, service support, transmission rate, and mobility support. These heterogeneous characteristics prompt the network as a whole to take full advantage of different networks and provide users absent service. In a heterogeneous network in which various wireless networks coexist, the primary problem is which wireless network the user should access, which can not only meet the user's different business service requirements, but also achieve optimal system performance.
在异构网络选择中，一般最优网络的选择取决于带宽、吞吐量、延迟、抖动、丢包率、功耗、成本等多个属性，即异构网络中最优网络的选择是一个多属性决策问题。目前，提出的异构网络选择方法多侧重于采用智能优化算法，即根据异构环境下各个网络的当前网络属性信息进行接入判决，此方法并未考虑选择目标网络进行接入之后，网络属性信息的动态变化对移动用户、网络回报函数和切换判决条件的影响。同时，现有的异构网络选择方法，其目标是最大化每个连接的期望总回报，并未区分移动用户的业务类型。但在实际场景中，移动用户可能同时运行着多个业务，不同业务(如语音、视频会议和视频流等)对网络的要求是不同的，如对于带宽，语音业务一般只需几十kb的速度，但对于视频会议是远不够的。由于5G中业务类型不断增长、移动用户对网络接入的响应时间和接入决策的准确性提出了更高的要求，单一的智能优化算法已无法满足5G超密集网络异构条件下的网络选择问题。In the selection of heterogeneous networks, the selection of the optimal network generally depends on multiple attributes such as bandwidth, throughput, delay, jitter, packet loss rate, power consumption, and cost. attribute decision problem. At present, the proposed heterogeneous network selection methods mostly focus on the use of intelligent optimization algorithms, that is, the access decision is made according to the current network attribute information of each network in a heterogeneous environment. This method does not consider the network attributes after selecting the target network for access. The influence of dynamic changes of information on mobile users, network reward function and handover decision conditions. At the same time, existing heterogeneous network selection methods, whose goal is to maximize the expected total return per connection, do not differentiate the service types of mobile users. However, in actual scenarios, mobile users may run multiple services at the same time, and different services (such as voice, video conferencing, and video streaming, etc.) have different network requirements. For example, for bandwidth, voice services generally only need tens of kilobytes. speed, but not enough for video conferencing. Due to the continuous growth of service types in 5G and the higher requirements of mobile users on the response time of network access and the accuracy of access decisions, a single intelligent optimization algorithm can no longer meet the network selection under the heterogeneous conditions of 5G ultra-dense networks. question.
因此，如何选择最优网络，提升用户体验是亟待解决的问题。Therefore, how to select an optimal network and improve user experience is an urgent problem to be solved.
发明内容SUMMARY OF THE INVENTION
本发明实施例所要解决的技术问题在于，提供一种智能选网的方法及相关装置，可实现选择最优网络，提升用户体验。The technical problem to be solved by the embodiments of the present invention is to provide a method and related device for intelligent network selection, which can realize the selection of an optimal network and improve user experience.
第一方面，本发明实施案例提供一种智能选网的方法，其特征在于，应用于终端设备，所述方法包括：确定所述终端设备的业务综合权重；所述业务综合权重包括所述终端设备的当前业务权重和预测业务权重；所述当前业务权重为基于所述终端设备当前业务属性信息得到的权重，所述预测业务权重为基于所述终端设备预测业务属性信息得到的权重；分别确定所述终端设备的M个候选网络的网络效用值；每个所述网络效用值包括对应网络的当前网络效用值和预测网络效用值；所述当前网络效用值为基于当前网络属性信息得到的值，所述预测网络效用值为基于预测网络属性信息得到的值；基于所述业务综合权重和所述M个候选网络的所述网络效用值，得到M个综合效用值；基于所述M个综合效用值，从所述M个候选网络中选取所述终端设备的目标网络。In a first aspect, an implementation case of the present invention provides a method for intelligent network selection, which is characterized in that, when applied to a terminal device, the method includes: determining a comprehensive service weight of the terminal device; the comprehensive service weight includes the terminal device. The current service weight and the predicted service weight of the device; the current service weight is the weight obtained based on the current service attribute information of the terminal device, and the predicted service weight is the weight obtained based on the predicted service attribute information of the terminal device; determine respectively network utility values of the M candidate networks of the terminal device; each of the network utility values includes the current network utility value and the predicted network utility value of the corresponding network; the current network utility value is a value obtained based on the current network attribute information , the predicted network utility value is a value obtained based on the predicted network attribute information; based on the service comprehensive weight and the network utility values of the M candidate networks, M comprehensive utility values are obtained; based on the M comprehensive utility values The utility value is to select the target network of the terminal device from the M candidate networks.
在智能选网场景中，现有技术中仅根据终端设备当前业务和当前网络状态进行选网，若终端设备运行的业务或网络状态发生变化，则会导致目标网络提供的通信服务不稳定，从而致使终端设备需要频繁切换网络，而又因为现有技术中并未考虑网络的未来状态，因而更无法保证网络切换的准确性，最终导致终端设备服务体验差。而在本发明实施例中，综合考虑终端设备的当前业务状态和未来一段时间内的业务状态，以及该终端设备候选的不同网络的当前状态和未来一段时间的状态，以确定终端设备的当前业务与未来一段时间的预测业务的综合业务权重，以及不同网络的当前效用值和未来一段时间的预测效用值，从而预测并选择出与该终端设备的业务意图最匹配的网络。因此通过实施本发明实施例提供的方法，终端设备可从多个候选网络中选出的目标网络，可以保证通信服务的稳定性，从而减少终端设备频繁切换网络，提升了通信性能指标，实现用户个性化业务体验。In the intelligent network selection scenario, the prior art only selects the network based on the current service of the terminal device and the current network state. If the service or network state of the terminal device changes, the communication service provided by the target network will be unstable. As a result, the terminal device needs to switch the network frequently, and because the future state of the network is not considered in the prior art, the accuracy of the network switching cannot be guaranteed, which eventually leads to poor service experience of the terminal device. However, in this embodiment of the present invention, the current service status of the terminal device and the service status in a future period of time, as well as the current status of different networks that are candidates for the terminal equipment and the status in a future period of time are comprehensively considered to determine the current service status of the terminal equipment. The integrated service weight of the predicted service in the future, as well as the current utility value of different networks and the predicted utility value of the future period, so as to predict and select the network that best matches the service intention of the terminal device. Therefore, by implementing the method provided by the embodiment of the present invention, the terminal device can select a target network from multiple candidate networks, which can ensure the stability of the communication service, thereby reducing the frequent switching of the network by the terminal device, improving the communication performance index, and realizing the user experience. Personalize the business experience.
在一种可能的实现方式中，所述确定所述终端设备的业务综合权重，包括：根据所述终端设备的所述当前业务属性信息得到所述当前业务权重，以及根据所述终端设备的所述预测业务属性信息得到所述预测业务权重；所述当前业务属性信息包括当前业务类型、当前业务开始时间、当前业务持续时间、业务流量、业务特征和用户偏好中的一种或多种，所述预测业务属性信息包括预测业务类型、预测业务开始时间、预测业务持续时间、预测业务流量、预测业务特征和所述用户偏好中的一种或多种。In a possible implementation manner, the determining the comprehensive service weight of the terminal device includes: obtaining the current service weight according to the current service attribute information of the terminal device, and obtaining the current service weight according to the current service attribute information of the terminal device; The predicted service attribute information is used to obtain the predicted service weight; the current service attribute information includes one or more of the current service type, the current service start time, the current service duration, the service flow, the service characteristics and the user preference. The predicted service attribute information includes one or more of predicted service type, predicted service start time, predicted service duration, predicted service flow, predicted service characteristics and the user preference.
在本发明实施例中，由于终端设备上记录了当前业务属性信息(如当前业务类型、当前业务持续时间等)和预测业务属性信息(如预测业务类型、预测业务持续时间等)，当终端设备需要从多个候选网络中选择出目标接入网络时，每次选网终端设备可根据当前记录的信息得到终端设备的业务综合权重，提高了终端设备智能选网的效率，且选出的目标网络可以保证通信服务的稳定性，从而减少终端设备频繁切换网络，提升了通信性能指标，实现用户个性化业务体验。In this embodiment of the present invention, since the current service attribute information (such as current service type, current service duration, etc.) and predicted service attribute information (such as predicted service type, predicted service duration, etc.) are recorded on the terminal device, when the terminal device When the target access network needs to be selected from multiple candidate networks, the terminal equipment for each network selection can obtain the comprehensive service weight of the terminal equipment according to the currently recorded information, which improves the efficiency of intelligent network selection of the terminal equipment, and the selected target The network can ensure the stability of communication services, thereby reducing the frequent switching of networks by terminal devices, improving communication performance indicators, and realizing personalized service experience for users.
在一种可能的实现方式中，所述确定所述终端设备的业务综合权重，包括：向网络设备发送所述终端设备的所述当前业务属性信息和所述预测业务属性信息；所述网络设备为与M个候选网络中的一个或多个候选网络关联的网络设备；接收所述网络设备发送的所述业务综合权重。In a possible implementation manner, the determining the comprehensive service weight of the terminal device includes: sending the current service attribute information and the predicted service attribute information of the terminal device to a network device; the network device is a network device associated with one or more candidate networks in the M candidate networks; and receives the comprehensive service weight sent by the network device.
在本发明实施例中，终端设备可将当前业务属性信息(如当前业务类型、当前业务持续时间等)和预测业务属性信息(如预测业务类型、预测业务持续时间等)发送给提供多个候选网络的一个或多个网络设备中的一个设备，然后由该网络设备来计算综合业务权重，在智能选网的过程中降低了终端设备对数据处理能力的要求，且选出的目标网络可以保证通信服务的稳定性，从而减少终端设备频繁切换网络，提升了通信性能指标，实现用户个性化业务体验。In this embodiment of the present invention, the terminal device may send current service attribute information (such as current service type, current service duration, etc.) and predicted service attribute information (such as predicted service type, predicted service duration, etc.) One device in one or more network devices of the network, and then the network device calculates the comprehensive service weight, which reduces the data processing capability requirements of the terminal device in the process of intelligent network selection, and the selected target network can guarantee. The stability of the communication service reduces the frequent network switching of terminal equipment, improves the communication performance index, and realizes the user's personalized service experience.
在一种可能的实现方式中，所述分别确定M个候选网络的网络效用值，包括：根据所述M个候选网络对应的所述当前网络属性信息确定每个所述候选网络的所述当前网络效用值，以及根据所述M个候选网络对应的所述预测网络属性信息确定每个所述候选网络的所述预测网络效用值，得到所述M个候选网络的所述网络效用值；所述当前网络属性信息包括网络当前的带宽、时延、抖动、丢包率中的一种或多种；所述预测网络属性信息包括网络未来的带宽、时延、抖动、丢包率中的一种或多种。In a possible implementation manner, the determining the network utility values of the M candidate networks respectively includes: determining the current network utility value of each candidate network according to the current network attribute information corresponding to the M candidate networks network utility value, and determining the predicted network utility value of each candidate network according to the predicted network attribute information corresponding to the M candidate networks, to obtain the network utility value of the M candidate networks; The current network attribute information includes one or more of the current network bandwidth, delay, jitter, and packet loss rate; the predicted network attribute information includes one or more of the network's future bandwidth, delay, jitter, and packet loss rate. one or more.
在本发明实施例中，终端设备可基于多个候选网络各自的当前网络属性信息(如网络当前的带宽、时延、抖动、丢包率等)确定当前网络效用值，以及预测网络属性信息(如网络未来的带宽、时延、抖动、丢包率等)确定预测网络效用值，从而得到多个候选网络的网络 效用值。当终端设备需要进行智能选网时，就可在终端设备上得到多个候选网络的网络效用值，提高了终端设备智能选网的效率，且选出的目标网络可以保证通信服务的稳定性，从而减少终端设备频繁切换网络，提升了通信性能指标，实现用户个性化业务体验。In this embodiment of the present invention, the terminal device may determine the current network utility value based on the respective current network attribute information (such as the current network bandwidth, delay, jitter, packet loss rate, etc.) of the multiple candidate networks, and predict the network attribute information ( Such as the future network bandwidth, delay, jitter, packet loss rate, etc.) to determine the predicted network utility value, so as to obtain the network utility value of multiple candidate networks. When the terminal device needs to perform intelligent network selection, the network utility values of multiple candidate networks can be obtained on the terminal device, which improves the efficiency of the terminal device's intelligent network selection, and the selected target network can ensure the stability of communication services. Thus, frequent network switching of terminal equipment is reduced, communication performance indicators are improved, and user personalized service experience is realized.
在一种可能的实现方式中，所述分别确定M个候选网络的网络效用值，包括：接收所述提供M个候选网络关联的一个或多个网络设备分别发送的所述M个候选网络的所述网络效用值。In a possible implementation manner, the determining the network utility values of the M candidate networks respectively includes: receiving the network utility values of the M candidate networks respectively sent by the one or more network devices associated with the M candidate networks. the network utility value.
在本发明实施例中，终端设备可接收由多个候选网络各自的网络设备所计算的本网络的网络效用值，然后终端设备可基于这些网络效用值从多个候选网络中选择出目标接入网络，降低了终端设备对数据处理能力的要求，且选出的目标网络可以保证通信服务的稳定性，从而减少终端设备频繁切换网络，提升了通信性能指标，实现用户个性化业务体验。In this embodiment of the present invention, the terminal device may receive the network utility values of the network calculated by the respective network devices of the multiple candidate networks, and then the terminal device may select the target access from the multiple candidate networks based on the network utility values. The network reduces the data processing capability requirements of the terminal equipment, and the selected target network can ensure the stability of the communication service, thereby reducing the frequent switching of the network by the terminal equipment, improving the communication performance index, and realizing the personalized service experience of the user.
在一种可能的实现方式中，所述基于所述业务综合权重和所述M个候选网络的所述网络效用值，得到M个综合效用值，包括：基于所述当前业务权重和每个所述候选网络的所述当前网络效用值，以及基于所述预测业务权重和每个所述候选网络的所述预测网络效用值，计算每个所述网络的综合效用值。In a possible implementation manner, the obtaining M comprehensive utility values based on the service comprehensive weight and the network utility values of the M candidate networks includes: based on the current service weight and each of the M candidate networks. the current network utility value of the candidate network, and a comprehensive utility value for each of the candidate networks is calculated based on the predicted traffic weight and the predicted network utility value for each of the candidate networks.
在本发明实施例中，将不同候选网络的当前网络效用值和预测网络效用值分别与对应的当前业务权重和预测业务权重进行加权，得到多个网络的综合效用值，该综合效用值的确定是基于了终端设备的当前业务状态和未来一段时间的业务状态，以及网络的当前状态和未来一段时间的状态，因此通过实施本发明实施例提供的方法，终端设备基于多个候选网络的综合效用值从多个候选网络中选出的目标网络，可以保证通信服务的稳定性，从而减少终端设备频繁切换网络，提升了通信性能指标，实现用户个性化业务体验。In the embodiment of the present invention, the current network utility value and the predicted network utility value of different candidate networks are respectively weighted with the corresponding current service weight and predicted service weight to obtain the comprehensive utility value of multiple networks. It is based on the current service state of the terminal device and the service state in the future, as well as the current state of the network and the state in the future. Therefore, by implementing the method provided by the embodiment of the present invention, the terminal device is based on the comprehensive utility of multiple candidate networks. The value of the target network selected from multiple candidate networks can ensure the stability of communication services, thereby reducing the frequent switching of networks by terminal equipment, improving communication performance indicators, and realizing user personalized service experience.
在一种可能的实现方式中，所述方法还包括：获取所述终端设备的第一历史记录信息；所述第一历史记录信息包括应用的访问记录、内容请求偏好、业务开始时间、业务流量、业务特征和使用时长中的一种或多种；根据所述第一历史记录信息进行预测得到所述预测业务属性信息。In a possible implementation manner, the method further includes: acquiring first history record information of the terminal device; the first history record information includes application access records, content request preferences, service start time, service traffic , one or more of service characteristics and usage duration; and predicting according to the first historical record information to obtain the predicted service attribute information.
在本发明实施例中，确定终端设备的综合业务权重时，需要考虑终端设备的预测业务属性信息(如预测业务类型、预测业务持续时间等)，该预测业务属性信息可由终端设备基于之前的历史记录信息(如应用的访问记录、内容请求偏好等)通过预设算法(如神经网络算法等)得到，实现终端设备在选网时基于该预测业务属性信息选出的目标网络，可以保证通信服务的稳定性，从而减少终端设备频繁切换网络，提升了通信性能指标，实现用户个性化业务体验。In this embodiment of the present invention, when determining the comprehensive service weight of a terminal device, it is necessary to consider the predicted service attribute information of the terminal device (such as predicted service type, predicted service duration, etc.), and the predicted service attribute information can be determined by the terminal device based on the previous history The record information (such as application access records, content request preferences, etc.) is obtained through a preset algorithm (such as neural network algorithm, etc.) to realize the target network selected by the terminal device based on the predicted service attribute information when selecting a network, which can ensure communication services. The stability of the network reduces the frequent network switching of terminal equipment, improves the communication performance index, and realizes the user's personalized service experience.
在一种可能的实现方式中，所述当前业务属性信息包括当前业务类型、当前业务持续时间和用户偏好；所述基于所述当前业务权重和每个所述候选网络的所述当前网络效用值，以及基于所述预测业务权重和每个所述候选网络的所述预测网络效用值，包括：根据所述当前业务属性信息得到所述当前业务的服务质量需求；所述服务质量需求包括当前所述终端设备所需要的带宽、时延、抖动和丢包率中的一种或多种；基于所述服务质量和所述用户偏好得到网络效用函数；根据每个所述候选网络的所述当前网络属性信息和所述网络效用函数得到所述当前网络效用值；根据每个所述候选网络的所述预测网络属性信息和所述网络效用函数得到所述预测网络效用值。In a possible implementation manner, the current service attribute information includes current service type, current service duration and user preference; the current service weight based on the current service weight and the current network utility value of each candidate network , and based on the predicted service weight and the predicted network utility value of each candidate network, including: obtaining the service quality requirement of the current service according to the current service attribute information; the service quality requirement includes the current one or more of the bandwidth, delay, jitter and packet loss rate required by the terminal device; obtain a network utility function based on the quality of service and the user preference; according to the current The current network utility value is obtained from the network attribute information and the network utility function; the predicted network utility value is obtained according to the predicted network attribute information and the network utility function of each candidate network.
在本发明实施例中，基于终端设备的当前用户业务的服务质量需求以及用户偏好构建网络效用函数，然后根据该网络效用函数得到网络效用值，终端设备基于该网络效用值进行个性化选网，使得选出的目标网络能够更加准确的反映用户业务意图，实现用户个性化业务体 验，且可以保证通信服务的稳定性，减少终端设备频繁切换网络，提升了通信性能指标。In the embodiment of the present invention, a network utility function is constructed based on the service quality requirements of the current user service of the terminal equipment and user preferences, and then a network utility value is obtained according to the network utility function, and the terminal equipment performs personalized network selection based on the network utility value, The selected target network can more accurately reflect the user's service intention, realize the user's personalized service experience, ensure the stability of the communication service, reduce the frequent network switching of terminal equipment, and improve the communication performance index.
在一种可能的实现方式中，所述基于所述服务质量和所述用户偏好得到网络效用函数，包括：基于所述服务质量需求得到L个网络决策属性效用函数；基于所述服务质量和所述用户偏好得到所述L个网络决策属性效用函数中的每个所述网络决策属性效用函数对应的网络决策属性权重；根据所述L个网络决策属性效用函数及其对应的所述网络决策属性权重，确定所述网络效用函数。In a possible implementation manner, the obtaining the network utility function based on the quality of service and the user preference includes: obtaining L network decision attribute utility functions based on the quality of service requirement; According to the user preference, the network decision attribute weight corresponding to each of the network decision attribute utility functions in the L network decision attribute utility functions is obtained; according to the L network decision attribute utility functions and their corresponding network decision attributes Weights, which determine the network utility function.
在本发明实施例中，终端设备的当前业务的服务质量需求中包括了L个网络决策属性(如带宽、时延等)，不同的网络决策属性由相应的网络决策属性效用函数来表征，再结合用户偏好信息确定不同网络决策属性效用函数的权重，进而得到网络效用函数。通过该网络效用函数可反映不同网络决策属性与用户满意度的关系，从而基于该网络效用函数选出的目标网络能够更加准确的反映用户业务意图，实现用户个性化业务体验，且可以保证通信服务的稳定性，减少终端设备频繁切换网络，提升了通信性能指标。In the embodiment of the present invention, the service quality requirements of the current service of the terminal device include L network decision attributes (such as bandwidth, delay, etc.), and different network decision attributes are represented by corresponding network decision attribute utility functions, and then Combined with user preference information, the weights of the utility functions of different network decision-making attributes are determined, and then the network utility functions are obtained. The network utility function can reflect the relationship between different network decision attributes and user satisfaction, so that the target network selected based on the network utility function can more accurately reflect the user's business intention, realize the user's personalized business experience, and ensure communication services. The stability of the network reduces the frequent switching of the network by the terminal equipment, and improves the communication performance index.
在一种可能的实现方式中，所述方法还包括：测量N个候选网络的参考信号接收功率或参考信号接收质量；判断所述参考信号接收功率或参考信号接收质量是否满足预评估条件；将满足所述预评估条件的M个候选网络确定为所述M个候选网络。In a possible implementation manner, the method further includes: measuring the reference signal received power or reference signal received quality of the N candidate networks; judging whether the reference signal received power or reference signal received quality meets a pre-evaluation condition; M candidate networks that satisfy the pre-evaluation conditions are determined as the M candidate networks.
在本发明实施例中，终端设备从N个候选网络中选择出满足预评估条件的M个候选网络，再从M个候选网络中选择出目标网络接入，终端设备仅计算M个候选网络的综合效用值，减少了终端设备在智能选网过程中的计算量。In the embodiment of the present invention, the terminal device selects M candidate networks that meet the pre-evaluation conditions from the N candidate networks, and then selects the target network to access from the M candidate networks, and the terminal device only calculates the M candidate networks. The comprehensive utility value reduces the calculation amount of terminal equipment in the process of intelligent network selection.
第二方面，本申请提供了一种智能选网的方法，其特征在于，应用于网络设备，所述网络设备提供S个候选网络；所述S个候选网络为所述终端设备的M个候选网络中的网络；所述方法包括：获取所述S个候选网络中的每个候选网络的第二历史记录信息；所述第二历史记录信息包括所述第二历史记录信息包括每个小区的历史带宽、时延、抖动、丢包率中的一种或多种；根据每个所述第二历史记录信息进行预测得到S个预测网络属性信息。In a second aspect, the present application provides a method for intelligent network selection, which is characterized in that it is applied to a network device, and the network device provides S candidate networks; the S candidate networks are M candidates for the terminal device A network in a network; the method includes: acquiring second history record information of each candidate network in the S candidate networks; the second history record information includes the second history record information including each cell's One or more of historical bandwidth, time delay, jitter, and packet loss rate; and S pieces of predicted network attribute information are obtained by performing prediction according to each of the second historical record information.
在本发明实施中，在终端设备确定候选网络的网络效用值时，需要考虑候选网络的预测网络属性信息(如网络未来的带宽、时延、抖动、丢包率等)，该预测网络属性信息可由候选网络的网络设备基于该网络之间的历史信息(如每个小区的历史带宽、时延、抖动、丢包率)通过预设算法(如神经网络算法等)得到，实现终端设备在选网时基于网络设备提供的候选网络的预测网络属性信息选出的目标网络，可以保证通信服务的稳定性，从而减少终端设备频繁切换网络，提升了通信性能指标，实现用户个性化业务体验。In the implementation of the present invention, when the terminal device determines the network utility value of the candidate network, it needs to consider the predicted network attribute information of the candidate network (such as the future network bandwidth, delay, jitter, packet loss rate, etc.), the predicted network attribute information It can be obtained by the network equipment of the candidate network based on the historical information between the networks (such as the historical bandwidth, delay, jitter, and packet loss rate of each cell) through a preset algorithm (such as neural network algorithm, etc.) The target network selected by NetTime based on the predicted network attribute information of the candidate network provided by the network device can ensure the stability of the communication service, thereby reducing the frequent network switching of the terminal device, improving the communication performance index, and realizing the personalized service experience of the user.
在一种可能的实现方式中，所述方法还包括：获取所述终端设备的当前业务属性信息和预测业务属性信息；根据所述当前业务属性信息确定所述当前业务权重，以及根据所述预测业务属性信息确定所述预测业务权重，得到所述终端设备的业务综合权重；所述当前业务属性信息包括当前业务类型、当前业务持续时间和用户偏好中的一种或多种，所述预测业务属性信息包括预测业务类型、预测业务持续时间和所述用户偏好中的一种或多种；向所述终端设备发送所述业务综合权重。In a possible implementation manner, the method further includes: acquiring current service attribute information and predicted service attribute information of the terminal device; determining the current service weight according to the current service attribute information, and determining the current service weight according to the prediction The service attribute information determines the predicted service weight, and obtains the comprehensive service weight of the terminal device; the current service attribute information includes one or more of the current service type, current service duration and user preference, and the predicted service The attribute information includes one or more of the predicted service type, the predicted service duration and the user preference; and the service comprehensive weight is sent to the terminal device.
在本发明实施中，终端设备可将当前业务属性信息(如当前业务类型、当前业务持续时间等)和预测业务属性信息(如预测业务类型、预测业务持续时间等)发送给提供多个候选网络的一个或多个网络设备中的一个设备，然后由该网络设备来计算综合业务权重。接下来网络设备再将得到的综合业务权重发送给终端设备，从而终端设备可基于该综合业务权重选出目标网络，提升了通信性能指标，实现用户个性化业务体验。In the implementation of the present invention, the terminal device can send current service attribute information (such as current service type, current service duration, etc.) and predicted service attribute information (such as predicted service type, predicted service duration, etc.) to the network providing multiple candidate networks One of the one or more network devices, and then the network device calculates the integrated service weight. Next, the network device sends the obtained comprehensive service weight to the terminal device, so that the terminal device can select the target network based on the comprehensive service weight, which improves the communication performance index and realizes the user's personalized service experience.
在一种可能的实现方式中，所述方法还包括：根据每个所候选网络对应的所述当前网络属性信息确定所述当前网络效用值，以及根据每个所述候选网络对应的所述预测网络属性信息确定所述预测网络效用值；所述当前网络属性信息包括网络当前的带宽、时延、抖动、丢包率中的一种或多种；所述预测网络属性信息包括网络未来的带宽、时延、抖动、丢包率中的一种或多种；基于每个所述当前网络效用值和每个所述预测网络效用值得到S个网络效用值；所述网络效用值包括同一个所述候选网络的当前网络效应值和预测网络效用值；向所述终端设备发送所述S个候选网络对应的所述S个网络效用值。In a possible implementation manner, the method further includes: determining the current network utility value according to the current network attribute information corresponding to each candidate network, and determining the current network utility value according to the prediction corresponding to each candidate network The network attribute information determines the predicted network utility value; the current network attribute information includes one or more of the current network bandwidth, delay, jitter, and packet loss rate; the predicted network attribute information includes the future network bandwidth , one or more of delay, jitter, and packet loss rate; S network utility values are obtained based on each of the current network utility values and each of the predicted network utility values; the network utility values include the same The current network effect value and the predicted network utility value of the candidate network; and the S network utility values corresponding to the S candidate networks are sent to the terminal device.
在本发明实施例中，网络设备可基于S个候选网络各自的当前网络属性信息(如网络当前的带宽、时延、抖动、丢包率等)得到当前网络效用值，以及预测网络属性信息(如网络未来的带宽、时延、抖动、丢包率等)确定预测网络效用值，从而得到S个候选网络的网络效用值，从而终端设备可基于网络效用值选出目标网络，提升了通信性能指标，实现用户个性化业务体验。In this embodiment of the present invention, the network device may obtain the current network utility value based on the current network attribute information (such as the current network bandwidth, delay, jitter, packet loss rate, etc.) of the S candidate networks, and predict the network attribute information ( Such as the future network bandwidth, delay, jitter, packet loss rate, etc.) to determine the predicted network utility value, so as to obtain the network utility value of the S candidate networks, so that the terminal device can select the target network based on the network utility value, which improves the communication performance. Metrics to realize user personalized service experience.
第三方面，本申请提供了一种智能选网的装置，其特征在于，应用于终端设备，所述装置包括：第一处理单元，用于确定所述终端设备的业务综合权重；所述业务综合权重包括所述终端设备的当前业务权重和预测业务权重；所述当前业务权重为基于所述终端设备当前业务属性信息得到的权重，所述预测业务权重为基于所述终端设备预测业务属性信息得到的权重；第二处理单元，用于分别确定所述终端设备的M个候选网络的网络效用值；每个所述网络效用值包括对应网络的当前网络效用值和预测网络效用值；所述当前网络效用值为基于当前网络属性信息得到的值，所述预测网络效用值为基于预测网络属性信息得到的值；第三处理单元，用于基于所述业务综合权重和所述M个候选网络的所述网络效用值，得到M个综合效用值；第四处理单元，用于基于所述M个综合效用值，从所述M个候选网络中选取所述终端设备的目标网络。In a third aspect, the present application provides an apparatus for intelligent network selection, which is characterized in that, when applied to terminal equipment, the apparatus includes: a first processing unit configured to determine a comprehensive service weight of the terminal equipment; the service The comprehensive weight includes the current service weight and the predicted service weight of the terminal equipment; the current service weight is a weight obtained based on the current service attribute information of the terminal equipment, and the predicted service weight is based on the predicted service attribute information of the terminal equipment. the obtained weight; a second processing unit, configured to respectively determine the network utility values of the M candidate networks of the terminal device; each of the network utility values includes the current network utility value and the predicted network utility value of the corresponding network; the The current network utility value is a value obtained based on the current network attribute information, and the predicted network utility value is a value obtained based on the predicted network attribute information; a third processing unit is configured to be based on the service comprehensive weight and the M candidate networks The network utility values obtained are obtained, and M comprehensive utility values are obtained; and the fourth processing unit is configured to select the target network of the terminal device from the M candidate networks based on the M comprehensive utility values.
在一种可能的实现方式中，所述第一处理单元，具有用于：根据所述终端设备的所述当前业务属性信息得到所述当前业务权重，以及根据所述终端设备的所述预测业务属性信息得到所述预测业务权重；所述当前业务属性信息包括当前业务类型、当前业务开始时间、当前业务持续时间、业务流量、业务特征和用户偏好中的一种或多种，所述预测业务属性信息包括预测业务类型、预测业务开始时间、预测业务持续时间、预测业务流量、预测业务特征和所述用户偏好中的一种或多种。In a possible implementation manner, the first processing unit is configured to: obtain the current service weight according to the current service attribute information of the terminal device, and obtain the current service weight according to the predicted service of the terminal device The attribute information obtains the weight of the predicted service; the current service attribute information includes one or more of the current service type, the current service start time, the current service duration, the service flow, the service feature and the user preference, and the predicted service The attribute information includes one or more of predicted service type, predicted service start time, predicted service duration, predicted service flow, predicted service characteristics and the user preference.
在一种可能的实现方式中，所述第一处理单元，具体用于：向网络设备发送所述终端设备的所述当前业务属性信息和所述预测业务属性信息；所述网络设备为与M个候选网络中的一个或多个候选网络关联的网络设备；接收所述网络设备发送的所述业务综合权重。In a possible implementation manner, the first processing unit is specifically configured to: send the current service attribute information and the predicted service attribute information of the terminal device to a network device; the network device is the same as the M A network device associated with one or more candidate networks in the candidate networks; and receiving the service comprehensive weight sent by the network device.
在一种可能的实现方式中，所述第二处理单元，具体用于：根据所述M个候选网络对应的所述当前网络属性信息确定每个所述候选网络的所述当前网络效用值，以及根据所述M个候选网络对应的所述预测网络属性信息确定每个所述候选网络的所述预测网络效用值，得到所述M个候选网络的所述网络效用值；所述当前网络属性信息包括网络当前的带宽、时延、抖动、丢包率中的一种或多种；所述预测网络属性信息包括网络未来的带宽、时延、抖动、丢包率中的一种或多种。In a possible implementation manner, the second processing unit is specifically configured to: determine the current network utility value of each of the candidate networks according to the current network attribute information corresponding to the M candidate networks, and determining the predicted network utility value of each of the candidate networks according to the predicted network attribute information corresponding to the M candidate networks, to obtain the network utility value of the M candidate networks; the current network attribute The information includes one or more of the current network bandwidth, delay, jitter, and packet loss rate; the predicted network attribute information includes one or more of the future network bandwidth, delay, jitter, and packet loss rate. .
在一种可能的实现方式中，所述第二处理单元，具体用于：接收所述提供M个候选网络关联的一个或多个网络设备分别发送的所述M个候选网络的所述网络效用值。In a possible implementation manner, the second processing unit is specifically configured to: receive the network utility of the M candidate networks respectively sent by the one or more network devices that provide M candidate network associations value.
在一种可能的实现方式中，所述第三处理单元，具体用于：基于所述当前业务权重和每 个所述候选网络的所述当前网络效用值，以及基于所述预测业务权重和每个所述候选网络的所述预测网络效用值，计算每个所述网络的综合效用值。In a possible implementation manner, the third processing unit is specifically configured to: based on the current service weight and the current network utility value of each candidate network, and based on the predicted service weight and each candidate network The predicted network utility values of each of the candidate networks, and a comprehensive utility value for each of the networks is calculated.
在一种可能的实现方式中，所述装置还包括：第一获取单元，用于获取所述终端设备的第一历史记录信息；所述第一历史记录信息包括应用的访问记录、内容请求偏好、业务开始时间、业务流量、业务特征和使用时长中的一种或多种；第五处理单元，用于根据所述第一历史记录信息进行预测得到所述预测业务属性信息。In a possible implementation manner, the apparatus further includes: a first acquiring unit, configured to acquire first history record information of the terminal device; the first history record information includes application access records, content request preferences , one or more of service start time, service flow, service characteristics and use duration; and a fifth processing unit, configured to perform prediction according to the first historical record information to obtain the predicted service attribute information.
在一种可能的实现方式中，所述第二处理单元，具体用于：根据所述当前业务属性信息得到所述当前业务的服务质量需求；所述服务质量需求包括当前所述终端设备所需要的带宽、时延、抖动和丢包率中的一种或多种；基于所述服务质量和所述用户偏好得到网络效用函数；根据每个所述候选网络的所述当前网络属性信息和所述网络效用函数得到所述当前网络效用值；根据每个所述候选网络的所述预测网络属性信息和所述网络效用函数得到所述预测网络效用值。In a possible implementation manner, the second processing unit is specifically configured to: obtain the service quality requirement of the current service according to the current service attribute information; the service quality requirement includes the current requirements of the terminal device one or more of the bandwidth, delay, jitter and packet loss rate; obtain a network utility function based on the quality of service and the user preference; according to the current network attribute information of each candidate network and all The current network utility value is obtained from the network utility function; the predicted network utility value is obtained according to the predicted network attribute information of each candidate network and the network utility function.
在一种可能的实现方式中，所述第二处理单元，具体用于：基于所述服务质量需求得到L个网络决策属性效用函数；基于所述服务质量和所述用户偏好得到所述L个网络决策属性效用函数中的每个所述网络决策属性效用函数对应的网络决策属性权重；根据所述L个网络决策属性效用函数及其对应的所述网络决策属性权重，确定所述网络效用函数。In a possible implementation manner, the second processing unit is specifically configured to: obtain L network decision attribute utility functions based on the service quality requirement; obtain the L network decision attribute utility functions based on the service quality and the user preference The network decision attribute weight corresponding to each of the network decision attribute utility functions in the network decision attribute utility function; according to the L network decision attribute utility functions and their corresponding network decision attribute weights, the network utility function is determined. .
在一种可能的实现方式中，所述装置还包括：第六处理单元，用于测量N个候选网络的参考信号接收功率或参考信号接收质量；判断所述参考信号接收功率或参考信号接收质量是否满足预评估条件；将满足所述预评估条件的M个候选网络确定为所述M个候选网络。In a possible implementation manner, the apparatus further includes: a sixth processing unit, configured to measure the reference signal received power or reference signal received quality of the N candidate networks; judge the reference signal received power or reference signal received quality Whether the pre-evaluation condition is satisfied; M candidate networks that satisfy the pre-evaluation condition are determined as the M candidate networks.
第四方面，本申请提供了一种智能选网的装置，其特征在于，应用于网络设备，所述网络设备提供S个候选网络；所述S个候选网络为所述终端设备的M个候选网络中的网络；所述装置包括：第一获取单元，用于获取所述S个候选网络中的每个候选网络的第二历史记录信息；所述第二历史记录信息包括每个小区的历史带宽、时延、抖动、丢包率中的一种或多种；第一处理单元，用于根据每个所述第二历史记录信息进行预测得到S个预测网络属性信息。In a fourth aspect, the present application provides a device for intelligent network selection, which is characterized in that, when applied to a network device, the network device provides S candidate networks; the S candidate networks are M candidates for the terminal device A network in a network; the apparatus includes: a first acquisition unit configured to acquire second history record information of each candidate network in the S candidate networks; the second history record information includes the history of each cell One or more of bandwidth, delay, jitter, and packet loss rate; a first processing unit, configured to perform prediction according to each of the second historical record information to obtain S pieces of predicted network attribute information.
在一种可能的实现方式中，所述装置还包括：第二处理单元，用于获取所述终端设备的当前业务属性信息和预测业务属性信息；根据所述当前业务属性信息确定所述当前业务权重，以及根据所述预测业务属性信息确定所述预测业务权重，得到所述终端设备的业务综合权重；所述当前业务属性信息包括当前业务类型、当前业务持续时间和用户偏好中的一种或多种，所述预测业务属性信息包括预测业务类型、预测业务持续时间和所述用户偏好中的一种或多种；第一发送单元，用于向所述终端设备发送所述业务综合权重。In a possible implementation manner, the apparatus further includes: a second processing unit, configured to acquire current service attribute information and predicted service attribute information of the terminal device; determine the current service attribute information according to the current service attribute information weight, and determine the predicted service weight according to the predicted service attribute information to obtain the comprehensive service weight of the terminal device; the current service attribute information includes one of the current service type, the current service duration and the user preference or The predicted service attribute information includes one or more of predicted service type, predicted service duration and the user preference; a first sending unit is configured to send the comprehensive weight of the service to the terminal device.
在一种可能的实现方式中，所述装置还包括：第三处理单元，用于根据每个所候选网络对应的所述当前网络属性信息确定所述当前网络效用值，以及根据每个所述候选网络对应的所述预测网络属性信息确定所述预测网络效用值；所述当前网络属性信息包括网络当前的带宽、时延、抖动、丢包率中的一种或多种；所述预测网络属性信息包括网络未来的带宽、时延、抖动、丢包率中的一种或多种；基于每个所述当前网络效用值和每个所述预测网络效用值得到S个网络效用值；所述网络效用值包括同一个所述候选网络的当前网络效应值和预测网络效用值；第二发送单元，用于向所述终端设备发送所述S个候选网络对应的所述S个网络效用值。In a possible implementation manner, the apparatus further includes: a third processing unit, configured to determine the current network utility value according to the current network attribute information corresponding to each candidate network, and according to each of the The predicted network attribute information corresponding to the candidate network determines the predicted network utility value; the current network attribute information includes one or more of the current network bandwidth, delay, jitter, and packet loss rate; the predicted network The attribute information includes one or more of the future bandwidth, delay, jitter, and packet loss rate of the network; S network utility values are obtained based on each of the current network utility values and each of the predicted network utility values; The network utility value includes the current network effect value and the predicted network utility value of the same candidate network; the second sending unit is configured to send the S network utility values corresponding to the S candidate networks to the terminal device .
第五方面，本申请提供了一种智能选网的方法，其特征在于，应用于网络设备，所述方法包括：确定终端设备的业务综合权重；所述业务综合权重包括所述终端设备的当前业务权重和预测业务权重；所述当前业务权重为基于所述终端设备当前业务属性信息得到的权重，所述预测业务权重为基于所述终端设备预测业务属性信息得到的权重；分别确定所述终端设备的M个候选网络的网络效用值；每个所述网络效用值包括对应网络的当前网络效用值和预测网络效用值；所述当前网络效用值为基于当前网络属性信息得到的值，所述预测网络效用值为基于预测网络属性信息得到的值；基于所述业务综合权重和所述M个候选网络的所述网络效用值，得到M个综合效用值；基于所述M个综合效用值，从所述M个候选网络中选取所述终端设备的目标网络。In a fifth aspect, the present application provides a method for intelligent network selection, which is characterized in that, when applied to a network device, the method includes: determining a comprehensive service weight of a terminal device; the comprehensive service weight includes the current value of the terminal device. Service weight and predicted service weight; the current service weight is the weight obtained based on the current service attribute information of the terminal equipment, and the predicted service weight is the weight obtained based on the predicted service attribute information of the terminal equipment; The network utility values of the M candidate networks of the device; each of the network utility values includes the current network utility value and the predicted network utility value of the corresponding network; the current network utility value is a value obtained based on the current network attribute information, and the The predicted network utility value is a value obtained based on the predicted network attribute information; based on the service comprehensive weight and the network utility value of the M candidate networks, M comprehensive utility values are obtained; based on the M comprehensive utility values, The target network of the terminal device is selected from the M candidate networks.
第六方面，本申请提供了一种计算机存储介质，其特征在于，所述计算机存储介质存储有计算机程序，该计算机程序被处理器执行时实现上述第一方面任意一项所述的方法。In a sixth aspect, the present application provides a computer storage medium, characterized in that, the computer storage medium stores a computer program, and when the computer program is executed by a processor, implements any one of the methods described in the first aspect above.
第七方面，本申请提供了一种计算机存储介质，其特征在于，所述计算机存储介质存储有计算机程序，该计算机程序被处理器执行时实现上述第二方面任意一项所述的方法。In a seventh aspect, the present application provides a computer storage medium, characterized in that, the computer storage medium stores a computer program, and when the computer program is executed by a processor, implements any one of the methods described in the second aspect above.
第八方面，本申请提供了一种计算机存储介质，其特征在于，所述计算机存储介质存储有计算机程序，该计算机程序被处理器执行时实现上述第一方面任意一项所述的方法。In an eighth aspect, the present application provides a computer storage medium, wherein the computer storage medium stores a computer program, and when the computer program is executed by a processor, implements any one of the methods described in the first aspect.
第九方面，本申请提供了一种计算机存储介质，其特征在于，所述计算机存储介质存储有计算机程序，该计算机程序被处理器执行时实现上述第二方面任意一项所述的方法。In a ninth aspect, the present application provides a computer storage medium, characterized in that the computer storage medium stores a computer program, and when the computer program is executed by a processor, implements any one of the methods described in the second aspect above.
第十方面，本发明实施例提供一种电子设备，该电子设备中包括处理器，处理器被配置为支持该电子设备实现第一方面提供的智能选网的方法中相应的功能。该电子设备还可以包括存储器，存储器用于与处理器耦合，其保存该电子设备必要的程序指令和数据。该电子设备还可以包括通信接口，用于该电子设备与其他设备或通信网络通信。In a tenth aspect, an embodiment of the present invention provides an electronic device, the electronic device includes a processor, and the processor is configured to support the electronic device to implement corresponding functions in the method for intelligent network selection provided in the first aspect. The electronic device may also include a memory for coupling with the processor that holds program instructions and data necessary for the electronic device. The electronic device may also include a communication interface for the electronic device to communicate with other devices or a communication network.
第十一方面，本发明实施例提供一种电子设备，该电子设备中包括处理器，处理器被配置为支持该电子设备实现第二方面提供的智能选网的方法中相应的功能。该电子设备还可以包括存储器，存储器用于与处理器耦合，其保存该电子设备必要的程序指令和数据。该电子设备还可以包括通信接口，用于该电子设备与其他设备或通信网络通信。In an eleventh aspect, an embodiment of the present invention provides an electronic device, the electronic device includes a processor, and the processor is configured to support the electronic device to implement corresponding functions in the method for intelligent network selection provided in the second aspect. The electronic device may also include a memory for coupling with the processor that holds program instructions and data necessary for the electronic device. The electronic device may also include a communication interface for the electronic device to communicate with other devices or a communication network.
第十二方面，本申请提供了一种芯片***，该芯片***包括处理器，用于支持电子设备实现上述第一方面中所涉及的功能，例如，生成或处理上述智能选网的方法中所涉及的信息。在一种可能的设计中，所述芯片***还包括存储器，所述存储器，用于保存电子设备必要的程序指令和数据。该芯片***，可以由芯片构成，也可以包含芯片和其他分立器件。In a twelfth aspect, the present application provides a chip system, the chip system includes a processor for supporting an electronic device to implement the functions involved in the above-mentioned first aspect, for example, generating or processing the above-mentioned method for intelligent network selection. information involved. In a possible design, the chip system further includes a memory for storing necessary program instructions and data of the electronic device. The chip system may be composed of chips, or may include chips and other discrete devices.
第十三方面，本申请提供了一种芯片***，该芯片***包括处理器，用于支持电子设备实现上述第二方面中所涉及的功能，例如，生成或处理上述智能选网的方法中所涉及的信息。在一种可能的设计中，所述芯片***还包括存储器，所述存储器，用于保存电子设备必要的程序指令和数据。该芯片***，可以由芯片构成，也可以包含芯片和其他分立器件。In a thirteenth aspect, the present application provides a chip system, the chip system includes a processor for supporting an electronic device to implement the functions involved in the above second aspect, for example, generating or processing the above method for intelligent network selection. information involved. In a possible design, the chip system further includes a memory for storing necessary program instructions and data of the electronic device. The chip system may be composed of chips, or may include chips and other discrete devices.
附图说明Description of drawings
图1A为本发明实施例提供的一种无线通信***架构示意图。FIG. 1A is a schematic diagram of an architecture of a wireless communication system according to an embodiment of the present invention.
图1B为本发明实施例提供的一种智能选网的架构示意图。FIG. 1B is a schematic structural diagram of an intelligent network selection provided by an embodiment of the present invention.
图1C为现有技术中的一种不同业务服务质量需求示意图。FIG. 1C is a schematic diagram of service quality requirements for different services in the prior art.
图1D为本发明实施例提供的一种预测用户业务模型示意图。FIG. 1D is a schematic diagram of a predicted user service model according to an embodiment of the present invention.
图2A为本申请实施例中的一种智能选网方法的流程示意图。FIG. 2A is a schematic flowchart of an intelligent network selection method in an embodiment of the present application.
图2B为本申请实施例中的另一种智能选网方法的流程示意图。FIG. 2B is a schematic flowchart of another method for intelligent network selection in an embodiment of the present application.
图3A为本发明实施例提供的一种智能选网方法示例图流程图。FIG. 3A is a flowchart illustrating an example of an intelligent network selection method provided by an embodiment of the present invention.
图3B为本发明实施例提供的一种确定网络效用函数方法示例性流程图。FIG. 3B is an exemplary flowchart of a method for determining a network utility function according to an embodiment of the present invention.
图4A为本发明实施例提供的本申请提供了一种智能选网的装置示意图。FIG. 4A is a schematic diagram of an apparatus for intelligent network selection provided by the present application according to an embodiment of the present invention.
图4B为本发明实施例提供的本申请提供了一种智能选网的装置示意图。FIG. 4B is a schematic diagram of an apparatus for intelligent network selection provided by the present application according to an embodiment of the present invention.
具体实施方式Detailed ways
下面将结合本发明实施例中的附图，对本发明实施例进行描述。The embodiments of the present invention will be described below with reference to the accompanying drawings in the embodiments of the present invention.
本申请的说明书和权利要求书及所述附图中的术语“第一”、“第二”、“第三”和“第四”等是用于区别不同对象，而不是用于描述特定顺序。此外，术语“包括”和“具有”以及它们任何变形，意图在于覆盖不排他的包含。例如包含了一系列步骤或单元的过程、方法、***、产品或设备没有限定于已列出的步骤或单元，而是可选地还包括没有列出的步骤或单元，或可选地还包括对于这些过程、方法、产品或设备固有的其它步骤或单元。The terms "first", "second", "third" and "fourth" in the description and claims of the present application and the drawings are used to distinguish different objects, rather than to describe a specific order . Furthermore, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device comprising a series of steps or units is not limited to the listed steps or units, but optionally also includes unlisted steps or units, or optionally also includes For other steps or units inherent to these processes, methods, products or devices.
在本文中提及“实施例”意味着，结合实施例描述的特定特征、结构或特性可以包含在本申请的至少一个实施例中。在说明书中的各个位置出现该短语并不一定均是指相同的实施例，也不是与其它实施例互斥的独立的或备选的实施例。本领域技术人员显式地和隐式地理解的是，本文所描述的实施例可以与其它实施例相结合。Reference herein to an "embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the present application. The appearances of the phrase in various places in the specification are not necessarily all referring to the same embodiment, nor a separate or alternative embodiment that is mutually exclusive of other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.
下面结合附图对本申请的实施例进行描述。The embodiments of the present application will be described below with reference to the accompanying drawings.
基于上述提出的技术问题为了便于理解本发明实施例，下面先对本发明实施例所基于的***架构进行描述。请参考见图1A，图1A是本发明实施例提供的一种无线通信***架构示意图，该无线通信***100用于为终端设备智能选择最优网络，提升用户体验。该***架构中可以包括终端设备101和网络设备102。其中，Based on the technical problems raised above, in order to facilitate the understanding of the embodiments of the present invention, the following first describes the system architecture on which the embodiments of the present invention are based. Please refer to FIG. 1A . FIG. 1A is a schematic diagram of the architecture of a wireless communication system according to an embodiment of the present invention. The wireless communication system 100 is used for intelligently selecting an optimal network for a terminal device to improve user experience. The system architecture may include a terminal device 101 and a network device 102 . in,
终端设备101，可以分布在整个无线通信***100中，可以是静止的，也可以是移动的。在本申请的一些实施例中，终端设备101可以是移动设备、移动台(mobile station)、移动单元(mobile unit)、M2M终端、无线单元，远程单元、终端代理、移动客户端等等。常见的移动终端包括智能手机、智能手表、平板电脑等设备。在未来通信***中，终端设备101也可以为本申请中的智能选网装置，或者被配置为包含有本申请中的智能选网装置的终端设备。当终端设备101接入网络设备102提供的网络时，终端设备101可通过网络使用不同的业务。例如，当智能手机处于基站1的A小区、基站3的B小区和基站2的C小区覆盖交叠区域时，智能手机可从三个小区中选择一个小区接入，然后使用该小区所提供的网络进行相应的业务(如语音业务，视频通话业务等)。The terminal devices 101 may be distributed in the entire wireless communication system 100, and may be stationary or mobile. In some embodiments of the present application, the terminal device 101 may be a mobile device, a mobile station, a mobile unit, an M2M terminal, a wireless unit, a remote unit, a terminal agent, a mobile client, and the like. Common mobile terminals include devices such as smartphones, smart watches, and tablet computers. In a future communication system, the terminal device 101 may also be the intelligent network selection device in this application, or a terminal device configured to include the intelligent network selection device in this application. When the terminal device 101 accesses the network provided by the network device 102, the terminal device 101 can use different services through the network. For example, when the smartphone is in the overlapping area covered by cell A of base station 1, cell B of base station 3, and cell C of base station 2, the smartphone can select one of the three cells to access, and then use the The network performs corresponding services (such as voice services, video calling services, etc.).
网络设备102，可以分布在整个通信***100中，可以为基站，如基站1、基站2、基站 3，基站可以是时分同步码分多址(Time Division Synchronous Code Division Multiple Access，TD-SCDMA)***中的基站收发台(Base Transceiver Station，BTS)，也可以是LTE***中的演进型基站(Evolutional Node B，eNB)，以及5G***、新空口(NR)***中的基站。另外，基站也可以为接入点(Access Point，AP)、传输节点(Trans TRP)、中心单元(Central Unit，CU)或其他网络实体，并且可以包括以上网络实体的功能中的一些或所有功能。需要说明的是，每个基站都可划分为一个或多个小区，且每个小区都能够为终端设备101提供网络。当终端设备101位于多个小区覆盖交叠区域时，可以选择一个目标小区选择接入，从而可以使用该小区提供的网络进行相应的业务。例如，当智能手机处于基站1的A小区、基站3的B小区和基站2的C小区覆盖交叠区域时，智能手机可以通过本发明实施例提供的智能选网的方法在三个小区中选取性能最好的小区接入，从而选择出最优网络。The network device 102 can be distributed in the entire communication system 100, and can be a base station, such as base station 1, base station 2, and base station 3, and the base station can be a time division synchronous code division multiple access (Time Division Synchronous Code Division Multiple Access, TD-SCDMA) system The base transceiver station (Base Transceiver Station, BTS) in the LTE system can also be an evolved base station (Evolutional Node B, eNB) in the LTE system, and the base station in the 5G system and the New Radio (NR) system. In addition, the base station may also be an access point (Access Point, AP), a transmission node (Trans TRP), a central unit (Central Unit, CU) or other network entities, and may include some or all of the functions of the above network entities. . It should be noted that each base station can be divided into one or more cells, and each cell can provide a network for the terminal device 101 . When the terminal device 101 is located in an overlapping area covered by multiple cells, a target cell can be selected for access, so that the network provided by the cell can be used to perform corresponding services. For example, when the smartphone is in the overlapping area covered by cell A of base station 1, cell B of base station 3, and cell C of base station 2, the smartphone can select among the three cells by using the intelligent network selection method provided in the embodiment of the present invention. The cell with the best performance is accessed to select the optimal network.
可以理解的是，图1A中的一种智能选网的***架构只是本申请实施例中的一种示例性的实施方式，本申请实施例中的智能选网***架构包括但不仅限于以上***架构。It can be understood that the system architecture of a kind of intelligent network selection in FIG. 1A is only an exemplary implementation in the embodiment of the application, and the intelligent network selection system architecture in the embodiment of the application includes but is not limited to the above system architecture. .
基于上述提出的技术问题为了便于理解本发明实施例，下面先对本发明实施例所基于的架构进行描述。请参考见图1B，图1B是本发明实施例提供的一种智能选网的架构示意图，该智能选网架构用于为终端设备智能选择最优网络，提升用户体验。该架构中可以包括业务识别模块201、业务预测模块202、策略映射模块203、意图映射模块204和分析决策模块205。其中，Based on the technical problems raised above, in order to facilitate the understanding of the embodiments of the present invention, the architecture on which the embodiments of the present invention are based is first described below. Please refer to FIG. 1B . FIG. 1B is a schematic diagram of the architecture of an intelligent network selection provided by an embodiment of the present invention. The intelligent network selection architecture is used to intelligently select an optimal network for a terminal device to improve user experience. The architecture may include a service identification module 201 , a service prediction module 202 , a policy mapping module 203 , an intention mapping module 204 and an analysis decision module 205 . in,
业务识别模块201，用于根据用户在终端设备上发起的业务的标识，如APP的ID、业务类型ID，或者根据业务包的长度、业务包到达间隔等识别业务的类型，从而输出当前业务的QoS需求，以5G为例，用一组整型值来表示业务不同的服务等级，包括优先级、时延、误包率等参数。如图1C所示，图1C为现有技术中的一种不同业务服务质量需求示意图，图中，比如会话语音业务的QoS需求包括：优先级是20，分组时延预算是100ms，误包率是10 -2；会话视频业务的QoS需求包括：优先级是40，分组时延预算是150ms，误包率是10 -3；实时游戏业务的QoS需求包括：优先级是30，分组时延预算是50ms，误包率是10 -3。 The service identification module 201 is used to identify the type of service according to the identification of the service initiated by the user on the terminal device, such as the ID of the APP, the service type ID, or according to the length of the service package, the arrival interval of the service package, etc., thereby outputting the current service. QoS requirements, taking 5G as an example, use a set of integer values to represent different service levels of a business, including parameters such as priority, delay, and packet error rate. As shown in FIG. 1C, FIG. 1C is a schematic diagram of service quality requirements of different services in the prior art. In the figure, for example, the QoS requirements of the conversational voice service include: the priority is 20, the packet delay budget is 100ms, the packet error rate is 10-2 ; the QoS requirements of the session video service include: the priority is 40, the packet delay budget is 150ms, and the packet error rate is 10-3 ; the QoS requirements of the real-time game service include: the priority is 30, the packet delay budget is is 50ms, and the packet error rate is 10 -3 .
业务预测模块202，用于预测用户未来一段时间内的业务类型和使用时长。如图1D所示，图1D为本发明实施例提供的一种预测用户业务模型示意图，图中业务预测模块的输入包括用户对应用的访问记录、内容请求偏好和使用时长统计等信息，输出为预测的业务类型和使用时长。可选的，终端设备可根据时空维度上用户的移动(如在不同时间的位置信息)和业务习惯的历史数据(如不同时间使用的APP或业务信息)，通过线下监督学习训练神经网络算法，预测用户未来一段时间的业务需求，预测的业务信息中可包括何时何地进行何种业务、业务持续时间等，并可结合终端设备的本地信息(如电量，用户偏好等)，获得用户预测业务属性信息。用户预测业务属性信息可包括用户在未来一段时间内的业务需求(如业务类型，业务持续时间等)和设备状态(如电量，用户偏好等)。The service prediction module 202 is used for predicting the service type and usage time of the user in a certain period of time in the future. As shown in FIG. 1D, FIG. 1D is a schematic diagram of a predictive user service model provided by an embodiment of the present invention. The input of the service prediction module in the figure includes information such as the user's access record to the application, content request preference, and usage duration statistics, and the output is Predicted business type and usage duration. Optionally, the terminal device can train the neural network algorithm through offline supervised learning according to the user's movement in the space-time dimension (such as location information at different times) and historical data of business habits (such as APP or business information used at different times). , predict the user's business needs for a period of time in the future. The predicted business information can include when and where to carry out what kind of business, business duration, etc., and can be combined with the local information of the terminal device (such as electricity, user preferences, etc.) to obtain user Predict business attribute information. The user-predicted service attribute information may include the user's service requirements (such as service type, service duration, etc.) and device status (such as power level, user preference, etc.) in a future period of time.
策略映射模块203，用于确定业务所需的网络决策属性效用函数；确定现在和将来的业务综合权重。可选的，终端设备可根据用户的QoS需求，确定业务所需的网络性能指标效用函数，即网络决策属性效用函数；根据用户当前业务属性信息和用户预测业务属性信息，确定现在和将来的业务综合权重。例如某一用户在看视频，预测到该用户未来一个小时内会进行云AR/VR游戏，由于云AR/VR游戏相对视频的QoS要求更高，因此，将来的业务综合权重需要比现在的业务权重大，比如可确定业务综合权重为[0.3,0.7]，表示相对于未来的网络性能和用户状态，用户对网络当前所提供服务的需求更低，对终端设备当前的资源需求如能 量需求更低。又例如，对于另一正在进行游戏业务的用户来说，预测到该用户未来十几分钟内会保持游戏业务，根据历史游戏业务信息可确认用户对未来的游戏业务的偏好使得对终端设备将来的资源需求(如能量需求)更低，因此，将来的业务综合权重可比现在的业务权重小，比如可确定业务综合权重为[0.6,0.4]。对于不同用户和不同业务，业务综合权重动态变化，用于在分析决策模块对各小区当前与未来的效用值进行加权综合计算。在本发明实施例中，策略映射模块加入了对用户业务需求和用户状态的预测，综合考虑用户当前和未来对业务的需求，得到动态变化的业务综合权重，有利于满足用户的不同需求，保证用户服务的连续性。The policy mapping module 203 is used to determine the network decision attribute utility function required by the service; and to determine the comprehensive weight of the current and future services. Optionally, the terminal device can determine the network performance index utility function required by the service according to the QoS requirements of the user, that is, the network decision attribute utility function; according to the user's current service attribute information and the user's predicted service attribute information, determine the current and future services. Comprehensive weight. For example, a user is watching a video, and it is predicted that the user will play cloud AR/VR games in the next hour. Since cloud AR/VR games have higher QoS requirements than videos, the comprehensive weight of future services needs to be higher than that of current services. The weight is heavy, for example, it can be determined that the comprehensive weight of the service is [0.3, 0.7], which means that compared with the future network performance and user status, the user's demand for the services currently provided by the network is lower, and the current resource demand of terminal equipment, such as energy demand, is higher. Low. For another example, for another user who is in the game business, it is predicted that the user will maintain the game business in the next ten minutes, and the user's preference for the future game business can be confirmed according to the historical game business information. Resource requirements (such as energy requirements) are lower, therefore, the comprehensive weight of future services may be smaller than the current weight of services. For example, the comprehensive weight of services can be determined to be [0.6, 0.4]. For different users and different services, the comprehensive weight of services changes dynamically, and is used to perform weighted comprehensive calculation on the current and future utility values of each cell in the analysis and decision module. In the embodiment of the present invention, the policy mapping module adds the prediction of user service requirements and user status, comprehensively considers the user's current and future service requirements, and obtains a dynamically changing service comprehensive weight, which is beneficial to meet the different needs of users and ensure that Continuity of user service.
意图映射模块204，用于确定对应网络决策属性权重。可选的，终端设备可根据用户的QoS需求和用户偏好，确定对应网络决策属性权重。一方面，用户在进行语音电话、游戏、看直播等不同业务时，对网络性能的需求是不同的。另一方面，不同的用户对同一种业务的通信服务偏好也是不同，有的用户偏好极致的业务体验，可以忽略网络流量的费用，追求大带宽高速率的业务体验，而有的用户则偏好低价的可用服务，对带宽速率的追求不大。因此，终端设备可根据业务意图进行快速、主动的智能网络接入选择，需要根据不同业务类型所对应的QoS需求和用户偏好对各网络决策属性分配权重。The intent mapping module 204 is configured to determine the corresponding network decision attribute weight. Optionally, the terminal device may determine the corresponding network decision attribute weight according to the user's QoS requirement and user preference. On the one hand, users have different requirements for network performance when conducting different services such as voice calls, games, and watching live broadcasts. On the other hand, different users have different communication service preferences for the same service. Some users prefer the ultimate service experience, ignore the cost of network traffic, and pursue the service experience with large bandwidth and high speed, while some users prefer low service experience. Available services at a low price, the pursuit of bandwidth speed is not large. Therefore, the terminal device can perform fast and active intelligent network access selection according to the service intention, and it is necessary to assign weights to each network decision attribute according to the QoS requirements and user preferences corresponding to different service types.
分析决策模块205，可根据策略映射模块所提供的各个网络决策属性效用函数和意图映射模块所提供的网络决策属性权重分配，对各单元效用函数进行组合，得到多元效用函数(例如，对单元效用函数求权重次幂，再连乘求积)；再将获得的各网络决策属性信息代入计算，得到网络中各小区当前的总体效用值和未来的总体效用值；然后根据策略映射模块所提供的业务综合权重对各小区当前和未来的效用值进行组合(如加权求和)，经过计算后选择当前与未来的综合效用值最大的小区，即与用户业务意图最匹配的小区。终端设备所获得的各网络属性信息，包括当前网络属性信息和预测网络属性信息，是基站根据终端设备的请求，下发给终端设备的。例如，基站通过广播消息下发的存放在邻小区列表(Neighbor Cell List，NCL)里的网络属性信息(如包含了网络环境中各小区当前的网络属性信息和所预测的未来的网络属性信息)。网络属性信息为各个小区的各个网络属性的当前值和预测值，如下表：The analysis and decision module 205 can combine the utility functions of each unit according to the utility function of each network decision attribute provided by the strategy mapping module and the weight distribution of the network decision attribute provided by the intention mapping module to obtain a multivariate utility function (for example, for the unit utility function). The function calculates the weight to the power, and then multiplies to calculate the product); then substitute the obtained network decision attribute information into the calculation to obtain the current overall utility value and the future overall utility value of each cell in the network; and then according to the policy mapping module. The comprehensive service weight combines the current and future utility values of each cell (eg, weighted summation), and selects the cell with the largest current and future comprehensive utility value after calculation, that is, the cell that best matches the user's business intention. The various network attribute information obtained by the terminal equipment, including the current network attribute information and the predicted network attribute information, are delivered by the base station to the terminal equipment according to the request of the terminal equipment. For example, the network attribute information stored in the Neighbor Cell List (NCL) delivered by the base station through a broadcast message (for example, including the current network attribute information and predicted future network attribute information of each cell in the network environment) . The network attribute information is the current value and predicted value of each network attribute of each cell, as shown in the following table:
表一网络属性信息的当前值和预测值示意表Table 1 Schematic representation of current and predicted values of network attribute information
网络属性信息Network attribute information 服务小区 service area 邻区1Neighborhood 1 …… 邻区NNeighborhood N
当前带宽current bandwidth BW 1 BW 1 BW 2 BW 2    BW N-1 BW N-1
预测带宽forecast bandwidth BW 1’ BW 1 ' BW 2’ BW 2 '    BW N-1’ BW N-1 '
……            
……            
当前时延current delay T 1 T1 T 2 T 2    T N-1 T N-1
预测时延prediction delay T 1’ T 1 ' T 2’ T 2 '    T N-1’ T N-1 '
分析决策模块可通过网络属性信息获得各小区的相应的网络属性(如当前带宽、预测带宽等)的值，从而可以计算出各小区的综合效用值。The analysis and decision module can obtain the values of the corresponding network attributes (such as current bandwidth, predicted bandwidth, etc.) of each cell through the network attribute information, so as to calculate the comprehensive utility value of each cell.
可以理解的是，图1B中的一种智能选网的架构只是本申请实施例中的一种示例性的实施方式，本申请实施例中的智能选网架构包括但不仅限于以上***架构。It can be understood that the architecture of an intelligent network selection in FIG. 1B is only an exemplary implementation in the embodiments of the present application, and the intelligent network selection architecture in the embodiments of the present application includes but is not limited to the above system architecture.
下面对本发明实施例所基于的具体方法架构进行描述。参见图2A，图2A是本申请实施例中的一种智能选网方法的流程示意图，下面将结合附图2A并基于上述图1A中的智能选网***架构从终端设备和网络设备的交互侧对本申请实施例中的智能选网方法进行描述。需要说明的是，为了更详细的描述本申请实施例中的智能选网方法，本申请在各个流程步骤中描述了相应的执行主体分别为终端设备或网络设备，但不代表本申请实施例只能通过所描述的执行主体进行对应的方法流程。The specific method architecture on which the embodiments of the present invention are based is described below. Referring to Fig. 2A, Fig. 2A is a schematic flowchart of an intelligent network selection method in the embodiment of the present application, below will be combined with Fig. 2A and based on the intelligent network selection system architecture in the above-mentioned Fig. 1A from the interactive side of terminal equipment and network equipment The intelligent network selection method in the embodiment of the present application is described. It should be noted that, in order to describe the intelligent network selection method in the embodiments of the present application in more detail, the present application describes in each process step that the corresponding execution subject is a terminal device or a network device, respectively, but this does not mean that the embodiments of the present application only The corresponding method flow can be performed through the described execution body.
步骤S301：终端设备确定所述终端设备的业务综合权重。Step S301: The terminal device determines the comprehensive service weight of the terminal device.
所述业务综合权重包括所述终端设备的当前业务权重和预测业务权重；所述当前业务权重为基于所述终端设备当前业务属性信息得到的权重，所述预测业务权重为基于所述终端设备预测业务属性信息得到的权重。具体地，终端设备可以理解为智能手机、智能手表、平板电脑等移动设备；业务综合权重可以理解为用于表示当前业务和预测业务分别对网络资源的需求；当前业务权重可以理解为终端设备当前运行的业务对网络资源的需求；预测业务权重可以理解为终端设备未来一段时间可能运行的业务对网络资源的需求；当前业务属性信息可以理解为终端设备当前运行的业务类型和当前业务持续时间等信息；预测业务属性信息可以理解为终端设备未来可能运行的业务类型和预测业务持续时间等信息。例如，某一用户当下正在使用视频软件看视频，预测到该用户未来一个小时内会进行化AR/VR游戏，由于化AR/VR游戏相对视频对网络的服务质量需求更高，因此，将来的业务综合权重需要比现在的业务权重大，比如可确定业务综合权重为[0.3,0.7]，表示相对于未来的网络性能和用户状态，用户对网络当前所提供服务质量需求更低，也可以理解为终端设备当前所需要的网络资源(如能量需求)比未来所需要的网络资源低。The service comprehensive weight includes the current service weight and the predicted service weight of the terminal equipment; the current service weight is a weight obtained based on the current service attribute information of the terminal equipment, and the predicted service weight is based on the terminal equipment prediction. The weight obtained by the business attribute information. Specifically, terminal devices can be understood as mobile devices such as smart phones, smart watches, and tablet computers; the comprehensive service weight can be understood as representing the network resource requirements of the current service and the predicted service respectively; the current service weight can be understood as the current service weight of the terminal device. The demand for network resources of the running service; the predicted service weight can be understood as the demand for network resources of the service that the terminal device may run in a period of time in the future; the current service attribute information can be understood as the current service type and current service duration of the terminal device. Information; predicted service attribute information can be understood as information such as the type of service that the terminal device may run in the future and the duration of the predicted service. For example, a user is currently using video software to watch videos, and it is predicted that the user will play AR/VR games in the next hour. Since AR/VR games have higher requirements for network service quality than videos, future The comprehensive weight of the business needs to be larger than the current business weight. For example, it can be determined that the comprehensive weight of the business is [0.3, 0.7], which means that compared with the future network performance and user status, the user has a lower demand for the service quality currently provided by the network, which is understandable. The network resources (eg energy requirements) currently required for the terminal device are lower than the network resources required in the future.
在一种可能的实现方式中，所述方法还包括：终端设备获取所述终端设备的第一历史记录信息；所述第一历史记录信息包括应用的访问记录、内容请求偏好、业务开始时间、业务流量、业务特征和使用时长中的一种或多种；根据所述第一历史记录信息进行预测得到所述预测业务属性信息。具体地，第一历史记录信息可以理解为终端设备之前的历史应用访问记录、用户内容请求偏好、应用使用时长等历史信息；应用的访问记录可以理解为在不同时间用户使用应用的记录；内容请求偏好可以理解为在不同时间用户对内容的请求偏好；使用时长为可以理解为业务持续时间；业务特征可以包括相应业务对应的网络切片，数据网络名称DNN或者接入点名称APN等；预测业务属性信息可以理解为基于终端设备历史记录信息通过预测算法(如神经网络算法)得到的终端设备未来一段时间可能运行的业务以及业务可能持续的时间等。例如，终端设备可根据时空维度上用户的移动(如在不同时间的位置信息)和业务习惯的历史数据(如不同时间使用的APP或业务信息)，通过线下监督学习训练神经网络算法，预测用户未来一段时间的业务需求，预测的业务信息中可包括何时何地进行何种业务、业务持续时间等，并可结合终端设备的本地信息(如电量，用户偏好等)，获得用户预测业务属性信息。用户预测业务属性信息可包括用户在未来一段时间内的业务需求(如业务类型，业务持续时间等)和设备状态(如电量，用户偏好等)。In a possible implementation manner, the method further includes: the terminal device obtains first history record information of the terminal device; the first history record information includes application access records, content request preferences, service start time, One or more of service flow, service characteristics, and usage duration; the predicted service attribute information is obtained by predicting according to the first historical record information. Specifically, the first historical record information can be understood as historical information such as historical application access records, user content request preferences, and application usage duration before the terminal device; application access records can be understood as records of users using applications at different times; content requests Preference can be understood as the user's request preference for content at different times; usage duration can be understood as service duration; service features can include network slices corresponding to corresponding services, data network name DNN or access point name APN, etc.; predicted service attributes The information can be understood as the possible operation of the terminal device for a period of time in the future and the possible duration of the service obtained through a prediction algorithm (such as a neural network algorithm) based on the terminal device's historical record information. For example, the terminal device can train a neural network algorithm through offline supervised learning based on the user's movement in time and space (such as location information at different times) and historical data of business habits (such as APP or business information used at different times) to predict The user's business needs for a period of time in the future, the predicted business information can include when and where to carry out what kind of business, business duration, etc., and can be combined with the local information of the terminal device (such as electricity, user preferences, etc.) to obtain the user's predicted business property information. The user-predicted service attribute information may include the user's service requirements (such as service type, service duration, etc.) and device status (such as power level, user preference, etc.) in a future period of time.
需要说明的是，在上述步骤S301中所涉及的确定所述终端设备的业务综合权重的实现可以包括如下两种实施方式：It should be noted that the implementation of determining the comprehensive service weight of the terminal device involved in the above step S301 may include the following two implementations:
实施方式一：由终端设备生成业务综合权重。Embodiment 1: The terminal device generates the comprehensive weight of the service.
在一种可能的实现方式中，所述终端设备确定所述终端设备的业务综合权重，包括：终端设备根据所述终端设备的所述当前业务属性信息得到所述当前业务权重，以及根据所述终 端设备的所述预测业务属性信息得到所述预测业务权重；所述当前业务属性信息包括当前业务类型、当前业务持续时间和用户偏好中的一种或多种，所述预测业务属性信息包括预测业务类型、预测业务持续时间和所述用户偏好中的一种或多种。具体地，当前业务属性信息可以理解为终端设备当前运行的业务以及当前业务持续的时间和用户偏好等信息；预测业务属性信息可以理解为基于终端设备历史记录信息通过预测算法(如神经网络算法)得到的终端设备未来一段时间可能运行的业务以及业务可能持续的时间等信息。终端设备可基于当前业务属性信息得到当前业务权重，基于预测业务属性信息得到预测业务权重，例如，对于正在进行游戏业务的用户来说，若预测到该用户未来十几分钟内会保持游戏业务，根据历史游戏业务信息可确认用户对未来的游戏业务的偏好使得对终端设备将来的资源需求(如能量需求)更低，因此，将来的业务综合权重可比现在的业务权重小，比如可确定业务综合权重为[0.6,0.4]。In a possible implementation manner, the terminal device determining the comprehensive service weight of the terminal device includes: obtaining the current service weight by the terminal device according to the current service attribute information of the terminal device, and obtaining the current service weight according to the current service attribute information of the terminal device; The predicted service attribute information of the terminal device obtains the predicted service weight; the current service attribute information includes one or more of current service type, current service duration and user preference, and the predicted service attribute information includes prediction One or more of service type, predicted service duration, and said user preference. Specifically, the current service attribute information can be understood as information such as the service currently running on the terminal device, the duration of the current service, and user preferences; the predicted service attribute information can be understood as a prediction algorithm (such as a neural network algorithm) based on the historical record information of the terminal device. The obtained information, such as the services that the terminal equipment may run in the future, and the possible duration of the services. The terminal device can obtain the current service weight based on the current service attribute information, and obtain the predicted service weight based on the predicted service attribute information. According to the historical game service information, it can be confirmed that the user's preference for the future game service makes the future resource demand (such as energy demand) of the terminal device lower. Therefore, the comprehensive weight of the future service can be smaller than the current service weight. For example, it can be determined that the comprehensive weight of the service The weights are [0.6, 0.4].
实施方式二：由网络设备生成业务综合权重。Embodiment 2: The comprehensive weight of the service is generated by the network device.
在一种可能的实现方式中，所述方法还包括：网络设备获取所述终端设备的当前业务属性信息和预测业务属性信息；根据所述当前业务属性信息确定所述当前业务权重，以及根据所述预测业务属性信息确定所述预测业务权重，得到所述终端设备的业务综合权重；所述当前业务属性信息包括当前业务类型、当前业务持续时间和用户偏好中的一种或多种，所述预测业务属性信息包括预测业务类型、预测业务持续时间和所述用户偏好中的一种或多种；向所述终端设备发送所述业务综合权重。在本发明实施中，终端设备可将当前业务属性信息(如当前业务类型、当前业务持续时间等)和预测业务属性信息(如预测业务类型、预测业务持续时间等)发送给提供多个候选网络的一个或多个网络设备中的一个设备，然后由该网络设备来计算综合业务权重。接下来网络设备再将得到的综合业务权重发送给终端设备，从而终端设备可基于该综合业务权重选出目标网络，提升了通信性能指标，实现用户个性化业务体验。In a possible implementation manner, the method further includes: acquiring the current service attribute information and predicted service attribute information of the terminal device by the network device; determining the current service weight according to the current service attribute information, and The predicted service attribute information determines the predicted service weight, and obtains the comprehensive service weight of the terminal device; the current service attribute information includes one or more of the current service type, current service duration and user preference, the The predicted service attribute information includes one or more of the predicted service type, the predicted service duration and the user preference; and the service comprehensive weight is sent to the terminal device. In the implementation of the present invention, the terminal device can send current service attribute information (such as current service type, current service duration, etc.) and predicted service attribute information (such as predicted service type, predicted service duration, etc.) to the network providing multiple candidate networks One of the one or more network devices, and then the network device calculates the integrated service weight. Next, the network device sends the obtained comprehensive service weight to the terminal device, so that the terminal device can select the target network based on the comprehensive service weight, which improves the communication performance index and realizes the user's personalized service experience.
在一种可能的实现方式中，所述终端设备确定所述终端设备的业务综合权重，包括：所述终端设备向网络设备发送所述终端设备的所述当前业务属性信息和所述预测业务属性信息；所述网络设备为与M个候选网络中的一个或多个候选网络关联的网络设备；接收所述网络设备发送的所述业务综合权重。具体地，当终端设备处于连接网络状态时，M个候选网络可以理解为终端设备当前接入的网络以及其相邻的网络，例如，4G网络、5G网络、蜂窝网络、WiFi网络等；当终端设备处于未连接网络状态时，M个候选网络可以理解为终端设备所处区域能够连接的网络。需要说明的是，终端设备处于连接网络状态，可以理解为终端与核心网之间存在信令连接(控制面连接)或者存在信令连接与数据传输的连接(用户面连接)，包括UE与基站之间的空口连接和基站与核心网之间的连接，处于连接网络状态的终端，可以直接与网络进行交互建立用于数据传输的连接或者在已有的数据传输连接上直接进行数据传输。对应地，终端设备处于未连接网络状态，可以理解为终端与网络之间不存在信令连接，或者存在信令连接但处于未激活的状态，终端不能直接与核心网进行信令交互或者数据传输，当终端需要与核心网进行信令交互或者数据传输时，需要先执行从未激活状态到激活状态的转变。候选网络关联的网络设备可以理解为当前接入网络的相邻网络关联的网络架构中涉及到的网络设备，也可以理解为提供候选网络的网络设备，即为候选网络提供网络服务的设备，例如提供候选网络的接入网设备(4G、5G以及未来网络的基站设备，或者ePDG、N3IWF以及未来网络的一些接入网设备)、核心网设备(4G中的MME、Serving-GW、P-GW等，或者5G中的AMF、SMF、UPF等，以及未来网络中的核心网设备)，同理，当前接入网络关 联的网络设备可以理解为当前接入网络的网络架构中涉及到的网络设备，也可以理解为提供当前接入网络的网络设备，即为当前接入的网络提供网络服务的设备，例如提供当前接入的接入网设备(4G、5G以及未来网络的基站设备，或者ePDG、N3IWF以及未来网络的一些接入网设备)、核心网设备(4G中的MME、Serving-GW、P-GW等，或者5G中的AMF、SMF、UPF等，以及未来网络中的核心网设备)。举例来说，所述终端当前接入网络A，对应的接入网设备为A1，核心网设备为A2，候选的网络为B、C，对应的接入网设备为B1、C1，核心网设备为B2、C2。则当前接入的网络所关联的网络设备为A1或者A2，候选的网络所关联的网络设备为B1、B2、C1、C2。在本发明实施例中，终端设备可接收由多个候选网络各自的网络设备所计算的本网络的网络效用值，然后终端设备可基于这些网络效用值从多个候选网络中选择出目标接入网络，降低了终端设备对数据处理能力的要求，且选出的目标网络可以保证通信服务的稳定性，从而减少终端设备频繁切换网络，提升了通信性能指标，实现用户个性化业务体验。In a possible implementation manner, the terminal device determining the comprehensive service weight of the terminal device includes: the terminal device sending the current service attribute information and the predicted service attribute of the terminal device to a network device information; the network device is a network device associated with one or more candidate networks in the M candidate networks; and the service comprehensive weight sent by the network device is received. Specifically, when the terminal device is in a network connection state, the M candidate networks can be understood as the network currently accessed by the terminal device and its adjacent networks, such as 4G network, 5G network, cellular network, WiFi network, etc.; When the device is not connected to the network, the M candidate networks can be understood as the networks that can be connected to the area where the terminal device is located. It should be noted that when the terminal device is in the network connection state, it can be understood that there is a signaling connection (control plane connection) between the terminal and the core network, or there is a signaling connection and data transmission connection (user plane connection), including the UE and the base station. Between the air interface connection and the connection between the base station and the core network, the terminal in the network connection state can directly interact with the network to establish a connection for data transmission or directly perform data transmission on the existing data transmission connection. Correspondingly, if the terminal device is not connected to the network, it can be understood that there is no signaling connection between the terminal and the network, or there is a signaling connection but is in an inactive state, and the terminal cannot directly perform signaling interaction or data transmission with the core network. , when the terminal needs to perform signaling interaction or data transmission with the core network, it needs to perform the transition from the inactive state to the active state first. The network device associated with the candidate network can be understood as the network device involved in the network architecture associated with the adjacent network currently accessing the network, or as the network device that provides the candidate network, that is, the device that provides network services for the candidate network, such as Provide access network equipment of candidate networks (base station equipment of 4G, 5G and future networks, or ePDG, N3IWF and some access network equipment of future networks), core network equipment (MME, Serving-GW, P-GW in 4G etc., or AMF, SMF, UPF, etc. in 5G, and core network equipment in future networks), similarly, the network equipment associated with the current access network can be understood as the network equipment involved in the network architecture of the current access network , can also be understood as the network equipment that provides the current access network, that is, the equipment that provides network services for the currently accessed network, such as the access network equipment that provides the current access (4G, 5G and future network base station equipment, or ePDG , N3IWF and some access network equipment in future networks), core network equipment (MME, Serving-GW, P-GW, etc. in 4G, or AMF, SMF, UPF, etc. in 5G, and core network equipment in future networks) ). For example, if the terminal currently accesses network A, the corresponding access network device is A1, the core network device is A2, the candidate networks are B and C, the corresponding access network devices are B1 and C1, and the core network device is For B2, C2. The network device associated with the currently accessed network is A1 or A2, and the network devices associated with the candidate network are B1, B2, C1, and C2. In this embodiment of the present invention, the terminal device may receive the network utility values of the network calculated by the respective network devices of the multiple candidate networks, and then the terminal device may select the target access from the multiple candidate networks based on the network utility values. The network reduces the data processing capability requirements of the terminal equipment, and the selected target network can ensure the stability of the communication service, thereby reducing the frequent switching of the network by the terminal equipment, improving the communication performance index, and realizing the personalized service experience of the user.
可选的，网络设备获取终端设备的第一历史记录信息；所述第一历史记录信息包括终端设备的应用的访问记录、内容请求偏好、业务开始时间、业务流量、业务特征和使用时长中的一种或多种；网络设备根据所述第一历史记录信息进行预测得到所述预测业务属性信息，进一步网络设备基于该预测业务属性信息和当前业务属性信息确定业务综合权重。Optionally, the network device obtains the first history record information of the terminal device; the first history record information includes the access record of the application of the terminal device, the content request preference, the service start time, the service flow, the service feature, and the usage duration. One or more: the network device performs prediction according to the first historical record information to obtain the predicted service attribute information, and further, the network device determines the comprehensive service weight based on the predicted service attribute information and the current service attribute information.
步骤S302：终端设备分别确定所述终端设备的M个候选网络的网络效用值。Step S302: The terminal device respectively determines the network utility values of the M candidate networks of the terminal device.
每个所述网络效用值包括对应网络的当前网络效用值和预测网络效用值；所述当前网络效用值为基于当前网络属性信息得到的值，所述预测网络效用值为基于预测网络属性信息得到的值。具体地，当终端设备处于连接网络状态时，M个候选网络可以理解为终端设备当前接入的网络以及其相邻的网络，例如，4G网络、5G网络、蜂窝网络、WiFi网络等；当终端设备处于未连接网络状态时，M个候选网络可以理解为终端设备所处区域能够连接的网络；候选网络的网络效用值可以理解为在该网络下用户的业务体验满意度，例如卡顿、流畅等；候选网络的当前网络效用值可以理解为在该网络下用户对当前业务体验的满意度；候选网络的预测网络效用值可以理解为在该网络下用户对预测业务体验满意度。Each of the network utility values includes a current network utility value and a predicted network utility value of the corresponding network; the current network utility value is a value obtained based on the current network attribute information, and the predicted network utility value is obtained based on the predicted network attribute information value of . Specifically, when the terminal device is in a network connection state, the M candidate networks can be understood as the network currently accessed by the terminal device and its adjacent networks, such as 4G network, 5G network, cellular network, WiFi network, etc.; When the device is not connected to the network, the M candidate networks can be understood as the networks that can be connected in the area where the terminal device is located; the network utility value of the candidate network can be understood as the user's service experience satisfaction under the network, such as stuck, smooth etc.; the current network utility value of the candidate network can be understood as the user's satisfaction with the current service experience under this network; the predicted network utility value of the candidate network can be understood as the user's satisfaction with the predicted service experience under this network.
在一种可能的实现方式中，应用于网络设备，所述网络设备提供S个候选网络；所述S个候选网络为所述终端设备的M个候选网络中的网络；所述方法包括：网络设备获取所述S个候选网络中的每个候选网络的第二历史记录信息；所述第二历史记录信息包括每个小区的历史带宽、时延、抖动、丢包率中的一种或多种；根据每个所述第二历史记录信息进行预测得到S个预测网络属性信息。可以理解的是，每个小区的历史带宽、时延、抖动、丢包率去预测网络属性信息时，可能会有一些限制条件，比如是相同时间段的，特定的位置区域等的。在本发明实施中，在终端设备确定候选网络的网络效用值时，需要考虑候选网络的预测网络属性信息(如网络未来的带宽、时延、抖动、丢包率等)，该预测网络属性信息可由候选网络的网络设备基于该网络之间的历史信息(如每个小区的历史带宽、时延、抖动、丢包率)通过预设算法(如神经网络算法等)得到，实现终端设备在选网时基于网络设备提供的候选网络的预测网络属性信息选出的目标网络，可以保证通信服务的稳定性，从而减少终端设备频繁切换网络，提升了通信性能指标，实现用户个性化业务体验。In a possible implementation manner, it is applied to a network device, and the network device provides S candidate networks; the S candidate networks are networks among M candidate networks of the terminal device; the method includes: a network The device obtains the second history record information of each candidate network in the S candidate networks; the second history record information includes one or more of the historical bandwidth, delay, jitter, and packet loss rate of each cell Prediction is performed according to each of the second historical record information to obtain S pieces of predicted network attribute information. It is understandable that there may be some restrictions when predicting the network attribute information based on the historical bandwidth, delay, jitter, and packet loss rate of each cell, such as the same time period, a specific location area, and so on. In the implementation of the present invention, when the terminal device determines the network utility value of the candidate network, it needs to consider the predicted network attribute information of the candidate network (such as the future network bandwidth, delay, jitter, packet loss rate, etc.), the predicted network attribute information It can be obtained by the network equipment of the candidate network based on the historical information between the networks (such as the historical bandwidth, delay, jitter, and packet loss rate of each cell) through a preset algorithm (such as neural network algorithm, etc.) The target network selected by NetTime based on the predicted network attribute information of the candidate network provided by the network device can ensure the stability of the communication service, thereby reducing the frequent network switching of the terminal device, improving the communication performance index, and realizing the personalized service experience of the user.
在一种可能的实现方式中，所述方法还包括：终端设备测量N个候选网络的参考信号接收功率或参考信号接收质量；判断所述参考信号接收功率或参考信号接收质量是否满足预评估条件；将满足所述预评估条件的M个候选网络确定为所述M个候选网络。In a possible implementation manner, the method further includes: the terminal device measures the reference signal received power or the reference signal received quality of the N candidate networks; and judges whether the reference signal received power or the reference signal received quality meets a pre-evaluation condition ; Determine the M candidate networks that satisfy the pre-evaluation conditions as the M candidate networks.
具体地，N个候选网络可以理解为能够为终端设备UE当前接入的小区以及相邻小区中 的至少一个小区，例如，UE当前接入A小区，B、C、D为A小区的相邻小区，则N个候选的网络可以理解为A、B、C、D四个小区中的至少一个小区，或者，也可以将N个候选的网络理解为UE当前接入的网络制式以及可以接入到的其他网络制式中的至少一个网络或者至少一个网络下的一个小区，例如，UE当前接入4G网络且可以接入到5G网络、WiFi网络等，则N个候选的网络可以理解为4G、5G、WiFi三种网络中的至少一个网络或者至少一个网络里的一个小区；参考信号接收功率参考和信号接收质量可以理解为代表无线信号强度和无线信号质量的关键参数；预评估条件可以理解为参考信号接收功率参考和/或信号接收质量满足一定的门限数值，通常为基站配置给终端的用于空闲态执行小区重选的参数以及连接态用于发起小区切换的事件。比如在空闲态的重选条件(邻小区的信号强度测量值较服务小区高于一定门限数值并持续一定时长，比如5G内，邻小区测量值为65dbm，服务小区的测量值为70dbm，并且保持了5s，当门限数值为3dB，时长为3s时，预评估条件满足)在连接态，基站配置给UE的A3事件(相同无线接入技术中，邻小区的信号强度测量值较服务小区高于一定门限数值并持续一定时长，比如5G内，邻小区测量值为65dbm，服务小区的测量值为70dbm，并且保持了5s，当门限数值为3dB，时长为3s则A3事件满足)或B2事件(服务小区的信号强度测量值低于第一门限，并且邻小区的信号强度测量值高于第二门限数值，其中服务小区和相邻小区属于不同无线接入技术的小区)；M个候选网络可以理解为当前接入的网络以及其相邻网络中满足预评估条件的网络。Specifically, the N candidate networks can be understood as at least one cell among the cell currently accessed by the terminal equipment UE and the adjacent cells, for example, the UE currently accesses the cell A, and B, C, and D are the neighbors of the cell A. cell, the N candidate networks can be understood as at least one cell among the four cells A, B, C, and D, or the N candidate networks can also be understood as the network standard currently accessed by the UE and the network that can be accessed. At least one network in other network standards or a cell under at least one network, for example, if the UE currently accesses a 4G network and can access a 5G network, a WiFi network, etc., the N candidate networks can be understood as 4G, At least one network or at least one cell in one of the three networks of 5G and WiFi; the reference signal reception power reference and signal reception quality can be understood as the key parameters representing the wireless signal strength and wireless signal quality; the pre-evaluation conditions can be understood as The reference signal received power reference and/or the received signal quality meet a certain threshold value, which is usually a parameter configured by the base station to the terminal for performing cell reselection in an idle state and an event for initiating a cell handover in a connected state. For example, in the re-selection condition in the idle state (the signal strength measurement value of the neighbor cell is higher than a certain threshold value and lasts for a certain period of time, for example, in 5G, the measurement value of the neighbor cell is 65dbm, and the measurement value of the serving cell is 70dbm, and keep When the threshold value is 3dB and the duration is 3s, the pre-evaluation condition is satisfied) In the connected state, the base station configures the A3 event for the UE (in the same wireless access technology, the signal strength measurement value of the neighboring cell is higher than that of the serving cell) A certain threshold value and lasts for a certain period of time. For example, in 5G, the measurement value of the neighboring cell is 65dbm, the measurement value of the serving cell is 70dbm, and it is maintained for 5s. When the threshold value is 3dB and the duration is 3s, the A3 event is satisfied) or the B2 event ( The signal strength measurement value of the serving cell is lower than the first threshold, and the signal strength measurement value of the neighboring cell is higher than the second threshold value, wherein the serving cell and the neighboring cell belong to different radio access technology cells); M candidate networks can be It is understood as the network currently accessed and the network that meets the pre-evaluation conditions in its neighboring networks.
需要说明的是，在上述步骤S302中所涉及的确定所述终端设备的M个候选网络的网络效用值的实现可以包括如下两种实施方式：It should be noted that the implementation of determining the network utility values of the M candidate networks of the terminal device involved in the foregoing step S302 may include the following two implementations:
实施方式一：由终端设备生成网络效用值。Embodiment 1: The network utility value is generated by the terminal device.
在一种可能的实现方式中，所述终端设备分别确定M个候选网络的网络效用值，包括：终端设备根据所述M个候选网络对应的所述当前网络属性信息确定每个所述候选网络的所述当前网络效用值，以及根据所述M个候选网络对应的所述预测网络属性信息确定每个所述候选网络的所述预测网络效用值，得到所述M个候选网络的所述网络效用值；所述当前网络属性信息包括网络当前的带宽、时延、抖动、丢包率中的一种或多种；所述预测网络属性信息包括网络未来的带宽、时延、抖动、丢包率中的一种或多种。具体地，候选网络的当前网络属性信息可以理解为网络当前的带宽、时延、抖动、丢包率等信息；候选网络的当前网络效用值可以理解为在该网络下用户对当前业务体验的满意度；候选网络的预测网络效用值可以理解为在该网络下用户对预测业务体验满意度。在本发明实施例中，终端设备可基于多个候选网络各自的当前网络属性信息(如网络当前的带宽、时延、抖动、丢包率等)确定当前网络效用值，以及预测网络属性信息(如网络未来的带宽、时延、抖动、丢包率等)确定预测网络效用值，从而得到多个候选网络的网络效用值。当终端设备需要进行智能选网时，就可在终端设备上得到多个候选网络的网络效用值，提高了终端设备智能选网的效率，且选出的目标网络可以保证通信服务的稳定性，从而减少终端设备频繁切换网络，提升了通信性能指标，实现用户个性化业务体验。In a possible implementation manner, the terminal device determining the network utility values of the M candidate networks respectively includes: the terminal device determining each candidate network according to the current network attribute information corresponding to the M candidate networks and the predicted network utility value of each candidate network is determined according to the predicted network attribute information corresponding to the M candidate networks, and the network of the M candidate networks is obtained. Utility value; the current network attribute information includes one or more of the current network bandwidth, delay, jitter, and packet loss rate; the predicted network attribute information includes future network bandwidth, delay, jitter, packet loss one or more of the rates. Specifically, the current network attribute information of the candidate network can be understood as the current network bandwidth, delay, jitter, packet loss rate and other information; the current network utility value of the candidate network can be understood as the user's satisfaction with the current service experience under the network degree; the predicted network utility value of the candidate network can be understood as the user's satisfaction with the predicted service experience under this network. In this embodiment of the present invention, the terminal device may determine the current network utility value based on the respective current network attribute information (such as the current network bandwidth, delay, jitter, packet loss rate, etc.) of the multiple candidate networks, and predict the network attribute information ( Such as the future network bandwidth, delay, jitter, packet loss rate, etc.) to determine the predicted network utility value, so as to obtain the network utility value of multiple candidate networks. When the terminal device needs to perform intelligent network selection, the network utility values of multiple candidate networks can be obtained on the terminal device, which improves the efficiency of the terminal device's intelligent network selection, and the selected target network can ensure the stability of communication services. Thus, frequent network switching of terminal devices is reduced, communication performance indicators are improved, and user personalized service experience is realized.
可选的，终端设备通过指定信令向网络设备(如基站)请求网络决策属性信息，网络下发满足切换预评估条件的邻区的网络决策属性信息，其中包括网络环境中各邻区当前和所预测的网络决策属性信息。具体情况可见下表举例：Optionally, the terminal device requests the network device (such as the base station) for network decision attribute information through designated signaling, and the network delivers the network decision attribute information of the neighboring cells that meet the handover pre-evaluation conditions, including the current and Predicted network decision attribute information. The specific situation can be seen in the table below for example:
表二获取网络决策属性信息情况表Table 2 Obtaining the information of network decision attributes
Figure PCTCN2022083216-appb-000001
Figure PCTCN2022083216-appb-000001
Figure PCTCN2022083216-appb-000002
Figure PCTCN2022083216-appb-000002
Figure PCTCN2022083216-appb-000003
Figure PCTCN2022083216-appb-000003
在一种可能的实现方式中，所述当前业务属性信息包括当前业务类型、当前业务持续时间和用户偏好；所述终端设备基于所述当前业务权重和每个所述候选网络的所述当前网络效用值，以及基于所述预测业务权重和每个所述候选网络的所述预测网络效用值，包括：终端设备根据所述当前业务属性信息得到所述当前业务的服务质量需求；所述服务质量需求包括当前所述终端设备所需要的带宽、时延、抖动和丢包率中的一种或多种；基于所述服务质量和所述用户偏好得到网络效用函数；根据每个所述候选网络的所述当前网络属性信息和所述网络效用函数得到所述当前网络效用值；根据每个所述候选网络的所述预测网络属性信息和所述网络效用函数得到所述预测网络效用值。具体地，服务质量需求可以理解为终端设备运行当前业务所需要的带宽、时延、抖动和丢包率等信息；用户偏好可以理解为用户对业务的通信服务偏好，例如，用户在进行语音电话、游戏、看直播等不同业务时，对网络性能的需求是不同的，此外，不同的用户对同一种业务的通信服务偏好也是不同，有的用户偏好极致的业务体验，可以忽略网络流量的费用，追求大带宽高速率的业务体验，而有的用户则偏好低价的可用服务，对带宽速率的追求不大；网络效用函数可以理解为表示网络服务与用户满意度之间的效用函数。在本发明实施例中，基于终端设备的当前用户业务的服务质量需求以及用户偏好构建网络效用函数，然后根据该网络效用函数得到网络效用值，终端设备基于该网络效用值进行个性化选网，使得选出的目标网络能够更加准确的反映用户业务意图，实现用户个性化业务体验，且可以保证通信服务的稳定性，减少终端设备频繁切换网络，提升了通信性能指标。In a possible implementation manner, the current service attribute information includes current service type, current service duration and user preference; the terminal device is based on the current service weight and the current network of each candidate network utility value, and the predicted network utility value based on the predicted service weight and each candidate network, including: the terminal device obtains the service quality requirement of the current service according to the current service attribute information; the service quality The requirements include one or more of the bandwidth, delay, jitter and packet loss rate currently required by the terminal device; a network utility function is obtained based on the quality of service and the user preference; according to each candidate network The current network utility value is obtained from the current network attribute information and the network utility function; the predicted network utility value is obtained according to the predicted network attribute information and the network utility function of each candidate network. Specifically, service quality requirements can be understood as information such as bandwidth, delay, jitter, and packet loss rate required by the terminal device to run the current service; user preferences can be understood as the user's communication service preferences for services, for example, when a user is making a voice call Different services, such as games, live streaming, etc., have different requirements for network performance. In addition, different users have different communication service preferences for the same service. Some users prefer the ultimate service experience, and the cost of network traffic can be ignored. , the pursuit of high-bandwidth and high-speed service experience, while some users prefer low-cost available services, and have little pursuit of bandwidth speed; network utility function can be understood as a utility function between network services and user satisfaction. In the embodiment of the present invention, a network utility function is constructed based on the service quality requirements of the current user service of the terminal equipment and user preferences, and then a network utility value is obtained according to the network utility function, and the terminal equipment performs personalized network selection based on the network utility value, The selected target network can more accurately reflect the user's service intention, realize the user's personalized service experience, ensure the stability of the communication service, reduce the frequent network switching of terminal equipment, and improve the communication performance index.
在一种可能的实现方式中，所述终端设备基于所述服务质量和所述用户偏好得到网络效用函数，包括：终端设备基于所述服务质量需求得到L个网络决策属性效用函数；基于所述服务质量和所述用户偏好得到所述L个网络决策属性效用函数中的每个所述网络决策属性效用函数对应的网络决策属性权重；根据所述L个网络决策属性效用函数及其对应的所述网络决策属性权重，确定所述网络效用函数。具体地，终端设备的当前业务的服务质量需求中包括了L个网络决策属性(如带宽、时延等)，不同的网络决策属性由相应的网络决策属性效用函数来表征，再结合用户偏好信息确定不同网络决策属性效用函数的权重，进而得到网络效用函数。通过该网络效用函数可反映不同网络决策属性与用户满意度的关系，从而基于该网络效用函数选出的目标网络能够更加准确的反映用户业务意图，实现用户个性化业务体验，且可以保证通信服务的稳定性，减少终端设备频繁切换网络，提升了通信性能指标。In a possible implementation manner, the terminal device obtains a network utility function based on the service quality and the user preference, including: the terminal device obtains L network decision attribute utility functions based on the service quality requirement; The quality of service and the user preference obtain the network decision attribute weight corresponding to each of the L network decision attribute utility functions; according to the L network decision attribute utility functions and their corresponding The network decision attribute weight is used to determine the network utility function. Specifically, the service quality requirements of the current service of the terminal device include L network decision attributes (such as bandwidth, delay, etc.), and different network decision attributes are represented by the corresponding network decision attribute utility functions, and then combined with user preference information. Determine the weight of the utility function of different network decision attributes, and then obtain the network utility function. The network utility function can reflect the relationship between different network decision attributes and user satisfaction, so that the target network selected based on the network utility function can more accurately reflect the user's business intention, realize the user's personalized business experience, and ensure communication services. The stability of the network reduces the frequent switching of the network by the terminal equipment, and improves the communication performance index.
本申请实施例中示例性给出一种确定网络效用函数的具体方法，但本申请实施例中的确定网络效用函数的具体方法包括但不仅限于以下描述。A specific method for determining a network utility function is exemplarily given in the embodiment of the present application, but the specific method for determining the network utility function in the embodiment of the present application includes but is not limited to the following description.
(1)确定L个网络决策属性效用函数：(1) Determine L network decision attribute utility functions:
不同的网络决策属性，需要相应的效用函数来表征，以便反映不同网络决策属性与用户满意度的关系。例如，用户对可用带宽的要求存在一个区间[x min,x max]，在该区间内，用户满意度随着带宽决策属性值的增加而提升，若可用带宽低于x min，表示为用户不满意服务，因为已经不能满足当前业务的基本要求。另一方面，若可用带宽高于x max，用户的满意度也不会得到太明显的提升，因为用户已经得到最高业务需求的满足，更多的带宽已经不能带来用户体验的提升。S型函数符合上述特征，因此，可以将S型函数作为可用带宽网络决策属性的效用函数。一个基本的S型函数表示形式如下： Different network decision attributes need to be represented by corresponding utility functions, so as to reflect the relationship between different network decision attributes and user satisfaction. For example, there is an interval [x min , x max ] in the user's requirement for the available bandwidth. In this interval, the user satisfaction increases with the increase of the bandwidth decision attribute value. If the available bandwidth is lower than x min , it means that the user does not Satisfied with the service, because it can no longer meet the basic requirements of the current business. On the other hand, if the available bandwidth is higher than x max , the user's satisfaction will not be significantly improved, because the user has been satisfied with the highest service requirements, and more bandwidth can no longer bring about an improvement in the user experience. The sigmoid function conforms to the above characteristics, therefore, the sigmoid function can be used as the utility function of the available bandwidth network decision properties. A basic sigmoid function representation is as follows:
Figure PCTCN2022083216-appb-000004
Figure PCTCN2022083216-appb-000004
其中，λ表示函数值随变量改变的敏感度，x m表示变量的期望值。根据带宽网络决策属性的性质，带宽网络决策属性的效用函数u(x 1)可表示为： where λ represents the sensitivity of the function value to changes in the variable, and x m represents the expected value of the variable. According to the nature of the decision attribute of the bandwidth network, the utility function u(x 1 ) of the decision attribute of the bandwidth network can be expressed as:
Figure PCTCN2022083216-appb-000005
Figure PCTCN2022083216-appb-000005
其中，x m表示与用户业务需求在当前通信环境下的等效频谱带宽，即用户的期望带宽，x 1表示当前网络的可用带宽，BW表示网络带宽上限。α表示用户对带宽属性的敏感程度，敏感程度越高，α的值越大。 Among them, x m represents the equivalent spectral bandwidth of the user's service requirements in the current communication environment, that is, the user's desired bandwidth, x 1 represents the available bandwidth of the current network, and BW represents the upper limit of the network bandwidth. α represents the user's sensitivity to the bandwidth attribute, and the higher the sensitivity, the larger the value of α.
又例如，对于时延等属性，用户的要求是数值越低越好，期望接入时延低的网络，随着时延的增加，用户的满意度越低，体验下降。所以时延网络决策属性的效用函数u(x 2)可以表示为： For another example, for attributes such as delay, the user's requirement is that the lower the value, the better, and the user expects to access a network with low delay. As the delay increases, the user's satisfaction is lower and the experience is degraded. So the utility function u(x 2 ) of the decision attribute of the delay network can be expressed as:
Figure PCTCN2022083216-appb-000006
Figure PCTCN2022083216-appb-000006
其中，x max表示无线网络中时延决策属性的上限值(表示超时)，x 2表示当前网络时延决策属性值，β表示用户业务对时延属性的敏感程度，敏感程度越高，β的值越大。 Among them, x max represents the upper limit of the delay decision attribute in the wireless network (indicating timeout), x 2 represents the current network delay decision attribute value, and β represents the sensitivity of the user service to the delay attribute. The higher the sensitivity, the β The larger the value is.
(2)确定网络决策属性权重：(2) Determine the weight of network decision attributes:
根据业务类型，确定终端设备的服务质量需求。如针对在线游戏业务类型，服务质量需求为[Bandwidth＝10Mbps,Latency＝30ms,Jitter＝20ms,Loss Tolerance＝0.0005]。终端设备的用户偏好为[4,2,1,3]，某个服务质量需求对应的用户偏好值越大，说明用户越重视。例如，在线游戏业务中，对该用户设备而言，可用带宽是业务质量的根本保证，而丢包率是导致图像质量恶化的根本原因，重要程度次之，而时延和抖动对用户来说不重要，因为该用户设备上运行在线游戏业务的软件有通过适当的缓存来减少时延和抖动对业务体验的影响的能力。因此，在各网络决策属性的权重分配中，可用带宽的权重最大，丢包率此次，时延和抖动最小。各网络决策属性的权重可采用如下公式确定：According to the service type, determine the service quality requirements of the terminal equipment. For example, for the online game business type, the service quality requirement is [Bandwidth=10Mbps, Latency=30ms, Jitter=20ms, Loss Tolerance=0.0005]. The user preference of the terminal device is [4, 2, 1, 3]. The larger the user preference value corresponding to a certain service quality requirement, the more attention the user pays. For example, in the online game business, for the user equipment, the available bandwidth is the fundamental guarantee of the service quality, and the packet loss rate is the root cause of the deterioration of the image quality. It is not important, because the software running the online game service on the user equipment has the ability to reduce the impact of delay and jitter on the service experience through appropriate caching. Therefore, in the weight distribution of each network decision attribute, the weight of the available bandwidth is the largest, the packet loss rate this time, and the delay and jitter are the smallest. The weight of each network decision attribute can be determined by the following formula:
Figure PCTCN2022083216-appb-000007
Figure PCTCN2022083216-appb-000007
其中，p i和ω i分别为第i个网络决策属性的用户偏好和权重，N为网络决策属性的数目，由此可获得在线游戏业务的权重分配为w1＝[ω 1,ω 2,ω 3,ω 4]＝[0.4,0.2,0.1,0.3]。 Among them, p i and ω i are the user preference and weight of the i-th network decision-making attribute, respectively, and N is the number of network decision-making attributes, so the weight distribution of the online game business can be obtained as w1=[ω 1 ,ω 2 ,ω 3 , ω 4 ]=[0.4, 0.2, 0.1, 0.3].
(3)确定网络效用函数：(3) Determine the network utility function:
根据网络决策属性的权重分配，以及相应的效用函数，将多个单元效用函数按照权重进 行乘法或加法，或乘法与加法组合，得到多元效用函数，即网络效用函数。例如，乘法组合的多元效用函数表达式为：According to the weight distribution of the network decision attributes and the corresponding utility function, the multi-unit utility function is multiplied or added according to the weight, or the multiplication and addition are combined to obtain the multi-functional utility function, that is, the network utility function. For example, the multiplicative utility function expression for the multiplicative combination is:
Figure PCTCN2022083216-appb-000008
Figure PCTCN2022083216-appb-000008
其中，u i(x)和w i分别表示第i个网络决策属性的效用函数和权重。 Among them, ui (x) and wi represent the utility function and weight of the i-th network decision attribute, respectively.
实施方式二：由网络设备生成网络效用值。Embodiment 2: The network utility value is generated by the network device.
在一种可能的实现方式中，所述网络设备根据每个所候选网络对应的所述当前网络属性信息确定所述当前网络效用值，以及根据每个所述候选网络对应的所述预测网络属性信息确定所述预测网络效用值；所述当前网络属性信息包括网络当前的带宽、时延、抖动、丢包率中的一种或多种；所述预测网络属性信息包括网络未来的带宽、时延、抖动、丢包率中的一种或多种；基于每个所述当前网络效用值和每个所述预测网络效用值得到S个网络效用值；所述网络效用值包括同一个所述候选网络的当前网络效应值和预测网络效用值；向所述终端设备发送所述S个候选网络对应的所述S个网络效用值。具体地，候选网络的当前网络属性信息可以理解为网络当前的带宽、时延、抖动、丢包率等信息；候选网络的当前网络效用值可以理解为在该网络下用户对当前业务体验的满意度；候选网络的预测网络效用值可以理解为在该网络下用户对预测业务体验满意度。在本发明实施例中，网络设备可基于S个候选网络各自的当前网络属性信息(如网络当前的带宽、时延、抖动、丢包率等)得到当前网络效用值，以及预测网络属性信息(如网络未来的带宽、时延、抖动、丢包率等)确定预测网络效用值，从而得到S个候选网络的网络效用值，从而终端设备可基于网络效用值选出目标网络，提升了通信性能指标，实现用户个性化业务体验。In a possible implementation manner, the network device determines the current network utility value according to the current network attribute information corresponding to each candidate network, and determines the current network utility value according to the predicted network attribute corresponding to each candidate network information to determine the predicted network utility value; the current network attribute information includes one or more of the current network bandwidth, delay, jitter, and packet loss rate; the predicted network attribute information includes future network bandwidth, time one or more of delay, jitter, and packet loss rate; S network utility values are obtained based on each of the current network utility values and each of the predicted network utility values; the network utility values include the same The current network effect value and the predicted network utility value of the candidate network; the S network utility values corresponding to the S candidate networks are sent to the terminal device. Specifically, the current network attribute information of the candidate network can be understood as the current network bandwidth, delay, jitter, packet loss rate and other information; the current network utility value of the candidate network can be understood as the user's satisfaction with the current service experience under the network degree; the predicted network utility value of the candidate network can be understood as the user's satisfaction with the predicted service experience under this network. In this embodiment of the present invention, the network device may obtain the current network utility value based on the current network attribute information (such as the current network bandwidth, delay, jitter, packet loss rate, etc.) of the S candidate networks, and predict the network attribute information ( Such as the future network bandwidth, delay, jitter, packet loss rate, etc.) to determine the predicted network utility value, so as to obtain the network utility value of the S candidate networks, so that the terminal device can select the target network based on the network utility value, which improves the communication performance. Metrics to realize user personalized service experience.
可选的，网络设备获取终端设备的当前业务属性信息，然后网络设备根据终端设备的当前业务属性信息得到所述当前业务的服务质量需求；服务质量需求可包括当前终端设备所需要的带宽、时延、抖动和丢包率中的一种或多种；网络设备基于该服务质量和用户偏好得到网络效用函数；将候选网络的当前网络属性信息输入所述网络效用函数，得到所述当前网络效用值；将每个候选网络的所述预测网络属性信息输入所述网络效用函数，得到所述预测网络效用值。需要说明的是，当前业务属性信息可包括当前业务类型、当前业务持续时间和用户偏好。Optionally, the network device obtains the current service attribute information of the terminal device, and then the network device obtains the service quality requirement of the current service according to the current service attribute information of the terminal device; one or more of delay, jitter and packet loss rate; network equipment obtains a network utility function based on the service quality and user preference; input the current network attribute information of the candidate network into the network utility function to obtain the current network utility value; input the predicted network attribute information of each candidate network into the network utility function to obtain the predicted network utility value. It should be noted that the current service attribute information may include current service type, current service duration and user preference.
可选的，所述网络设备基于所述服务质量和所述用户偏好得到网络效用函数，包括：网络设备基于所述服务质量需求得到L个网络决策属性效用函数；网络设备基于所述服务质量和所述用户偏好得到所述L个网络决策属性效用函数中的每个所述网络决策属性效用函数对应的网络决策属性权重；根据所述L个网络决策属性效用函数及其对应的所述网络决策属性权重，确定所述网络效用函数。Optionally, the network device obtains a network utility function based on the service quality and the user preference, including: the network device obtains L network decision attribute utility functions based on the service quality requirement; the network device obtains L network decision attribute utility functions based on the service quality and the user preference; The user preference obtains the network decision attribute weight corresponding to each of the network decision attribute utility functions in the L network decision attribute utility functions; according to the L network decision attribute utility functions and the corresponding network decision Attribute weights, which determine the network utility function.
在一种可能的实现方式中，所述终端设备分别确定M个候选网络的网络效用值，包括：终端设备接收所述提供M个候选网络关联的一个或多个网络设备分别发送的所述M个候选网络的所述网络效用值。在本发明实施例中，终端设备可接收由多个候选网络各自的网络设备所计算的本网络的网络效用值，然后终端设备可基于这些网络效用值从多个候选网络中选择出目标接入网络，降低了终端设备对数据处理能力的要求，且选出的目标网络可以保证通信服务的稳定性，从而减少终端设备频繁切换网络，提升了通信性能指标，实现用户个性化业务体验。In a possible implementation manner, the terminal device determining the network utility values of the M candidate networks respectively includes: the terminal device receiving the M messages respectively sent by the one or more network devices that provide associations of the M candidate networks. the network utility values for each candidate network. In this embodiment of the present invention, the terminal device may receive the network utility values of the network calculated by the respective network devices of the multiple candidate networks, and then the terminal device may select the target access from the multiple candidate networks based on the network utility values. The network reduces the data processing capability requirements of the terminal equipment, and the selected target network can ensure the stability of the communication service, thereby reducing the frequent switching of the network by the terminal equipment, improving the communication performance index, and realizing the personalized service experience of the user.
需要说明的是，本申请对于上述步骤S301和步骤S302的执行先后顺序不作具体限定，可以先执行步骤S301再执行步骤S302，也可以先执行步骤S302再执行步骤S301，还可以同时执行步骤S301和步骤S302。It should be noted that this application does not specifically limit the order of execution of the above steps S301 and S302. Step S301 may be executed first and then step S302 may be executed, or step S302 may be executed first and then step S301 may be executed, or steps S301 and S301 may be executed simultaneously. Step S302.
步骤S303：终端设备基于所述业务综合权重和所述M个候选网络的所述网络效用值，得到M个综合效用值。Step S303: The terminal device obtains M comprehensive utility values based on the service comprehensive weight and the network utility values of the M candidate networks.
在一种可能的实现方式中，所述终端设备基于所述业务综合权重和所述M个候选网络的所述网络效用值，得到M个综合效用值，包括：终端设备基于所述当前业务权重和每个所述候选网络的所述当前网络效用值，以及基于所述预测业务权重和每个所述候选网络的所述预测网络效用值，计算每个所述网络的综合效用值。在本发明实施例中，将不同候选网络的当前网络效用值和预测网络效用值分别与对应的当前业务权重和预测业务权重进行加权，得到多个网络的综合效用值，该综合效用值的确定是基于了终端设备的当前业务状态和未来一段时间的业务状态，以及网络的当前状态和未来一段时间的状态，因此通过实施本发明实施例提供的方法，终端设备基于多个候选网络的综合效用值从多个候选网络中选出的目标网络，可以保证通信服务的稳定性，从而减少终端设备频繁切换网络，提升了通信性能指标，实现用户个性化业务体验。In a possible implementation manner, the terminal device obtains M comprehensive utility values based on the service comprehensive weight and the network utility values of the M candidate networks, including: the terminal device is based on the current service weight and the current network utility value for each of the candidate networks, and a composite utility value for each of the candidate networks is calculated based on the predicted traffic weight and the predicted network utility value for each of the candidate networks. In the embodiment of the present invention, the current network utility value and the predicted network utility value of different candidate networks are respectively weighted with the corresponding current service weight and predicted service weight to obtain the comprehensive utility value of multiple networks. It is based on the current service state of the terminal device and the service state in the future, as well as the current state of the network and the state in the future. Therefore, by implementing the method provided by the embodiment of the present invention, the terminal device is based on the comprehensive utility of multiple candidate networks. The value of the target network selected from multiple candidate networks can ensure the stability of communication services, thereby reducing the frequent switching of networks by terminal equipment, improving communication performance indicators, and realizing user personalized service experience.
例如，根据业务综合权重，对各小区当前和未来的效用值进行综合加权计算，即综合考虑各小区当前与未来的效用值，选择综合效用值最大的作为目标小区。综合效用值计算表达式为：For example, the current and future utility values of each cell are comprehensively weighted according to the service comprehensive weight, that is, the current and future utility values of each cell are comprehensively considered, and the target cell with the largest comprehensive utility value is selected. The comprehensive utility value calculation expression is:
Figure PCTCN2022083216-appb-000009
Figure PCTCN2022083216-appb-000009
其中，w cur和
Figure PCTCN2022083216-appb-000010
分别为当前的业务权重和当前的网络决策属性效用值，w fut和
Figure PCTCN2022083216-appb-000011
分别为将来的业务权重和将来的网络决策属性效用值。 where w cur and
Figure PCTCN2022083216-appb-000010
are the current business weight and the current network decision attribute utility value, w fut and
Figure PCTCN2022083216-appb-000011
are the future business weight and the future network decision attribute utility value, respectively.
步骤S304：终端设备基于所述M个综合效用值，从所述M个候选网络中选取所述终端设备的目标网络。Step S304: The terminal device selects a target network of the terminal device from the M candidate networks based on the M comprehensive utility values.
具体地，通过上述步骤可以分别获得M个候选网络的综合效用值，然后可将综合效用值分数最高的候选网络作为目标网络。进一步，终端设备可上报目标小区，进入切换执行过程。例如，假设有4个小区{0,1,2,3}，根据其当前和将来的网络决策属性，可以获得各小区当前和将来的效用值。如：Specifically, through the above steps, the comprehensive utility values of the M candidate networks can be obtained respectively, and then the candidate network with the highest comprehensive utility value score can be used as the target network. Further, the terminal equipment can report the target cell and enter the handover execution process. For example, assuming that there are 4 cells {0, 1, 2, 3}, according to their current and future network decision attributes, the current and future utility values of each cell can be obtained. like:
Figure PCTCN2022083216-appb-000012
Figure PCTCN2022083216-appb-000012
然后基于业务综合权重w2＝[0.55,0.45]，分别得出各小区的综合效用值为{0.7625,0.6455,0.8625,0.8115}，因此选择2号小区为目标小区。Then, based on the comprehensive service weight w2=[0.55, 0.45], the comprehensive utility value of each cell is obtained as {0.7625, 0.6455, 0.8625, 0.8115}, so cell No. 2 is selected as the target cell.
可选的，终端设备在执行完上述步骤S301至步骤S304之后，终端设备还可执行目标小区选择。例如，当终端设备处于空闲状态(如未接入网络状态)时，终端设备自主选择到目标小区；当终端设备处于连接态时，终端向网络发送测量报告消息，消息中指示UE优选的目标小区，或者优先的目标小区排序等，UE结合UE发送的信息，选择一个目标小区执行向目标小区的切换过程。Optionally, after the terminal device performs the above steps S301 to S304, the terminal device may also perform target cell selection. For example, when the terminal device is in an idle state (such as not accessing the network), the terminal device autonomously selects the target cell; when the terminal device is in a connected state, the terminal sends a measurement report message to the network, which indicates the UE's preferred target cell. , or priority target cell ranking, etc., the UE selects a target cell to perform the handover process to the target cell in combination with the information sent by the UE.
在本发明实施例中，综合考虑终端设备的当前业务状态和未来一段时间内的业务状态， 以及不同网络的当前状态和未来一段时间的状态，确定终端设备的当前业务与未来一段时间的预测业务的综合业务权重，以及不同网络的当前效用值和未来一段时间的预测效用值，从而选择出与用户业务意图最匹配的网络，避免了仅根据用户当前业务和当前网络状态通过多元网络效用函数进行选网，终端设备切换至目标网络后，若终端设备运行的业务或网络状态发生变化，导致目标网络提供的通信服务不稳定终端设备需要频繁切换网络的问题，以及现有技术中未考虑网络的未来状态，无法保证网络的选择是否准确和切换网络是否必要的问题。因此通过实施本发明实施例提供的方法，终端设备从多个候选网络中选出的目标网络，可以保证通信服务的稳定性，从而减少终端设备频繁切换网络，提升了通信性能指标，实现用户个性化业务体验。In the embodiment of the present invention, the current service status of the terminal device and the service status in a future period of time, as well as the current status of different networks and the status of a future period of time, are comprehensively considered to determine the current service of the terminal equipment and the predicted service for a future period of time. The comprehensive service weight of the network, as well as the current utility value of different networks and the predicted utility value for a period of time in the future, so as to select the network that best matches the user's service intention, avoiding the need to use multiple network utility functions only based on the user's current service and current network status. After the terminal device switches to the target network, if the service or network status of the terminal device changes, the communication service provided by the target network is unstable, and the terminal device needs to switch the network frequently, and the prior art does not consider the network problem. In the future, there is no guarantee whether the selection of the network is accurate and whether it is necessary to switch the network. Therefore, by implementing the method provided by the embodiment of the present invention, the target network selected by the terminal device from multiple candidate networks can ensure the stability of the communication service, thereby reducing the frequent switching of the network by the terminal device, improving the communication performance index, and realizing the user's individuality. business experience.
针对上述步骤S301、步骤S302、步骤S303、步骤S304是以终端设备为执行主体进行描述，但本申请实施例中的智能选网方法中还可包括执行主体为其他设备情况，以下示例性举例说明，参见图2B，图2B是本申请实施例中的另一种智能选网方法的流程示意图，下面将结合附图2B并基于上述图1A中的智能选网***架构从终端设备和网络设备的交互侧对本申请实施例中的智能选网方法进行描述。需要说明的是，为了更详细的描述本申请实施例中的智能选网方法，本申请在各个流程步骤中描述了相应的执行主体分别为网络设备，但不代表本申请实施例只能通过所描述的执行主体进行对应的方法流程。The above steps S301, S302, S303, and S304 are described with the terminal device as the execution subject, but the intelligent network selection method in this embodiment of the present application may also include the execution subject as other devices. The following exemplary examples illustrate the situation , referring to Fig. 2B, Fig. 2B is the schematic flow chart of another kind of intelligent network selection method in the embodiment of the present application, below will be combined with accompanying drawing 2B and based on the intelligent network selection system architecture in above-mentioned Fig. 1A from terminal equipment and network equipment The interactive side describes the intelligent network selection method in the embodiment of the present application. It should be noted that, in order to describe the intelligent network selection method in the embodiments of the present application in more detail, the present application describes in each process step that the corresponding execution subjects are network devices, respectively, but this does not mean that the embodiments of the present application can only be implemented through all The described execution body performs the corresponding method flow.
步骤S3001：网络设备确定终端设备的业务综合权重。Step S3001: The network device determines the comprehensive service weight of the terminal device.
所述业务综合权重包括所述终端设备的当前业务权重和预测业务权重；所述当前业务权重为基于所述终端设备当前业务属性信息得到的权重，所述预测业务权重为基于所述终端设备预测业务属性信息得到的权重。具体地，网络设备可以理解为基站，例如基站可以是时分同步码分多址(Time Division Synchronous Code Division Multiple Access，TD-SCDMA)***中的基站收发台(Base Transceiver Station，BTS)，也可以是LTE***中的演进型基站(Evolutional Node B，eNB)，以及5G***、新空口(NR)***中的基站等；终端设备可以理解为智能手机、智能手表、平板电脑等移动设备；业务综合权重可以理解为用于表示当前业务和预测业务分别对网络资源的需求；当前业务权重可以理解为终端设备当前运行的业务对网络资源的需求；预测业务权重可以理解为终端设备未来一段时间可能运行的业务对网络资源的需求；当前业务属性信息可以理解为终端设备当前运行的业务类型和当前业务持续时间等信息；预测业务属性信息可以理解为终端设备未来可能运行的业务类型和预测业务持续时间等信息。例如，某一用户当下正在使用视频软件看视频，预测到该用户未来一个小时内会进行化AR/VR游戏，由于化AR/VR游戏相对视频对网络的服务质量需求更高，因此，将来的业务综合权重需要比现在的业务权重大，比如可确定业务综合权重为[0.3,0.7]，表示相对于未来的网络性能和用户状态，用户对网络当前所提供服务质量需求更低，也可以理解为终端设备当前所需要的网络资源(如能量需求)比未来所需要的网络资源低。The service comprehensive weight includes the current service weight and the predicted service weight of the terminal equipment; the current service weight is a weight obtained based on the current service attribute information of the terminal equipment, and the predicted service weight is based on the terminal equipment prediction. The weight obtained by the business attribute information. Specifically, a network device can be understood as a base station, for example, a base station can be a base transceiver station (Base Transceiver Station, BTS) in a Time Division Synchronous Code Division Multiple Access (TD-SCDMA) system, or a Evolved base station (Evolutional Node B, eNB) in LTE system, as well as base station in 5G system, new air interface (NR) system, etc.; terminal equipment can be understood as mobile devices such as smartphones, smart watches, tablet computers, etc.; business comprehensive weight It can be understood as representing the requirements of the current service and the predicted service for network resources respectively; the current service weight can be understood as the demand for network resources of the service currently running on the terminal device; the predicted service weight can be understood as the terminal equipment may run in the future Service requirements for network resources; current service attribute information can be understood as information such as the current service type and current service duration of the terminal equipment; predicted service attribute information can be understood as the terminal equipment may run in the future. Service type and predicted service duration, etc. information. For example, a user is currently using video software to watch videos, and it is predicted that the user will play AR/VR games in the next hour. Since AR/VR games have higher requirements for network service quality than videos, future The comprehensive weight of the business needs to be larger than the current business weight. For example, it can be determined that the comprehensive weight of the business is [0.3, 0.7], which means that compared with the future network performance and user status, the user has a lower demand for the service quality currently provided by the network, which is understandable. The network resources (eg energy requirements) currently required for the terminal device are lower than the network resources required in the future.
在一种可能的实现方式中，终端设备获取所述终端设备的第一历史记录信息；所述第一历史记录信息包括应用的访问记录、内容请求偏好、业务开始时间、业务流量、业务特征和使用时长中的一种或多种；根据所述第一历史记录信息进行预测得到所述预测业务属性信息。具体地，第一历史记录信息可以理解为终端设备之前的历史应用访问记录、用户内容请求偏好、应用使用时长等历史信息；应用的访问记录可以理解为在不同时间用户使用应用的记录；内容请求偏好可以理解为在不同时间用户对内容的请求偏好；使用时长为可以理解为业务持续时间；业务特征可以业务特征包括这个业务对应的网络切片，数据网络名称DNN或者接 入点名称APN等；预测业务属性信息可以理解为基于终端设备历史记录信息通过预测算法(如神经网络算法)得到的终端设备未来一段时间可能运行的业务以及业务可能持续的时间等。In a possible implementation manner, the terminal device acquires first history record information of the terminal device; the first history record information includes application access records, content request preferences, service start time, service flow, service characteristics and Use one or more of the durations; perform prediction according to the first historical record information to obtain the predicted service attribute information. Specifically, the first historical record information can be understood as historical information such as historical application access records, user content request preferences, and application usage duration before the terminal device; application access records can be understood as records of users using applications at different times; content requests Preference can be understood as the user's request preference for content at different times; usage duration can be understood as service duration; service features can include the network slice corresponding to this service, data network name DNN or access point name APN, etc.; prediction The service attribute information can be understood as the service that the terminal device may run for a period of time in the future and the possible duration of the service obtained by a prediction algorithm (such as a neural network algorithm) based on the historical record information of the terminal device.
可选的，网络设备获取终端设备的第一历史记录信息；所述第一历史记录信息包括终端设备的应用的访问记录、内容请求偏好、业务开始时间、业务流量、业务特征和使用时长中的一种或多种；网络设备根据所述第一历史记录信息进行预测得到所述预测业务属性信息，进一步网络设备基于该预测业务属性信息和当前业务属性信息确定业务综合权重。Optionally, the network device obtains the first history record information of the terminal device; the first history record information includes the access record of the application of the terminal device, the content request preference, the service start time, the service flow, the service feature, and the usage duration. One or more: the network device performs prediction according to the first historical record information to obtain the predicted service attribute information, and further, the network device determines the comprehensive service weight based on the predicted service attribute information and the current service attribute information.
需要说明的是，在上述步骤S3001中所涉及的确定终端设备的业务综合权重的实现可以包括如下两种实施方式：It should be noted that the implementation of determining the comprehensive service weight of the terminal device involved in the above step S3001 may include the following two implementations:
实施方式一：由终端设备生成业务综合权重，然后发送给网络设备。Embodiment 1: The terminal device generates the comprehensive service weight, and then sends it to the network device.
在一种可能的实现方式中，终端设备根据所述终端设备的所述当前业务属性信息得到所述当前业务权重，以及根据所述终端设备的所述预测业务属性信息得到所述预测业务权重；向所述网络设备发送所述业务综合权重；所述当前业务属性信息包括当前业务类型、当前业务持续时间和用户偏好中的一种或多种，所述预测业务属性信息包括预测业务类型、预测业务持续时间和所述用户偏好中的一种或多种。具体地，当前业务属性信息可以理解为终端设备当前运行的业务以及当前业务持续的时间和用户偏好等信息；预测业务属性信息可以理解为基于终端设备历史记录信息通过预测算法(如神经网络算法)得到的终端设备未来一段时间可能运行的业务以及业务可能持续的时间等信息。In a possible implementation manner, the terminal device obtains the current service weight according to the current service attribute information of the terminal device, and obtains the predicted service weight according to the predicted service attribute information of the terminal device; Send the comprehensive service weight to the network device; the current service attribute information includes one or more of current service type, current service duration and user preference, and the predicted service attribute information includes predicted service type, prediction One or more of service duration and said user preference. Specifically, the current service attribute information can be understood as information such as the service currently running on the terminal device, the duration of the current service, and user preferences; the predicted service attribute information can be understood as a prediction algorithm (such as a neural network algorithm) based on the historical record information of the terminal device. The obtained information, such as the services that the terminal equipment may run in the future, and the possible duration of the services.
在一种可能的实现方式中，所述网络设备确定所述终端设备的业务综合权重，包括：网络设备接收所述终端设备发送的所述业务综合权重。In a possible implementation manner, the network device determining the comprehensive service weight of the terminal device includes: the network device receiving the comprehensive service weight sent by the terminal device.
实施方式二：由网络设备生成业务综合权重。Embodiment 2: The comprehensive weight of the service is generated by the network device.
在一种可能的实现方式中，所述方法还包括：网络设备获取所述终端设备的当前业务属性信息和预测业务属性信息；根据所述当前业务属性信息确定所述当前业务权重，以及根据所述预测业务属性信息确定所述预测业务权重，得到所述终端设备的业务综合权重；所述当前业务属性信息包括当前业务类型、当前业务持续时间和用户偏好中的一种或多种，所述预测业务属性信息包括预测业务类型、预测业务持续时间和所述用户偏好中的一种或多种。In a possible implementation manner, the method further includes: acquiring the current service attribute information and predicted service attribute information of the terminal device by the network device; determining the current service weight according to the current service attribute information, and The predicted service attribute information determines the predicted service weight, and obtains the comprehensive service weight of the terminal device; the current service attribute information includes one or more of the current service type, current service duration and user preference, the The predicted service attribute information includes one or more of predicted service type, predicted service duration and the user preference.
步骤S3002：网络设备分别确定所述终端设备的M个候选网络的网络效用值。Step S3002: The network device respectively determines the network utility values of the M candidate networks of the terminal device.
每个所述网络效用值包括对应网络的当前网络效用值和预测网络效用值；所述当前网络效用值为基于当前网络属性信息得到的值，所述预测网络效用值为基于预测网络属性信息得到的值。具体地，当终端设备处于连接网络状态时，M个候选网络可以理解为终端设备当前接入的网络以及其相邻的网络，例如，4G网络、5G网络、蜂窝网络、WiFi网络等；当终端设备处于未连接网络状态时，M个候选网络可以理解为终端设备所处区域能够连接的网络；候选网络的网络效用值可以理解为在该网络下用户的业务体验满意度，例如卡顿、流畅等；候选网络的当前网络效用值可以理解为在该网络下用户对当前业务体验的满意度；候选网络的预测网络效用值可以理解为在该网络下用户对预测业务体验满意度。Each of the network utility values includes a current network utility value and a predicted network utility value of the corresponding network; the current network utility value is a value obtained based on the current network attribute information, and the predicted network utility value is obtained based on the predicted network attribute information value of . Specifically, when the terminal device is in a network connection state, the M candidate networks can be understood as the network currently accessed by the terminal device and its adjacent networks, such as 4G network, 5G network, cellular network, WiFi network, etc.; When the device is not connected to the network, the M candidate networks can be understood as the networks that can be connected in the area where the terminal device is located; the network utility value of the candidate network can be understood as the user's service experience satisfaction under the network, such as stuck, smooth etc.; the current network utility value of the candidate network can be understood as the user's satisfaction with the current service experience under this network; the predicted network utility value of the candidate network can be understood as the user's satisfaction with the predicted service experience under this network.
在一种可能的实现方式中，所述方法还包括：网络设备分别获取M个候选网络的当前网络属性信息和预测网络属性信息；所述当前业务属性信息包括当前业务类型、当前业务持续时间和用户偏好中的一种或多种，所述预测业务属性信息包括预测业务类型、预测业务持续时间和所述用户偏好中的一种或多种；基于每个所述候选网络的当前网络属性信息得到所述当前网络效用值；基于每个所述候选网络的预测网络属性信息得到所述预测网络效用值。In a possible implementation manner, the method further includes: the network device obtains current network attribute information and predicted network attribute information of the M candidate networks respectively; the current service attribute information includes the current service type, current service duration and One or more of user preferences, the predicted service attribute information includes one or more of predicted service type, predicted service duration and the user preference; current network attribute information based on each of the candidate networks obtaining the current network utility value; obtaining the predicted network utility value based on the predicted network attribute information of each candidate network.
在一种可能的实现方式中，所述方法还包括：终端设备测量N个候选网络的参考信号接 收功率或参考信号接收质量；判断所述参考信号接收功率或参考信号接收质量是否满足预评估条件；将满足所述预评估条件的M个候选网络确定为所述M个候选网络。具体地，N个候选网络可以理解为能够为UE当前接入的小区以及相邻小区中的至少一个小区，例如，UE当前接入A小区，B、C、D为A小区的相邻小区，则N个候选的网络可以理解为A、B、C、D四个小区中的至少一个小区，或者，也可以将N个候选的网络理解为UE当前接入的网络制式以及可以接入到的其他网络制式中的至少一个网络或者至少一个网络下的一个小区，例如，UE当前接入4G网络且可以接入到5G网络、WiFi网络等，则N个候选的网络可以理解为4G、5G、WiFi三种网络中的至少一个网络或者至少一个网络里的一个小区；参考信号接收功率参考和信号接收质量可以理解为代表无线信号强度和无线信号质量的关键参数；预评估条件可以理解为参考信号接收功率参考和/或信号接收质量满足一定的门限数值，通常为基站配置给终端的用于空闲态执行小区重选的参数以及连接态用于发起小区切换的事件。比如在空闲态的重选条件(邻小区的信号强度测量值较服务小区高于一定门限数值并持续一定时长，比如5G内，邻小区测量值为65dbm，服务小区的测量值为70dbm，并且保持了5s，当门限数值为3dB，时长为3s时，预评估条件满足)在连接态，基站配置给UE的A3事件(相同无线接入技术中，邻小区的信号强度测量值较服务小区高于一定门限数值并持续一定时长，比如5G内，邻小区测量值为65dbm，服务小区的测量值为70dbm，并且保持了5s，当门限数值为3dB，时长为3s则A3事件满足)或B2事件(服务小区的信号强度测量值低于第一门限，并且邻小区的信号强度测量值高于第二门限数值，其中服务小区和相邻小区属于不同无线接入技术的小区)；M个候选网络可以理解为当前接入的网络以及其相邻网络中满足预评估条件的网络。In a possible implementation manner, the method further includes: the terminal device measures the reference signal received power or the reference signal received quality of the N candidate networks; and judges whether the reference signal received power or the reference signal received quality meets a pre-evaluation condition ; Determine the M candidate networks that satisfy the pre-evaluation conditions as the M candidate networks. Specifically, the N candidate networks can be understood as being able to be at least one of the cell that the UE is currently accessing and the neighboring cells, for example, the UE is currently accessing the A cell, B, C, and D are the neighboring cells of the A cell, Then the N candidate networks can be understood as at least one cell among the four cells A, B, C, and D, or the N candidate networks can also be understood as the network standard currently accessed by the UE and the accessible network. At least one network in other network standards or a cell under at least one network, for example, if the UE currently accesses a 4G network and can access a 5G network, a WiFi network, etc., the N candidate networks can be understood as 4G, 5G, At least one of the three WiFi networks or a cell in at least one network; the reference signal received power reference and signal received quality can be understood as the key parameters representing the wireless signal strength and wireless signal quality; the pre-evaluation condition can be understood as the reference signal The received power reference and/or the received signal quality meet a certain threshold value, which are usually parameters configured by the base station to the terminal for performing cell reselection in the idle state and events for initiating cell handover in the connected state. For example, in the re-selection condition in the idle state (the signal strength measurement value of the neighbor cell is higher than a certain threshold value and lasts for a certain period of time, for example, in 5G, the measurement value of the neighbor cell is 65dbm, and the measurement value of the serving cell is 70dbm, and keep When the threshold value is 3dB and the duration is 3s, the pre-evaluation condition is satisfied) In the connected state, the base station configures the A3 event for the UE (in the same wireless access technology, the signal strength measurement value of the neighboring cell is higher than that of the serving cell) A certain threshold value and lasts for a certain period of time. For example, in 5G, the measurement value of the neighboring cell is 65dbm, the measurement value of the serving cell is 70dbm, and it is maintained for 5s. When the threshold value is 3dB and the duration is 3s, the A3 event is satisfied) or the B2 event ( The signal strength measurement value of the serving cell is lower than the first threshold, and the signal strength measurement value of the neighboring cell is higher than the second threshold value, wherein the serving cell and the neighboring cell belong to different radio access technology cells); M candidate networks can be It is understood as the network currently accessed and the network that meets the pre-evaluation conditions in its neighboring networks.
需要说明的是，在上述步骤S3002中所涉及的网络设备确定所述终端设备的M个候选网络的网络效用值的实现可以包括如下两种实施方式：It should be noted that, the network device involved in the above step S3002 determines the network utility value of the M candidate networks of the terminal device. The implementation may include the following two implementations:
实施方式一：由终端设备生成网络效用值。Embodiment 1: The network utility value is generated by the terminal device.
在一种可能的实现方式中，所述终端设备分别确定M个候选网络的网络效用值，包括：终端设备根据所述M个候选网络对应的所述当前网络属性信息确定每个所述候选网络的所述当前网络效用值，以及根据所述M个候选网络对应的所述预测网络属性信息确定每个所述候选网络的所述预测网络效用值，得到所述M个候选网络的所述网络效用值；将所述M个候选网络的所述网络效用值发送给所述网络设备；所述当前网络属性信息包括网络当前的带宽、时延、抖动、丢包率中的一种或多种；所述预测网络属性信息包括网络未来的带宽、时延、抖动、丢包率中的一种或多种。具体地，候选网络的当前网络属性信息可以理解为网络当前的带宽、时延、抖动、丢包率等信息；候选网络的当前网络效用值可以理解为在该网络下用户对当前业务体验的满意度；候选网络的预测网络效用值可以理解为在该网络下用户对预测业务体验满意度。在本发明实施例中，终端设备可基于多个候选网络各自的当前网络属性信息(如网络当前的带宽、时延、抖动、丢包率等)确定当前网络效用值，以及预测网络属性信息(如网络未来的带宽、时延、抖动、丢包率等)确定预测网络效用值，从而得到多个候选网络的网络效用值。当终端设备需要进行智能选网时，就可在终端设备上得到多个候选网络的网络效用值，便于网络设备为终端设备选出的目标网络可以保证通信服务的稳定性，从而减少终端设备频繁切换网络，提升了通信性能指标，实现用户个性化业务体验。In a possible implementation manner, the terminal device determining the network utility values of the M candidate networks respectively includes: the terminal device determining each candidate network according to the current network attribute information corresponding to the M candidate networks and the predicted network utility value of each candidate network is determined according to the predicted network attribute information corresponding to the M candidate networks, and the network of the M candidate networks is obtained. utility value; sending the network utility value of the M candidate networks to the network device; the current network attribute information includes one or more of the current network bandwidth, delay, jitter, and packet loss rate ; The predicted network attribute information includes one or more of the future network bandwidth, time delay, jitter, and packet loss rate. Specifically, the current network attribute information of the candidate network can be understood as the current network bandwidth, delay, jitter, packet loss rate and other information; the current network utility value of the candidate network can be understood as the user's satisfaction with the current service experience under the network degree; the predicted network utility value of the candidate network can be understood as the user's satisfaction with the predicted service experience under this network. In this embodiment of the present invention, the terminal device may determine the current network utility value based on the respective current network attribute information (such as the current network bandwidth, delay, jitter, packet loss rate, etc.) of the multiple candidate networks, and predict the network attribute information ( Such as the future network bandwidth, delay, jitter, packet loss rate, etc.) to determine the predicted network utility value, so as to obtain the network utility value of multiple candidate networks. When the terminal device needs to perform intelligent network selection, the network utility values of multiple candidate networks can be obtained on the terminal device, so that the target network selected by the network device for the terminal device can ensure the stability of communication services, thereby reducing the frequency of terminal devices. Switching the network improves the communication performance index and realizes the user's personalized service experience.
在一种可能的实现方式中，所述当前业务属性信息包括当前业务类型、当前业务持续时间和用户偏好；所述终端设备基于所述当前业务权重和每个所述候选网络的所述当前网络效用值，以及基于所述预测业务权重和每个所述候选网络的所述预测网络效用值，包括：终端 设备根据所述当前业务属性信息得到所述当前业务的服务质量需求；所述服务质量需求包括当前所述终端设备所需要的带宽、时延、抖动和丢包率中的一种或多种；基于所述服务质量和所述用户偏好得到网络效用函数；根据每个所述候选网络的所述当前网络属性信息和所述网络效用函数得到所述当前网络效用值；根据每个所述候选网络的所述预测网络属性信息和所述网络效用函数得到所述预测网络效用值。具体地，服务质量需求可以理解为终端设备运行当前业务所需要的带宽、时延、抖动和丢包率等信息；用户偏好可以理解为用户对业务的通信服务偏好，例如，用户在进行语音电话、游戏、看直播等不同业务时，对网络性能的需求是不同的，此外，不同的用户对同一种业务的通信服务偏好也是不同，有的用户偏好极致的业务体验，可以忽略网络流量的费用，追求大带宽高速率的业务体验，而有的用户则偏好低价的可用服务，对带宽速率的追求不大；网络效用函数可以理解为表示网络服务与用户满意度之间的效用函数。In a possible implementation manner, the current service attribute information includes current service type, current service duration and user preference; the terminal device is based on the current service weight and the current network of each candidate network utility value, and the predicted network utility value based on the predicted service weight and each candidate network, including: the terminal device obtains the service quality requirement of the current service according to the current service attribute information; the service quality The requirements include one or more of the bandwidth, delay, jitter and packet loss rate currently required by the terminal device; a network utility function is obtained based on the quality of service and the user preference; according to each candidate network The current network utility value is obtained from the current network attribute information and the network utility function; the predicted network utility value is obtained according to the predicted network attribute information and the network utility function of each candidate network. Specifically, service quality requirements can be understood as information such as bandwidth, delay, jitter, and packet loss rate required by the terminal device to run the current service; user preferences can be understood as the user's communication service preferences for services, for example, when a user is making a voice call Different services, such as games, live streaming, etc., have different requirements for network performance. In addition, different users have different communication service preferences for the same service. Some users prefer the ultimate service experience, and the cost of network traffic can be ignored. , the pursuit of high-bandwidth and high-speed service experience, while some users prefer low-cost available services, and have little pursuit of bandwidth speed; network utility function can be understood as a utility function between network services and user satisfaction.
在一种可能的实现方式中，所述终端设备基于所述服务质量和所述用户偏好得到网络效用函数，包括：终端设备基于所述服务质量需求得到L个网络决策属性效用函数；基于所述服务质量和所述用户偏好得到所述L个网络决策属性效用函数中的每个所述网络决策属性效用函数对应的网络决策属性权重；根据所述L个网络决策属性效用函数及其对应的所述网络决策属性权重，确定所述网络效用函数。具体地，终端设备的当前业务的服务质量需求中包括了L个网络决策属性(如带宽、时延等)，不同的网络决策属性由相应的网络决策属性效用函数来表征，再结合用户偏好信息确定不同网络决策属性效用函数的权重，进而得到网络效用函数。通过该网络效用函数可反映不同网络决策属性与用户满意度的关系，从而基于该网络效用函数选出的目标网络能够更加准确的反映用户业务意图，实现用户个性化业务体验，且可以保证通信服务的稳定性，减少终端设备频繁切换网络，提升了通信性能指标。In a possible implementation manner, the terminal device obtains a network utility function based on the service quality and the user preference, including: the terminal device obtains L network decision attribute utility functions based on the service quality requirement; The quality of service and the user preference obtain the network decision attribute weight corresponding to each of the L network decision attribute utility functions; according to the L network decision attribute utility functions and their corresponding The network decision attribute weight is used to determine the network utility function. Specifically, the service quality requirements of the current service of the terminal device include L network decision attributes (such as bandwidth, delay, etc.), and different network decision attributes are represented by the corresponding network decision attribute utility functions, and then combined with user preference information. Determine the weight of the utility function of different network decision attributes, and then obtain the network utility function. The network utility function can reflect the relationship between different network decision attributes and user satisfaction, so that the target network selected based on the network utility function can more accurately reflect the user's business intention, realize the user's personalized business experience, and ensure communication services. The stability of the network reduces the frequent switching of the network by the terminal equipment, and improves the communication performance index.
在一种可能的实现方式中，所述网络设备分别确定M个候选网络的网络效用值，包括：接收所述终端设备发送的M个候选网络的所述网络效用值。In a possible implementation manner, the network device respectively determining the network utility values of the M candidate networks includes: receiving the network utility values of the M candidate networks sent by the terminal device.
实施方式二：由网络设备生成网络效用值。Embodiment 2: The network utility value is generated by the network device.
在一种可能的实现方式中，所述方法还包括：网络设备分别获取M个候选网络的当前网络属性信息和预测网络属性信息；所述当前网络属性信息包括网络当前的带宽、时延、抖动、丢包率中的一种或多种；所述预测网络属性信息包括网络未来的带宽、时延、抖动、丢包率中的一种或多种。In a possible implementation manner, the method further includes: the network device obtains current network attribute information and predicted network attribute information of the M candidate networks respectively; the current network attribute information includes the current network bandwidth, delay, jitter , one or more of packet loss rate; the predicted network attribute information includes one or more of future network bandwidth, delay, jitter, and packet loss rate.
在一种可能的实现方式中，所述网络设备根据每个所候选网络对应的所述当前网络属性信息确定所述当前网络效用值，以及根据每个所述候选网络对应的所述预测网络属性信息确定所述预测网络效用值；所述当前网络属性信息包括网络当前的带宽、时延、抖动、丢包率中的一种或多种；所述预测网络属性信息包括网络未来的带宽、时延、抖动、丢包率中的一种或多种；基于每个所述当前网络效用值和每个所述预测网络效用值得到每个网络效用值；所述网络效用值包括同一个所述候选网络的当前网络效应值和预测网络效用值。In a possible implementation manner, the network device determines the current network utility value according to the current network attribute information corresponding to each candidate network, and determines the current network utility value according to the predicted network attribute corresponding to each candidate network information to determine the predicted network utility value; the current network attribute information includes one or more of the current network bandwidth, delay, jitter, and packet loss rate; the predicted network attribute information includes future network bandwidth, time one or more of delay, jitter, and packet loss rate; each network utility value is obtained based on each of the current network utility values and each of the predicted network utility values; the network utility value includes the same The current network effect value and predicted network effect value of the candidate network.
可选的，网络设备获取终端设备的当前业务属性信息，然后网络设备根据终端设备的当前业务属性信息得到所述当前业务的服务质量需求；服务质量需求可包括当前终端设备所需要的带宽、时延、抖动和丢包率中的一种或多种；基于该服务质量和用户偏好得到网络效用函数；将候选网络的当前网络属性信息输入所述网络效用函数，得到所述当前网络效用值；将每个候选网络的所述预测网络属性信息输入所述网络效用函数，得到所述预测网络效用值。具体地，服务质量需求可以理解为终端设备运行当前业务所需要的带宽、时延、抖动和丢包率等信息；用户偏好可以理解为用户对业务的通信服务偏好，例如，用户在进行语音电话、 游戏、看直播等不同业务时，对网络性能的需求是不同的，此外，不同的用户对同一种业务的通信服务偏好也是不同，有的用户偏好极致的业务体验，可以忽略网络流量的费用，追求大带宽高速率的业务体验，而有的用户则偏好低价的可用服务，对带宽速率的追求不大；网络效用函数可以理解为表示网络服务与用户满意度之间的效用函数。Optionally, the network device obtains the current service attribute information of the terminal device, and then the network device obtains the service quality requirement of the current service according to the current service attribute information of the terminal device; one or more of delay, jitter and packet loss rate; obtain a network utility function based on the service quality and user preference; input the current network attribute information of the candidate network into the network utility function to obtain the current network utility value; The predicted network attribute information of each candidate network is input into the network utility function to obtain the predicted network utility value. Specifically, service quality requirements can be understood as information such as bandwidth, delay, jitter, and packet loss rate required by the terminal device to run the current service; user preferences can be understood as the user's communication service preferences for services, for example, when a user is making a voice call Different services such as games, live broadcasts, etc. have different requirements for network performance. In addition, different users have different preferences for communication services of the same service. Some users prefer the ultimate service experience, and the cost of network traffic can be ignored. , the pursuit of high-bandwidth and high-speed service experience, while some users prefer low-cost available services, and have little pursuit of bandwidth speed; network utility function can be understood as a utility function between network services and user satisfaction.
可选的，所述网络设备基于所述服务质量和所述用户偏好得到网络效用函数，包括：网络设备基于所述服务质量需求得到L个网络决策属性效用函数；网络设备基于所述服务质量和所述用户偏好得到所述L个网络决策属性效用函数中的每个所述网络决策属性效用函数对应的网络决策属性权重；根据所述L个网络决策属性效用函数及其对应的所述网络决策属性权重，确定所述网络效用函数。Optionally, the network device obtains a network utility function based on the service quality and the user preference, including: the network device obtains L network decision attribute utility functions based on the service quality requirement; the network device obtains L network decision attribute utility functions based on the service quality and the user preference; The user preference obtains the network decision attribute weight corresponding to each of the network decision attribute utility functions in the L network decision attribute utility functions; according to the L network decision attribute utility functions and the corresponding network decision Attribute weights, which determine the network utility function.
需要说明的是，本申请对于上述步骤S3001和步骤S3002的执行先后顺序不作具体限定，可以先执行步骤S3001再执行步骤S3002，也可以先执行步骤S3002再执行步骤S3001，还可以同时执行步骤S3001和步骤S3002。It should be noted that this application does not specifically limit the execution sequence of the above steps S3001 and S3002. Step S3001 may be executed first and then step S3002 may be executed, or step S3002 may be executed first and then step S3001 may be executed, or steps S3001 and S3001 may be executed simultaneously. Step S3002.
步骤S3003：网络设备基于所述业务综合权重和所述M个候选网络的所述网络效用值，得到M个综合效用值。Step S3003: The network device obtains M comprehensive utility values based on the service comprehensive weight and the network utility values of the M candidate networks.
在一种可能的实现方式中，所述网络设备基于所述业务综合权重和所述M个候选网络的所述网络效用值，得到M个综合效用值，包括：网络设备基于所述当前业务权重和每个所述候选网络的所述当前网络效用值，以及基于所述预测业务权重和每个所述候选网络的所述预测网络效用值，计算每个所述网络的综合效用值。In a possible implementation manner, the network device obtains M comprehensive utility values based on the service comprehensive weight and the network utility values of the M candidate networks, including: the network device is based on the current service weight and the current network utility value for each of the candidate networks, and a composite utility value for each of the candidate networks is calculated based on the predicted traffic weight and the predicted network utility value for each of the candidate networks.
步骤S3004：网络设备基于所述M个综合效用值，从所述M个候选网络中选取所述终端设备的目标网络。Step S3004: The network device selects the target network of the terminal device from the M candidate networks based on the M comprehensive utility values.
具体地，通过上述步骤S3001至步骤S3004可以分别获得M个候选网络的综合效用值，然后可将综合效用值分数最高的候选网络作为目标网络。Specifically, through the above steps S3001 to S3004, the comprehensive utility values of the M candidate networks can be obtained respectively, and then the candidate network with the highest comprehensive utility value score can be used as the target network.
在本发明实施例中，综合考虑终端设备的当前业务状态和未来一段时间内的业务状态，以及该终端设备候选的不同网络的当前状态和未来一段时间的状态，以确定终端设备的当前业务与未来一段时间的预测业务的综合业务权重，以及不同网络的当前效用值和未来一段时间的预测效用值，从而预测并选择出与该终端设备的业务意图最匹配的网络。因此通过实施本发明实施例提供的方法，终端设备可从多个候选网络中选出的目标网络，可以保证通信服务的稳定性，从而减少终端设备频繁切换网络，提升了通信性能指标，实现用户个性化业务体验。In this embodiment of the present invention, the current service status of the terminal device and the service status in a future period of time, as well as the current status of different networks candidate for the terminal equipment and the status of a future period of time are comprehensively considered, so as to determine the current service status of the terminal equipment and the status of the terminal equipment. The comprehensive service weight of the predicted service in the future period, as well as the current utility value of different networks and the predicted utility value of the future period of time, so as to predict and select the network that best matches the service intention of the terminal device. Therefore, by implementing the method provided by the embodiment of the present invention, the terminal device can select a target network from multiple candidate networks, which can ensure the stability of the communication service, thereby reducing the frequent switching of the network by the terminal device, improving the communication performance index, and realizing the user experience. Personalize the business experience.
为了更详细的描述本申请实施例中的智能选网方法，将基于上述智能选网***架构，下面结合图3A和图3B进行示例性描述，图3A为本发明实施例提供的一种智能选网方法示例图流程图，图3B为本发明实施例提供的一种确定网络效用函数方法示例性流程图。其中In order to describe the intelligent network selection method in the embodiments of the present application in more detail, based on the above-mentioned intelligent network selection system architecture, an exemplary description will be given below with reference to FIG. 3A and FIG. 3B , and FIG. 3A is an intelligent network selection method provided by an embodiment of the present invention. An example flow chart of a network method, and FIG. 3B is an example flow chart of a method for determining a network utility function provided by an embodiment of the present invention. in
(1)终端设备智能选网流程可如下：(1) The intelligent network selection process of terminal equipment can be as follows:
步骤401：终端设备测量RSRP或RSRQ。Step 401: The terminal device measures RSRP or RSRQ.
步骤402：终端设备判断网络是否满足预评估条件。Step 402: The terminal device determines whether the network satisfies the pre-evaluation condition.
步骤403：终端设备获取邻小区网络属性信息。Step 403: The terminal device acquires the network attribute information of neighboring cells.
步骤404：终端设备基于用户业务属性信息和网络属性信息，选择综合效用值最大的邻区。Step 404: Based on the user service attribute information and the network attribute information, the terminal device selects the neighbor cell with the largest comprehensive utility value.
步骤405：终端设备执行目标小区选择。例如，当终端设备处于空闲状态(如未接入网络状态)时，终端设备自主选择到目标小区；当终端设备处于连接态时，终端向网络发送测 量报告消息，消息中指示UE优选的目标小区，或者优先的目标小区排序等，UE结合UE发送的信息，选择一个目标小区执行向目标小区的切换过程。Step 405: The terminal equipment performs target cell selection. For example, when the terminal device is in an idle state (such as not accessing the network), the terminal device autonomously selects the target cell; when the terminal device is in a connected state, the terminal sends a measurement report message to the network, which indicates the UE's preferred target cell. , or priority target cell ranking, etc., the UE selects a target cell to perform the handover process to the target cell in combination with the information sent by the UE.
(2)确定网络效用函数的流程可如下：(2) The process of determining the network utility function can be as follows:
步骤1：确定决策属性的权重。Step 1: Determine the weights of the decision attributes.
步骤2：匹配效用函数，确定业务综合权重。Step 2: Match the utility function to determine the comprehensive weight of the business.
步骤3：组合多元效用函数并综合计算。Step 3: Combine multivariate utility functions and comprehensively calculate.
步骤4：网络接入选择。Step 4: Network access selection.
在本发明实施例中，综合考虑终端设备的当前业务状态和未来一段时间内的业务状态，以及不同网络的当前状态和未来一段时间的状态，确定终端设备的当前业务与未来一段时间的预测业务的综合业务权重，以及不同网络的当前效用值和未来一段时间的预测效用值，从而选择出与用户业务意图最匹配的网络。In the embodiment of the present invention, the current service status of the terminal device and the service status in a future period of time, as well as the current status of different networks and the status of a future period of time, are comprehensively considered to determine the current service of the terminal equipment and the predicted service for a future period of time. The comprehensive service weight of the network, as well as the current utility value of different networks and the predicted utility value for a period of time in the future, so as to select the network that best matches the user's service intention.
上述详细阐述了本发明实施例的方法，下面提供了本发明实施例的相关装置。The methods of the embodiments of the present invention are described in detail above, and the related apparatuses of the embodiments of the present invention are provided below.
请参见图4A，图4A是本发明实施例提供的本申请提供了一种智能选网的装置示意图，应用于终端设备，该智能选网装置50，可以包括第一处理单元501、第二处理单元502、第三处理单元503、第四处理单元504、第一获取单元505、第五处理单元506、第六处理单元507，其中各个模块的详细描述如下。Please refer to FIG. 4A . FIG. 4A is a schematic diagram of an apparatus for intelligent network selection provided by the present application according to an embodiment of the present invention, which is applied to terminal equipment. The intelligent network selection apparatus 50 may include a first processing unit 501 and a second processing unit 501 . The unit 502, the third processing unit 503, the fourth processing unit 504, the first obtaining unit 505, the fifth processing unit 506, and the sixth processing unit 507, wherein the detailed description of each module is as follows.
第一处理单元501，用于确定所述终端设备的业务综合权重；所述业务综合权重包括所述终端设备的当前业务权重和预测业务权重；所述当前业务权重为基于所述终端设备当前业务属性信息得到的权重，所述预测业务权重为基于所述终端设备预测业务属性信息得到的权重；The first processing unit 501 is configured to determine the comprehensive service weight of the terminal equipment; the comprehensive service weight includes the current service weight and the predicted service weight of the terminal equipment; the current service weight is based on the current service of the terminal equipment. The weight obtained by the attribute information, the predicted service weight is the weight obtained based on the predicted service attribute information of the terminal device;
第二处理单元502，用于分别确定所述终端设备的M个候选网络的网络效用值；每个所述网络效用值包括对应网络的当前网络效用值和预测网络效用值；所述当前网络效用值为基于当前网络属性信息得到的值，所述预测网络效用值为基于预测网络属性信息得到的值；The second processing unit 502 is configured to respectively determine the network utility values of the M candidate networks of the terminal device; each of the network utility values includes the current network utility value and the predicted network utility value of the corresponding network; the current network utility value The value is a value obtained based on the current network attribute information, and the predicted network utility value is a value obtained based on the predicted network attribute information;
第三处理单元503，用于基于所述业务综合权重和所述M个候选网络的所述网络效用值，得到M个综合效用值；a third processing unit 503, configured to obtain M comprehensive utility values based on the service comprehensive weight and the network utility values of the M candidate networks;
第四处理单元504，用于基于所述M个综合效用值，从所述M个候选网络中选取所述终端设备的目标网络。The fourth processing unit 504 is configured to select a target network of the terminal device from the M candidate networks based on the M comprehensive utility values.
在一种可能的实现方式中，所述第一处理单元501，具有用于：根据所述终端设备的所述当前业务属性信息得到所述当前业务权重，以及根据所述终端设备的所述预测业务属性信息得到所述预测业务权重；所述当前业务属性信息包括当前业务类型、当前业务开始时间、当前业务持续时间、业务流量、业务特征和用户偏好中的一种或多种，所述预测业务属性信息包括预测业务类型、预测业务开始时间、预测业务持续时间、预测业务流量、预测业务特征和所述用户偏好中的一种或多种。In a possible implementation manner, the first processing unit 501 is configured to: obtain the current service weight according to the current service attribute information of the terminal device, and obtain the current service weight according to the prediction of the terminal device The predicted service weight is obtained from the service attribute information; the current service attribute information includes one or more of the current service type, the current service start time, the current service duration, the service flow, the service characteristics and the user preference. The service attribute information includes one or more of predicted service type, predicted service start time, predicted service duration, predicted service flow, predicted service characteristics and the user preference.
在一种可能的实现方式中，所述第一处理单元501，具体用于：向网络设备发送所述终端设备的所述当前业务属性信息和所述预测业务属性信息；所述网络设备为与M个候选网络中的一个或多个候选网络关联的网络设备；接收所述网络设备发送的所述业务综合权重。In a possible implementation manner, the first processing unit 501 is specifically configured to: send the current service attribute information and the predicted service attribute information of the terminal device to a network device; the network device is a A network device associated with one or more candidate networks in the M candidate networks; and receiving the service comprehensive weight sent by the network device.
在一种可能的实现方式中，所述第二处理单元502，具体用于：根据所述M个候选网络对应的所述当前网络属性信息确定每个所述候选网络的所述当前网络效用值，以及根据所述M个候选网络对应的所述预测网络属性信息确定每个所述候选网络的所述预测网络效用值，得到所述M个候选网络的所述网络效用值；所述当前网络属性信息包括网络当前的带宽、时 延、抖动、丢包率中的一种或多种；所述预测网络属性信息包括网络未来的带宽、时延、抖动、丢包率中的一种或多种。In a possible implementation manner, the second processing unit 502 is specifically configured to: determine the current network utility value of each candidate network according to the current network attribute information corresponding to the M candidate networks , and determine the predicted network utility value of each candidate network according to the predicted network attribute information corresponding to the M candidate networks to obtain the network utility value of the M candidate networks; the current network The attribute information includes one or more of the current network bandwidth, delay, jitter, and packet loss rate; the predicted network attribute information includes one or more of the future network bandwidth, delay, jitter, and packet loss rate. kind.
在一种可能的实现方式中，所述第二处理单元502，具体用于：接收所述提供M个候选网络关联的一个或多个网络设备分别发送的所述M个候选网络的所述网络效用值。In a possible implementation manner, the second processing unit 502 is specifically configured to: receive the network of the M candidate networks respectively sent by the one or more network devices that provide M candidate network associations utility value.
在一种可能的实现方式中，所述第三处理单元503，具体用于：基于所述当前业务权重和每个所述候选网络的所述当前网络效用值，以及基于所述预测业务权重和每个所述候选网络的所述预测网络效用值，计算每个所述网络的综合效用值。In a possible implementation manner, the third processing unit 503 is specifically configured to: based on the current service weight and the current network utility value of each candidate network, and based on the predicted service weight and The predicted network utility value of each of the candidate networks, and a comprehensive utility value of each of the networks is calculated.
在一种可能的实现方式中，所述装置还包括：第一获取单元505，用于获取所述终端设备的第一历史记录信息；所述第一历史记录信息包括应用的访问记录、内容请求偏好、业务开始时间、业务流量、业务特征和使用时长中的一种或多种；第五处理单元506，用于根据所述第一历史记录信息进行预测得到所述预测业务属性信息。In a possible implementation manner, the apparatus further includes: a first obtaining unit 505, configured to obtain first history record information of the terminal device; the first history record information includes application access records and content requests One or more of preferences, service start time, service flow, service characteristics and usage duration; the fifth processing unit 506 is configured to perform prediction according to the first historical record information to obtain the predicted service attribute information.
在一种可能的实现方式中，所述第二处理单元502，具体用于：根据所述当前业务属性信息得到所述当前业务的服务质量需求；所述服务质量需求包括当前所述终端设备所需要的带宽、时延、抖动和丢包率中的一种或多种；基于所述服务质量和所述用户偏好得到网络效用函数；根据每个所述候选网络的所述当前网络属性信息和所述网络效用函数得到所述当前网络效用值；根据每个所述候选网络的所述预测网络属性信息和所述网络效用函数得到所述预测网络效用值。In a possible implementation manner, the second processing unit 502 is specifically configured to: obtain the service quality requirement of the current service according to the current service attribute information; the service quality requirement includes the current status of the terminal device. One or more of the required bandwidth, delay, jitter and packet loss rate; obtain a network utility function based on the quality of service and the user preference; according to the current network attribute information of each of the candidate networks and The network utility function obtains the current network utility value; and the predicted network utility value is obtained according to the predicted network attribute information of each candidate network and the network utility function.
在一种可能的实现方式中，所述第二处理单元502，具体用于：基于所述服务质量需求得到L个网络决策属性效用函数；基于所述服务质量和所述用户偏好得到所述L个网络决策属性效用函数中的每个所述网络决策属性效用函数对应的网络决策属性权重；根据所述L个网络决策属性效用函数及其对应的所述网络决策属性权重，确定所述网络效用函数。In a possible implementation manner, the second processing unit 502 is specifically configured to: obtain L network decision attribute utility functions based on the service quality requirement; obtain the L network decision attribute utility functions based on the service quality and the user preference The network decision attribute weight corresponding to each of the network decision attribute utility functions in the network decision attribute utility functions; according to the L network decision attribute utility functions and their corresponding network decision attribute weights, the network utility is determined function.
在一种可能的实现方式中，所述装置还包括：In a possible implementation, the apparatus further includes:
第六处理单元507，用于测量N个候选网络的参考信号接收功率或参考信号接收质量；判断所述参考信号接收功率或参考信号接收质量是否满足预评估条件；将满足所述预评估条件的M个候选网络确定为所述M个候选网络。The sixth processing unit 507 is configured to measure the reference signal received power or reference signal received quality of the N candidate networks; determine whether the reference signal received power or reference signal received quality satisfies the pre-evaluation condition; M candidate networks are determined as the M candidate networks.
需要说明的是，本发明实施例中所描述的智能选网装置50中各功能单元的功能可参见上述方法实施例中终端设备所执行的步骤的相关描述，此处不再赘述。It should be noted that, for the functions of each functional unit in the intelligent network selection device 50 described in the embodiments of the present invention, reference may be made to the relevant description of the steps performed by the terminal device in the above method embodiments, and details are not repeated here.
请参见图4B，图4B是本发明实施例提供的本申请提供了一种智能选网的装置示意图，应用于网络设备，该智能选网装置60，可以包括第一获取单元601、第一处理单元602、第二处理单元603、第一发送单元604、第三处理单元605、第二发送单元606，其中各个模块的详细描述如下。Please refer to FIG. 4B . FIG. 4B is a schematic diagram of an apparatus for intelligent network selection provided by the present application according to an embodiment of the present invention, which is applied to network equipment. The intelligent network selection apparatus 60 may include a first obtaining unit 601 , a first processing The unit 602, the second processing unit 603, the first sending unit 604, the third processing unit 605, and the second sending unit 606, wherein the detailed description of each module is as follows.
第一获取单元601，用于获取所述S个候选网络中的每个候选网络的第二历史记录信息；所述第二历史记录信息包括每个小区的历史带宽、时延、抖动、丢包率中的一种或多种中的一种或多种；The first obtaining unit 601 is configured to obtain the second historical record information of each candidate network in the S candidate networks; the second historical record information includes the historical bandwidth, time delay, jitter, and packet loss of each cell one or more of one or more of the rates;
第一处理单元602，用于根据每个所述第二历史记录信息进行预测得到S个预测网络属性信息。The first processing unit 602 is configured to perform prediction according to each of the second historical record information to obtain S pieces of predicted network attribute information.
在一种可能的实现方式中，所述装置还包括：第二处理单元603，用于获取所述终端设备的当前业务属性信息和预测业务属性信息；根据所述当前业务属性信息确定所述当前业务权重，以及根据所述预测业务属性信息确定所述预测业务权重，得到所述终端设备的业务综合权重；所述当前业务属性信息包括当前业务类型、当前业务持续时间和用户偏好中的一种 或多种，所述预测业务属性信息包括预测业务类型、预测业务持续时间和所述用户偏好中的一种或多种；第一发送单元604，用于向所述终端设备发送所述业务综合权重。In a possible implementation manner, the apparatus further includes: a second processing unit 603, configured to acquire current service attribute information and predicted service attribute information of the terminal device; determine the current service attribute information according to the current service attribute information service weight, and determining the predicted service weight according to the predicted service attribute information to obtain the comprehensive service weight of the terminal device; the current service attribute information includes one of the current service type, current service duration and user preference or more, the predicted service attribute information includes one or more of the predicted service type, the predicted service duration and the user preference; the first sending unit 604 is configured to send the comprehensive service to the terminal device Weights.
在一种可能的实现方式中，所述装置还包括：第三处理单元605，用于根据每个所候选网络对应的所述当前网络属性信息确定所述当前网络效用值，以及根据每个所述候选网络对应的所述预测网络属性信息确定所述预测网络效用值；所述当前网络属性信息包括网络当前的带宽、时延、抖动、丢包率中的一种或多种；所述预测网络属性信息包括网络未来的带宽、时延、抖动、丢包率中的一种或多种；基于每个所述当前网络效用值和每个所述预测网络效用值得到S个网络效用值；所述网络效用值包括同一个所述候选网络的当前网络效应值和预测网络效用值；第二发送单元606，用于向所述终端设备发送所述S个候选网络对应的所述S个网络效用值。In a possible implementation manner, the apparatus further includes: a third processing unit 605, configured to determine the current network utility value according to the current network attribute information corresponding to each candidate network, and according to each candidate network The predicted network attribute information corresponding to the candidate network determines the predicted network utility value; the current network attribute information includes one or more of the current network bandwidth, delay, jitter, and packet loss rate; the predicted network The network attribute information includes one or more of future network bandwidth, delay, jitter, and packet loss rate; S network utility values are obtained based on each of the current network utility values and each of the predicted network utility values; The network utility value includes the current network effect value and the predicted network utility value of the same candidate network; the second sending unit 606 is configured to send the S networks corresponding to the S candidate networks to the terminal device utility value.
需要说明的是，本发明实施例中所描述的智能选网装置50中各功能单元的功能可参见上述方法实施例中网络设备所执行的步骤的相关描述，此处不再赘述。It should be noted that, for the functions of each functional unit in the intelligent network selection device 50 described in the embodiments of the present invention, reference may be made to the relevant descriptions of the steps performed by the network equipment in the foregoing method embodiments, which will not be repeated here.
本申请提供了一种计算机存储介质，其特征在于，所述计算机存储介质存储有计算机程序，该计算机程序被处理器执行时实现上述任意一种智能选网的方法。The present application provides a computer storage medium, characterized in that the computer storage medium stores a computer program, and when the computer program is executed by a processor, any one of the above-mentioned methods for intelligent network selection is implemented.
本申请提供了一种计算机存储介质，其特征在于，所述计算机存储介质存储有计算机程序，该计算机程序被处理器执行时实现上述任意另一种智能选网的方法。The present application provides a computer storage medium, characterized in that, the computer storage medium stores a computer program, and when the computer program is executed by a processor, any of the above-mentioned methods for intelligent network selection is implemented.
本申请提供了一种计算机存储介质，其特征在于，所述计算机存储介质存储有计算机程序，该计算机程序被处理器执行时实现上述任意一种智能选网的方法。The present application provides a computer storage medium, characterized in that the computer storage medium stores a computer program, and when the computer program is executed by a processor, any one of the above-mentioned methods for intelligent network selection is implemented.
本申请提供了一种计算机存储介质，其特征在于，所述计算机存储介质存储有计算机程序，该计算机程序被处理器执行时实现上述任意另一种智能选网的方法。The present application provides a computer storage medium, characterized in that, the computer storage medium stores a computer program, and when the computer program is executed by a processor, any of the above-mentioned methods for intelligent network selection is implemented.
本发明实施例提供一种电子设备，该电子设备中包括处理器，处理器被配置为支持该电子设备实现上述任意一种智能选网的方法中相应的功能。该电子设备还可以包括存储器，存储器用于与处理器耦合，其保存该电子设备必要的程序指令和数据。该电子设备还可以包括通信接口，用于该电子设备与其他设备或通信网络通信。An embodiment of the present invention provides an electronic device, the electronic device includes a processor, and the processor is configured to support the electronic device to implement corresponding functions in any of the foregoing methods for intelligent network selection. The electronic device may also include a memory for coupling with the processor that holds program instructions and data necessary for the electronic device. The electronic device may also include a communication interface for the electronic device to communicate with other devices or a communication network.
本发明实施例提供一种电子设备，该电子设备中包括处理器，处理器被配置为支持该电子设备实现上述任意另一种智能选网的方法中相应的功能。该电子设备还可以包括存储器，存储器用于与处理器耦合，其保存该电子设备必要的程序指令和数据。该电子设备还可以包括通信接口，用于该电子设备与其他设备或通信网络通信。An embodiment of the present invention provides an electronic device, the electronic device includes a processor, and the processor is configured to support the electronic device to implement corresponding functions in any of the above-mentioned methods for intelligent network selection. The electronic device may also include a memory for coupling with the processor that holds program instructions and data necessary for the electronic device. The electronic device may also include a communication interface for the electronic device to communicate with other devices or a communication network.
本申请提供了一种芯片***，该芯片***包括处理器，用于支持电子设备实现上述所涉及的功能，例如，生成或处理上述一种智能选网的方法中所涉及的信息。在一种可能的设计中，所述芯片***还包括存储器，所述存储器，用于保存电子设备必要的程序指令和数据。该芯片***，可以由芯片构成，也可以包含芯片和其他分立器件。The present application provides a chip system, the chip system includes a processor for supporting an electronic device to implement the above-mentioned functions, for example, generating or processing the information involved in the above-mentioned method for intelligent network selection. In a possible design, the chip system further includes a memory for storing necessary program instructions and data of the electronic device. The chip system may be composed of chips, or may include chips and other discrete devices.
本申请提供了一种芯片***，该芯片***包括处理器，用于支持电子设备实现上述所涉 及的功能，例如，生成或处理上述另一种智能选网的方法中所涉及的信息。在一种可能的设计中，所述芯片***还包括存储器，所述存储器，用于保存电子设备必要的程序指令和数据。该芯片***，可以由芯片构成，也可以包含芯片和其他分立器件。The present application provides a chip system, the chip system includes a processor, which is used for supporting the electronic device to realize the above-mentioned functions, for example, generating or processing the information involved in the above-mentioned another method for intelligent network selection. In a possible design, the chip system further includes a memory for storing necessary program instructions and data of the electronic device. The chip system may be composed of chips, or may include chips and other discrete devices.
在上述实施例中，对各个实施例的描述都各有侧重，某个实施例中没有详述的部分，可以参见其他实施例的相关描述。In the above-mentioned embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference may be made to the relevant descriptions of other embodiments.
需要说明的是，对于前述的各方法实施例，为了简单描述，故将其都表述为一系列的动作组合，但是本领域技术人员应该知悉，本申请并不受所描述的动作顺序的限制，因为依据本申请，某些步骤可能可以采用其他顺序或者同时进行。其次，本领域技术人员也应该知悉，说明书中所描述的实施例均属于优选实施例，所涉及的动作和模块并不一定是本申请所必须的。It should be noted that, for the sake of simple description, the foregoing method embodiments are all expressed as a series of action combinations, but those skilled in the art should know that the present application is not limited by the described action sequence. Because in accordance with the present application, certain steps may be performed in other orders or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.
在本申请所提供的几个实施例中，应该理解到，所揭露的装置，可通过其它的方式实现。例如，以上所描述的装置实施例仅仅是示意性的，例如上述单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个***，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，装置或单元的间接耦合或通信连接，可以是电性或其它的形式。In the several embodiments provided in this application, it should be understood that the disclosed apparatus may be implemented in other manners. For example, the device embodiments described above are only illustrative. For example, the division of the above-mentioned units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated. to another system, or some features can be ignored, or not implemented. On the other hand, the shown or discussed mutual coupling or direct coupling or communication connection may be through some interfaces, indirect coupling or communication connection of devices or units, and may be in electrical or other forms.
上述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。The units described above as separate components may or may not be physically separated, and components shown as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution in this embodiment.
另外，在本申请各实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现，也可以采用软件功能单元的形式实现。In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically alone, or two or more units may be integrated into one unit. The above-mentioned integrated units may be implemented in the form of hardware, or may be implemented in the form of software functional units.
上述集成的单元如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的全部或部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以为个人计算机、服务器或者网络设备等，具体可以是计算机设备中的处理器)执行本申请各个实施例上述方法的全部或部分步骤。其中，而前述的存储介质可包括：U盘、移动硬盘、磁碟、光盘、只读存储器(Read-Only Memory，缩写：ROM)或者随机存取存储器(Random Access Memory，缩写：RAM)等各种可以存储程序代码的介质。If the above-mentioned integrated units are implemented in the form of software functional units and sold or used as independent products, they may be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the present application can be embodied in the form of software products in essence, or the parts that contribute to the prior art, or all or part of the technical solutions, and the computer software products are stored in a storage medium , including several instructions to cause a computer device (which may be a personal computer, a server, or a network device, etc., specifically a processor in the computer device) to execute all or part of the steps of the foregoing methods in the various embodiments of the present application. Wherein, the aforementioned storage medium may include: U disk, mobile hard disk, magnetic disk, optical disk, Read-Only Memory (Read-Only Memory, abbreviation: ROM) or Random Access Memory (Random Access Memory, abbreviation: RAM), etc. A medium that can store program code.
以上所述，以上实施例仅用以说明本申请的技术方案，而非对其限制；尽管参照前述实施例对本申请进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述各实施例所记载的技术方案进行修改，或者对其中部分技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本申请各实施例技术方案的精神和范围。As mentioned above, the above embodiments are only used to illustrate the technical solutions of the present application, but not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand: The technical solutions described in the embodiments are modified, or some technical features thereof are equivalently replaced; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions in the embodiments of the present application.

Claims (29)

  1. 一种智能选网的方法，其特征在于，应用于终端设备，所述方法包括：A method for intelligent network selection, characterized in that it is applied to terminal equipment, the method comprising:
    确定所述终端设备的业务综合权重；所述业务综合权重包括所述终端设备的当前业务权重和预测业务权重；所述当前业务权重为基于所述终端设备当前业务属性信息得到的权重，所述预测业务权重为基于所述终端设备预测业务属性信息得到的权重；Determine the comprehensive service weight of the terminal device; the comprehensive service weight includes the current service weight and the predicted service weight of the terminal device; the current service weight is a weight obtained based on the current service attribute information of the terminal device, the The predicted service weight is a weight obtained based on the predicted service attribute information of the terminal device;
    分别确定所述终端设备的M个候选网络的网络效用值；每个所述网络效用值包括对应网络的当前网络效用值和预测网络效用值；所述当前网络效用值为基于当前网络属性信息得到的值，所述预测网络效用值为基于预测网络属性信息得到的值；Determine the network utility values of the M candidate networks of the terminal device respectively; each of the network utility values includes the current network utility value and the predicted network utility value of the corresponding network; the current network utility value is obtained based on the current network attribute information The value of the predicted network utility value is obtained based on the predicted network attribute information;
    基于所述业务综合权重和所述M个候选网络的所述网络效用值，得到M个综合效用值；Obtain M comprehensive utility values based on the service comprehensive weight and the network utility values of the M candidate networks;
    基于所述M个综合效用值，从所述M个候选网络中选取所述终端设备的目标网络。Based on the M comprehensive utility values, a target network of the terminal device is selected from the M candidate networks.
  2. 如权利要求1所述的方法，其特征在于，所述确定所述终端设备的业务综合权重，包括：The method according to claim 1, wherein the determining the comprehensive service weight of the terminal device comprises:
    根据所述终端设备的所述当前业务属性信息得到所述当前业务权重，以及根据所述终端设备的所述预测业务属性信息得到所述预测业务权重；所述当前业务属性信息包括当前业务类型、当前业务开始时间、当前业务持续时间、业务流量、业务特征和用户偏好中的一种或多种，所述预测业务属性信息包括预测业务类型、预测业务开始时间、预测业务持续时间、预测业务流量、预测业务特征和所述用户偏好中的一种或多种。The current service weight is obtained according to the current service attribute information of the terminal device, and the predicted service weight is obtained according to the predicted service attribute information of the terminal device; the current service attribute information includes the current service type, One or more of the current service start time, current service duration, service flow, service characteristics and user preference, and the predicted service attribute information includes predicted service type, predicted service start time, predicted service duration, predicted service flow , one or more of predicted business characteristics and said user preferences.
  3. 如权利要求1所述的方法，其特征在于，所述确定所述终端设备的业务综合权重，包括：The method according to claim 1, wherein the determining the comprehensive service weight of the terminal device comprises:
    向网络设备发送所述终端设备的所述当前业务属性信息和所述预测业务属性信息；所述网络设备为与M个候选网络中的一个或多个候选网络关联的网络设备；sending the current service attribute information and the predicted service attribute information of the terminal device to a network device; the network device is a network device associated with one or more candidate networks in the M candidate networks;
    接收所述网络设备发送的所述业务综合权重。The comprehensive weight of the service sent by the network device is received.
  4. 如权利要求1所述的方法，其特征在于，所述分别确定M个候选网络的网络效用值，包括：The method of claim 1, wherein the determining the network utility values of the M candidate networks respectively comprises:
    根据所述M个候选网络对应的所述当前网络属性信息确定每个所述候选网络的所述当前网络效用值，以及根据所述M个候选网络对应的所述预测网络属性信息确定每个所述候选网络的所述预测网络效用值，得到所述M个候选网络的所述网络效用值；所述当前网络属性信息包括网络当前的带宽、时延、抖动、丢包率中的一种或多种；所述预测网络属性信息包括网络未来的带宽、时延、抖动、丢包率中的一种或多种。The current network utility value of each candidate network is determined according to the current network attribute information corresponding to the M candidate networks, and the current network utility value of each candidate network is determined according to the predicted network attribute information corresponding to the M candidate networks. The predicted network utility value of the candidate network is obtained, and the network utility value of the M candidate networks is obtained; the current network attribute information includes one of the current network bandwidth, delay, jitter, and packet loss rate or The predicted network attribute information includes one or more of the future network bandwidth, delay, jitter, and packet loss rate.
  5. 如权利要求1所述的方法，其特征在于，所述分别确定M个候选网络的网络效用值，包括：The method of claim 1, wherein the determining the network utility values of the M candidate networks respectively comprises:
    接收所述提供M个候选网络关联的一个或多个网络设备分别发送的所述M个候选网络的所述网络效用值。The network utility values of the M candidate networks respectively sent by the one or more network devices providing the M candidate network associations are received.
  6. 如权利要求1-5任意一项所述的方法，其特征在于，所述基于所述业务综合权重和所述M个候选网络的所述网络效用值，得到M个综合效用值，包括：The method according to any one of claims 1-5, wherein the obtaining M comprehensive utility values based on the service comprehensive weight and the network utility values of the M candidate networks, comprising:
    基于所述当前业务权重和每个所述候选网络的所述当前网络效用值，以及基于所述预测业务权重和每个所述候选网络的所述预测网络效用值，计算每个所述网络的综合效用值。Calculate the network utility value for each of the candidate networks based on the current traffic weight and the current network utility value for each of the candidate networks, and based on the predicted traffic weight and the predicted network utility value for each of the candidate networks Comprehensive utility value.
  7. 如权利要求1-6任意一项所述的方法，其特征在于，所述方法还包括：The method according to any one of claims 1-6, wherein the method further comprises:
    获取所述终端设备的第一历史记录信息；所述第一历史记录信息包括应用的访问记录、内容请求偏好、业务开始时间、业务流量、业务特征和使用时长中的一种或多种；Obtain the first history record information of the terminal device; the first history record information includes one or more of application access records, content request preferences, service start time, service flow, service characteristics and usage duration;
    根据所述第一历史记录信息进行预测得到所述预测业务属性信息。The predicted service attribute information is obtained by predicting according to the first historical record information.
  8. 如权利要求4所述的方法，其特征在于，所述当前业务属性信息包括当前业务类型、当前业务持续时间和用户偏好；所述基于所述当前业务权重和每个所述候选网络的所述当前网络效用值，以及基于所述预测业务权重和每个所述候选网络的所述预测网络效用值，包括：The method of claim 4, wherein the current service attribute information includes current service type, current service duration and user preference; the current service weight and the The current network utility value, and the predicted network utility value based on the predicted traffic weight and each of the candidate networks, include:
    根据所述当前业务属性信息得到所述当前业务的服务质量需求；所述服务质量需求包括当前所述终端设备所需要的带宽、时延、抖动和丢包率中的一种或多种；Obtain the service quality requirement of the current service according to the current service attribute information; the service quality requirement includes one or more of the bandwidth, delay, jitter and packet loss rate currently required by the terminal device;
    基于所述服务质量和所述用户偏好得到网络效用函数；obtaining a network utility function based on the quality of service and the user preference;
    根据每个所述候选网络的所述当前网络属性信息和所述网络效用函数得到所述当前网络效用值；obtaining the current network utility value according to the current network attribute information and the network utility function of each of the candidate networks;
    根据每个所述候选网络的所述预测网络属性信息和所述网络效用函数得到所述预测网络效用值。The predicted network utility value is obtained according to the predicted network attribute information and the network utility function of each of the candidate networks.
  9. 如权利要求8所述的方法，其特征在于，所述基于所述服务质量和所述用户偏好得到网络效用函数，包括：The method of claim 8, wherein the obtaining a network utility function based on the quality of service and the user preference comprises:
    基于所述服务质量需求得到L个网络决策属性效用函数；obtaining L network decision attribute utility functions based on the service quality requirements;
    基于所述服务质量和所述用户偏好得到所述L个网络决策属性效用函数中的每个所述网络决策属性效用函数对应的网络决策属性权重；Obtaining a network decision attribute weight corresponding to each of the network decision attribute utility functions in the L network decision attribute utility functions based on the service quality and the user preference;
    根据所述L个网络决策属性效用函数及其对应的所述网络决策属性权重，确定所述网络效用函数。The network utility function is determined according to the L network decision attribute utility functions and their corresponding network decision attribute weights.
  10. 如权利要求1-9任意一项所述的方法，其特征在于，所述方法还包括：The method according to any one of claims 1-9, wherein the method further comprises:
    测量N个候选网络的参考信号接收功率或参考信号接收质量；Measure the reference signal received power or reference signal received quality of the N candidate networks;
    判断所述参考信号接收功率或参考信号接收质量是否满足预评估条件；Judging whether the received power of the reference signal or the received quality of the reference signal satisfies the pre-evaluation condition;
    将满足所述预评估条件的M个候选网络确定为所述M个候选网络。M candidate networks that satisfy the pre-evaluation conditions are determined as the M candidate networks.
  11. 一种智能选网的方法，其特征在于，应用于网络设备，所述网络设备提供S个候选网络；所述S个候选网络为所述终端设备的M个候选网络中的网络；所述方法包括：A method for intelligent network selection, characterized in that it is applied to a network device, and the network device provides S candidate networks; the S candidate networks are networks among M candidate networks of the terminal device; the method include:
    获取所述S个候选网络中的每个候选网络的第二历史记录信息；所述第二历史记录信息包括每个小区的历史带宽、时延、抖动、丢包率中的一种或多种；Obtain the second history record information of each candidate network in the S candidate networks; the second history record information includes one or more of the historical bandwidth, time delay, jitter, and packet loss rate of each cell ;
    根据每个所述第二历史记录信息进行预测得到S个预测网络属性信息。S pieces of predicted network attribute information are obtained by performing prediction according to each of the second historical record information.
  12. 如权利要求11所述的方法，其特征在于，所述方法还包括：The method of claim 11, wherein the method further comprises:
    获取所述终端设备的当前业务属性信息和预测业务属性信息；Acquire current service attribute information and predicted service attribute information of the terminal device;
    根据所述当前业务属性信息确定所述当前业务权重，以及根据所述预测业务属性信息确定所述预测业务权重，得到所述终端设备的业务综合权重；所述当前业务属性信息包括当前业务类型、当前业务持续时间和用户偏好中的一种或多种，所述预测业务属性信息包括预测业务类型、预测业务持续时间和所述用户偏好中的一种或多种；The current service weight is determined according to the current service attribute information, and the predicted service weight is determined according to the predicted service attribute information to obtain the comprehensive service weight of the terminal device; the current service attribute information includes the current service type, One or more of current service duration and user preference, and the predicted service attribute information includes one or more of predicted service type, predicted service duration and user preference;
    向所述终端设备发送所述业务综合权重。Send the service comprehensive weight to the terminal device.
  13. 如权利要求11所述的方法，其特征在于，所述方法还包括：The method of claim 11, wherein the method further comprises:
    根据每个所候选网络对应的所述当前网络属性信息确定所述当前网络效用值，以及根据每个所述候选网络对应的所述预测网络属性信息确定所述预测网络效用值；所述当前网络属性信息包括网络当前的带宽、时延、抖动、丢包率中的一种或多种；所述预测网络属性信息包括网络未来的带宽、时延、抖动、丢包率中的一种或多种；The current network utility value is determined according to the current network attribute information corresponding to each candidate network, and the predicted network utility value is determined according to the predicted network attribute information corresponding to each candidate network; the current network The attribute information includes one or more of the current network bandwidth, delay, jitter, and packet loss rate; the predicted network attribute information includes one or more of the future network bandwidth, delay, jitter, and packet loss rate. kind;
    基于每个所述当前网络效用值和每个所述预测网络效用值得到S个网络效用值；所述网 络效用值包括同一个所述候选网络的当前网络效应值和预测网络效用值；S network utility values are obtained based on each of the current network utility values and each of the predicted network utility values; the network utility values include the current network effect value and the predicted network utility value of the same candidate network;
    向所述终端设备发送所述S个候选网络对应的所述S个网络效用值。The S network utility values corresponding to the S candidate networks are sent to the terminal device.
  14. 一种智能选网的装置，其特征在于，应用于终端设备，所述装置包括：A device for intelligent network selection, characterized in that it is applied to terminal equipment, the device comprising:
    第一处理单元，用于确定所述终端设备的业务综合权重；所述业务综合权重包括所述终端设备的当前业务权重和预测业务权重；所述当前业务权重为基于所述终端设备当前业务属性信息得到的权重，所述预测业务权重为基于所述终端设备预测业务属性信息得到的权重；a first processing unit, configured to determine the comprehensive service weight of the terminal equipment; the comprehensive service weight includes the current service weight and the predicted service weight of the terminal equipment; the current service weight is based on the current service attribute of the terminal equipment The weight obtained by the information, the predicted service weight is the weight obtained based on the predicted service attribute information of the terminal device;
    第二处理单元，用于分别确定所述终端设备的M个候选网络的网络效用值；每个所述网络效用值包括对应网络的当前网络效用值和预测网络效用值；所述当前网络效用值为基于当前网络属性信息得到的值，所述预测网络效用值为基于预测网络属性信息得到的值；a second processing unit, configured to respectively determine the network utility values of the M candidate networks of the terminal device; each of the network utility values includes the current network utility value and the predicted network utility value of the corresponding network; the current network utility value is a value obtained based on the current network attribute information, and the predicted network utility value is a value obtained based on the predicted network attribute information;
    第三处理单元，用于基于所述业务综合权重和所述M个候选网络的所述网络效用值，得到M个综合效用值；a third processing unit, configured to obtain M comprehensive utility values based on the service comprehensive weight and the network utility values of the M candidate networks;
    第四处理单元，用于基于所述M个综合效用值，从所述M个候选网络中选取所述终端设备的目标网络。The fourth processing unit is configured to select a target network of the terminal device from the M candidate networks based on the M comprehensive utility values.
  15. 如权利要求14所述的装置，其特征在于，所述第一处理单元，具有用于：The apparatus of claim 14, wherein the first processing unit is configured to:
    根据所述终端设备的所述当前业务属性信息得到所述当前业务权重，以及根据所述终端设备的所述预测业务属性信息得到所述预测业务权重；所述当前业务属性信息包括当前业务类型、当前业务开始时间、当前业务持续时间、业务流量、业务特征和用户偏好中的一种或多种，所述预测业务属性信息包括预测业务类型、预测业务开始时间、预测业务持续时间、预测业务流量、预测业务特征和所述用户偏好中的一种或多种。The current service weight is obtained according to the current service attribute information of the terminal device, and the predicted service weight is obtained according to the predicted service attribute information of the terminal device; the current service attribute information includes the current service type, One or more of the current service start time, current service duration, service flow, service characteristics and user preference, and the predicted service attribute information includes predicted service type, predicted service start time, predicted service duration, predicted service flow , one or more of predicted business characteristics and said user preferences.
  16. 如权利要求14所述的装置，其特征在于，所述第一处理单元，具体用于：The apparatus of claim 14, wherein the first processing unit is specifically configured to:
    向网络设备发送所述终端设备的所述当前业务属性信息和所述预测业务属性信息；所述网络设备为与M个候选网络中的一个或多个候选网络关联的网络设备；sending the current service attribute information and the predicted service attribute information of the terminal device to a network device; the network device is a network device associated with one or more candidate networks in the M candidate networks;
    接收所述网络设备发送的所述业务综合权重。The comprehensive weight of the service sent by the network device is received.
  17. 如权利要求14所述的装置，其特征在于，所述第二处理单元，具体用于：The apparatus of claim 14, wherein the second processing unit is specifically configured to:
    根据所述M个候选网络对应的所述当前网络属性信息确定每个所述候选网络的所述当前网络效用值，以及根据所述M个候选网络对应的所述预测网络属性信息确定每个所述候选网络的所述预测网络效用值，得到所述M个候选网络的所述网络效用值；所述当前网络属性信息包括网络当前的带宽、时延、抖动、丢包率中的一种或多种；所述预测网络属性信息包括网络未来的带宽、时延、抖动、丢包率中的一种或多种。The current network utility value of each candidate network is determined according to the current network attribute information corresponding to the M candidate networks, and the current network utility value of each candidate network is determined according to the predicted network attribute information corresponding to the M candidate networks. The predicted network utility value of the candidate network is obtained, and the network utility value of the M candidate networks is obtained; the current network attribute information includes one of the current network bandwidth, delay, jitter, and packet loss rate or The predicted network attribute information includes one or more of the future network bandwidth, delay, jitter, and packet loss rate.
  18. 如权利要求14所述的装置，其特征在于，所述第二处理单元，具体用于：The apparatus of claim 14, wherein the second processing unit is specifically configured to:
    接收所述提供M个候选网络关联的一个或多个网络设备分别发送的所述M个候选网络的所述网络效用值。The network utility values of the M candidate networks respectively sent by the one or more network devices providing the M candidate network associations are received.
  19. 如权利要求14-18任意一项所述的装置，其特征在于，所述第三处理单元，具体用于：The apparatus according to any one of claims 14-18, wherein the third processing unit is specifically configured to:
    基于所述当前业务权重和每个所述候选网络的所述当前网络效用值，以及基于所述预测业务权重和每个所述候选网络的所述预测网络效用值，计算每个所述网络的综合效用值。Calculate the network utility value for each of the candidate networks based on the current traffic weight and the current network utility value for each of the candidate networks, and based on the predicted traffic weight and the predicted network utility value for each of the candidate networks Comprehensive utility value.
  20. 如权利要求14-19任意一项所述的装置，其特征在于，所述装置还包括：The device according to any one of claims 14-19, wherein the device further comprises:
    第一获取单元，用于获取所述终端设备的第一历史记录信息；所述第一历史记录信息包括应用的访问记录、内容请求偏好、业务开始时间、业务流量、业务特征和使用时长中的一种或多种；The first obtaining unit is used to obtain the first history record information of the terminal device; the first history record information includes application access records, content request preferences, service start time, service flow, service characteristics and usage duration. one or more;
    第五处理单元，用于根据所述第一历史记录信息进行预测得到所述预测业务属性信息。A fifth processing unit, configured to perform prediction according to the first historical record information to obtain the predicted service attribute information.
  21. 如权利要求17所述的装置，其特征在于，所述第二处理单元，具体用于：The apparatus of claim 17, wherein the second processing unit is specifically configured to:
    根据所述当前业务属性信息得到所述当前业务的服务质量需求；所述服务质量需求包括当前所述终端设备所需要的带宽、时延、抖动和丢包率中的一种或多种；Obtain the service quality requirement of the current service according to the current service attribute information; the service quality requirement includes one or more of the bandwidth, delay, jitter and packet loss rate currently required by the terminal device;
    基于所述服务质量和所述用户偏好得到网络效用函数；obtaining a network utility function based on the quality of service and the user preference;
    根据每个所述候选网络的所述当前网络属性信息和所述网络效用函数得到所述当前网络效用值；obtaining the current network utility value according to the current network attribute information and the network utility function of each of the candidate networks;
    根据每个所述候选网络的所述预测网络属性信息和所述网络效用函数得到所述预测网络效用值。The predicted network utility value is obtained according to the predicted network attribute information and the network utility function of each of the candidate networks.
  22. 如权利要求21所述的装置，其特征在于，所述第二处理单元，具体用于：The apparatus of claim 21, wherein the second processing unit is specifically configured to:
    基于所述服务质量需求得到L个网络决策属性效用函数；obtaining L network decision attribute utility functions based on the service quality requirements;
    基于所述服务质量和所述用户偏好得到所述L个网络决策属性效用函数中的每个所述网络决策属性效用函数对应的网络决策属性权重；Obtaining a network decision attribute weight corresponding to each of the network decision attribute utility functions in the L network decision attribute utility functions based on the service quality and the user preference;
    根据所述L个网络决策属性效用函数及其对应的所述网络决策属性权重，确定所述网络效用函数。The network utility function is determined according to the L network decision attribute utility functions and their corresponding network decision attribute weights.
  23. 如权利要求14-22任意一项所述的装置，其特征在于，所述装置还包括：The device according to any one of claims 14-22, wherein the device further comprises:
    第六处理单元，用于测量N个候选网络的参考信号接收功率或参考信号接收质量；判断所述参考信号接收功率或参考信号接收质量是否满足预评估条件；将满足所述预评估条件的M个候选网络确定为所述M个候选网络。The sixth processing unit is used to measure the reference signal received power or the reference signal received quality of the N candidate networks; judge whether the reference signal received power or the reference signal received quality meets the pre-evaluation condition; candidate networks are determined as the M candidate networks.
  24. 一种智能选网的装置，其特征在于，应用于网络设备，所述网络设备提供S个候选网络；所述S个候选网络为所述终端设备的M个候选网络中的网络；所述装置包括：A device for intelligent network selection, characterized in that, it is applied to a network device, and the network device provides S candidate networks; the S candidate networks are networks among M candidate networks of the terminal device; the device include:
    第一获取单元，用于获取所述S个候选网络中的每个候选网络的第二历史记录信息；所述第二历史记录信息包括每个小区的历史带宽、时延、抖动、丢包率中的一种或多种；A first obtaining unit, configured to obtain second historical record information of each candidate network in the S candidate networks; the second historical record information includes historical bandwidth, time delay, jitter, and packet loss rate of each cell one or more of;
    第一处理单元，用于根据每个所述第二历史记录信息进行预测得到S个预测网络属性信息。The first processing unit is configured to perform prediction according to each of the second historical record information to obtain S pieces of predicted network attribute information.
  25. 如权利要求24所述的装置，其特征在于，所述装置还包括：The apparatus of claim 24, wherein the apparatus further comprises:
    第二处理单元，用于获取所述终端设备的当前业务属性信息和预测业务属性信息；根据所述当前业务属性信息确定所述当前业务权重，以及根据所述预测业务属性信息确定所述预测业务权重，得到所述终端设备的业务综合权重；所述当前业务属性信息包括当前业务类型、当前业务持续时间和用户偏好中的一种或多种，所述预测业务属性信息包括预测业务类型、预测业务持续时间和所述用户偏好中的一种或多种；a second processing unit, configured to acquire current service attribute information and predicted service attribute information of the terminal device; determine the current service weight according to the current service attribute information, and determine the predicted service according to the predicted service attribute information weight, to obtain the comprehensive service weight of the terminal device; the current service attribute information includes one or more of the current service type, current service duration and user preference, and the predicted service attribute information includes the predicted service type, predicted service one or more of business duration and said user preferences;
    第一发送单元，用于向所述终端设备发送所述业务综合权重。A first sending unit, configured to send the comprehensive service weight to the terminal device.
  26. 如权利要求24所述的装置，其特征在于，所述装置还包括：The apparatus of claim 24, wherein the apparatus further comprises:
    第三处理单元，用于根据每个所候选网络对应的所述当前网络属性信息确定所述当前网络效用值，以及根据每个所述候选网络对应的所述预测网络属性信息确定所述预测网络效用值；所述当前网络属性信息包括网络当前的带宽、时延、抖动、丢包率中的一种或多种；所述预测网络属性信息包括网络未来的带宽、时延、抖动、丢包率中的一种或多种；基于每个所述当前网络效用值和每个所述预测网络效用值得到S个网络效用值；所述网络效用值包括同一个所述候选网络的当前网络效应值和预测网络效用值；a third processing unit, configured to determine the current network utility value according to the current network attribute information corresponding to each candidate network, and determine the predicted network according to the predicted network attribute information corresponding to each candidate network Utility value; the current network attribute information includes one or more of the current network bandwidth, delay, jitter, and packet loss rate; the predicted network attribute information includes future network bandwidth, delay, jitter, packet loss one or more of the rates; S network utility values are obtained based on each of the current network utility values and each of the predicted network utility values; the network utility values include the current network effect of the same candidate network value and predicted network utility value;
    第二发送单元，用于向所述终端设备发送所述S个候选网络对应的所述S个网络效用值。A second sending unit, configured to send the S network utility values corresponding to the S candidate networks to the terminal device.
  27. 一种智能选网的方法，其特征在于，应用于网络设备，所述方法包括：A method for intelligent network selection, characterized in that, applied to network equipment, the method comprising:
    确定终端设备的业务综合权重；所述业务综合权重包括所述终端设备的当前业务权重和预测业务权重；所述当前业务权重为基于所述终端设备当前业务属性信息得到的权重，所述预测业务权重为基于所述终端设备预测业务属性信息得到的权重；Determine the service comprehensive weight of the terminal device; the service comprehensive weight includes the current service weight and the predicted service weight of the terminal device; the current service weight is a weight obtained based on the current service attribute information of the terminal device, and the predicted service weight The weight is a weight obtained based on the predicted service attribute information of the terminal device;
    分别确定所述终端设备的M个候选网络的网络效用值；每个所述网络效用值包括对应网络的当前网络效用值和预测网络效用值；所述当前网络效用值为基于当前网络属性信息得到的值，所述预测网络效用值为基于预测网络属性信息得到的值；The network utility values of the M candidate networks of the terminal device are respectively determined; each of the network utility values includes the current network utility value and the predicted network utility value of the corresponding network; the current network utility value is obtained based on the current network attribute information The value of the predicted network utility value is obtained based on the predicted network attribute information;
    基于所述业务综合权重和所述M个候选网络的所述网络效用值，得到M个综合效用值；Obtain M comprehensive utility values based on the service comprehensive weight and the network utility values of the M candidate networks;
    基于所述M个综合效用值，从所述M个候选网络中选取所述终端设备的目标网络。Based on the M comprehensive utility values, a target network of the terminal device is selected from the M candidate networks.
  28. 一种计算机存储介质，其特征在于，所述计算机存储介质存储有计算机程序，该计算机程序被处理器执行时实现上述权利要求1-10或11-13中任意一项所述的方法。A computer storage medium, characterized in that the computer storage medium stores a computer program, and when the computer program is executed by a processor, implements the method described in any one of the above claims 1-10 or 11-13.
  29. 一种计算机程序，其特征在于，所述计算机程序包括指令，当所述计算机程序被计算机执行时，使得所述计算机执行如权利要求1-10或11-13中任意一项所述的方法。A computer program, characterized in that the computer program includes instructions, which, when executed by a computer, cause the computer to perform the method according to any one of claims 1-10 or 11-13.
PCT/CN2022/083216 2021-04-02 2022-03-26 Intelligent network selection method and related apparatus WO2022206633A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202110365608.4A CN115190496A (en) 2021-04-02 2021-04-02 Intelligent network selection method and related device
CN202110365608.4 2021-04-02

Publications (1)

Publication Number Publication Date
WO2022206633A1 true WO2022206633A1 (en) 2022-10-06

Family

ID=83455563

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/083216 WO2022206633A1 (en) 2021-04-02 2022-03-26 Intelligent network selection method and related apparatus

Country Status (2)

Country Link
CN (1) CN115190496A (en)
WO (1) WO2022206633A1 (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117580132A (en) * 2024-01-16 2024-02-20 杭州海康威视数字技术股份有限公司 Heterogeneous network access method, device and equipment for mobile equipment based on reinforcement learning

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101860885A (en) * 2010-06-11 2010-10-13 西安电子科技大学 Access network selection method based on neural network and fuzzy logic
WO2017000158A1 (en) * 2015-06-30 2017-01-05 Orange Method for scheduling user equipment in a heterogeneous network
CN108419274A (en) * 2018-02-28 2018-08-17 华南理工大学 A kind of selecting method for isomeric wireless network based on utility function
CN110113803A (en) * 2019-03-20 2019-08-09 西安电子科技大学 Network insertion selection method based on user experience in a kind of heterogeneous network
CN110519783A (en) * 2019-09-26 2019-11-29 东华大学 5G network based on enhancing study is sliced resource allocation methods
CN110545568A (en) * 2019-10-15 2019-12-06 中国联合网络通信集团有限公司 Heterogeneous network switching method, switching device, control equipment and storage medium

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104185221A (en) * 2014-02-19 2014-12-03 上海物联网有限公司 Vertical network switching method based on self-adaptive weight adjustment
CN105813123B (en) * 2016-03-25 2019-10-25 广东顺德中山大学卡内基梅隆大学国际联合研究院 Heterogeneous network access selection method based on prediction call blocking rate prescreening network
US11184778B2 (en) * 2018-02-20 2021-11-23 Intel Corporation Mobile service chain placement
CN108901058A (en) * 2018-07-06 2018-11-27 北方工业大学 Internet of things node access channel optimization selection method

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101860885A (en) * 2010-06-11 2010-10-13 西安电子科技大学 Access network selection method based on neural network and fuzzy logic
WO2017000158A1 (en) * 2015-06-30 2017-01-05 Orange Method for scheduling user equipment in a heterogeneous network
CN108419274A (en) * 2018-02-28 2018-08-17 华南理工大学 A kind of selecting method for isomeric wireless network based on utility function
CN110113803A (en) * 2019-03-20 2019-08-09 西安电子科技大学 Network insertion selection method based on user experience in a kind of heterogeneous network
CN110519783A (en) * 2019-09-26 2019-11-29 东华大学 5G network based on enhancing study is sliced resource allocation methods
CN110545568A (en) * 2019-10-15 2019-12-06 中国联合网络通信集团有限公司 Heterogeneous network switching method, switching device, control equipment and storage medium

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117580132A (en) * 2024-01-16 2024-02-20 杭州海康威视数字技术股份有限公司 Heterogeneous network access method, device and equipment for mobile equipment based on reinforcement learning
CN117580132B (en) * 2024-01-16 2024-04-12 杭州海康威视数字技术股份有限公司 Heterogeneous network access method, device and equipment for mobile equipment based on reinforcement learning

Also Published As

Publication number Publication date
CN115190496A (en) 2022-10-14

Similar Documents

Publication Publication Date Title
Trestian et al. Game theory-based network selection: Solutions and challenges
US9380508B2 (en) System, apparatus and method for managing network device connectivity on heterogenous networks
US11997647B2 (en) Device contexts, operational modes, and policy driven enhancements for paging in advanced networks
Nguyen-Vuong et al. A user-centric and context-aware solution to interface management and access network selection in heterogeneous wireless environments
Zhu et al. Adaptive multiservice heterogeneous network selection scheme in mobile edge computing
Sousa et al. A survey on QoE-oriented wireless resources scheduling
CN107426781A (en) A kind of double frequency WiFi automatic switching methods, device and computer-readable recording medium
US11757748B2 (en) Policy determining method, system, and apparatus
WO2021077630A1 (en) Communication method, apparatus, device, and system
Ndashimye et al. A Multi-criteria based handover algorithm for vehicle-to-infrastructure communications
US20140056153A1 (en) Wireless communications system having selective wireless communications network and related methods
WO2022228267A1 (en) Communication processing method, communication apparatus and communication system
Wu et al. A terminal‐controlled vertical handover decision scheme in IEEE 802.21‐enabled heterogeneous wireless networks
Abbas et al. An optimized approach to video traffic splitting in heterogeneous wireless networks with energy and QoE considerations
WO2022206633A1 (en) Intelligent network selection method and related apparatus
Louta et al. Towards realization of the ABC vision: A comparative survey of Access Network Selection
KR20090065328A (en) Handover method and apparatus in mobile network
CN107484236A (en) Wi-Fi hotspot scan control method, mobile terminal and storage medium
US20240089849A1 (en) Network selection policy determining method and apparatus
US10582508B2 (en) Facilitation of network resources
Mazza et al. A user-satisfaction based offloading technique for smart city applications
WO2012073059A1 (en) Methods, apparatuses and computer program products for using downlink and uplink over separate radio channels or different operators
Trestian User-centric power-friendly quality-based network selection strategy for heterogeneous wireless environments
Raschellà et al. A Centralized Win‐Win Cooperative Framework for Wi‐Fi and 5G Radio Access Networks
Singhal et al. eM‐SON: efficient multimedia service over self‐organising Wi‐Fi network

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22778815

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 22778815

Country of ref document: EP

Kind code of ref document: A1