WO2021238277A1 - 网络优化方法、服务器、网络侧设备、***和存储介质 - Google Patents

网络优化方法、服务器、网络侧设备、***和存储介质 Download PDF

Info

Publication number
WO2021238277A1
WO2021238277A1 PCT/CN2021/073544 CN2021073544W WO2021238277A1 WO 2021238277 A1 WO2021238277 A1 WO 2021238277A1 CN 2021073544 W CN2021073544 W CN 2021073544W WO 2021238277 A1 WO2021238277 A1 WO 2021238277A1
Authority
WO
WIPO (PCT)
Prior art keywords
network
side device
network side
measurement
server
Prior art date
Application number
PCT/CN2021/073544
Other languages
English (en)
French (fr)
Inventor
刘壮
高音
陈嘉君
Original Assignee
中兴通讯股份有限公司
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 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Priority to US17/923,098 priority Critical patent/US20230180025A1/en
Priority to MX2022014678A priority patent/MX2022014678A/es
Priority to EP21812406.3A priority patent/EP4161128A4/en
Priority to AU2021279510A priority patent/AU2021279510B2/en
Publication of WO2021238277A1 publication Critical patent/WO2021238277A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W24/00—Supervisory, monitoring or testing arrangements
    • H04W24/02—Arrangements for optimising operational condition
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/16—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks using machine learning or artificial intelligence
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W24/00—Supervisory, monitoring or testing arrangements
    • H04W24/10—Scheduling measurement reports ; Arrangements for measurement reports
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W24/00—Supervisory, monitoring or testing arrangements
    • H04W24/08—Testing, supervising or monitoring using real traffic

Definitions

  • This application relates to the field of communication technology, for example, to a network optimization method, server, network side device, system, and storage medium.
  • the 5th Generation Wireless Systems Network (The 5th Generation Wireless Systems Network, 5G) is being deployed and is expected to surpass 5G (Beyond 5G, B5G) networks in the future.
  • 5G Beyond 5G, B5G
  • AI artificial intelligence
  • This application provides a network optimization method, server, network side equipment, system, and storage medium.
  • the embodiment of the present application provides a network optimization method, including:
  • the terminal equipment of the network side equipment performs measurement configuration; receives the measurement report of the specified network side equipment and the measurement report of the terminal equipment; based on the measurement data and strategy information in the received measurement report, performs machine learning optimized for the network to obtain the network Optimization operation instructions, network optimization operation instructions are used to instruct designated network side devices and terminal devices to perform network optimization according to the network optimization operation instructions.
  • the embodiment of the application also provides a network optimization method, including:
  • the embodiment of the present application also provides a server, including:
  • the measurement control information determining module is used to determine the network side measurement control information according to the pre-obtained policy information that the network needs to meet; the measurement configuration request module is used to send a session establishment request to the specified network side device to request the specified network side
  • the equipment performs measurement configuration according to the network side measurement control information and performs measurement configuration on the terminal equipment connected to the designated network side equipment; the measurement report receiving module is used to receive the measurement report of the designated network side equipment and the measurement report of the terminal equipment;
  • the learning module is used to perform machine learning for network optimization based on the measurement data and strategy information in the received measurement report to obtain network optimization operation instructions.
  • the network optimization operation instructions are used to instruct designated network side equipment and terminal equipment to optimize according to the network Operating instructions for network optimization.
  • the embodiment of the present application also provides a network side device, including:
  • the measurement configuration module is used to perform measurement configuration according to the network side measurement control information in the session establishment request and to perform measurement configuration on the terminal device connected to the current network side device in response to receiving a session establishment request from a predetermined server; sending a measurement report
  • the module is used to send the measurement report of the current network side device and the received measurement report of the terminal device to the predetermined server.
  • the measurement report of the current network side device and the measurement report of the terminal device are used in the predetermined server. For machine learning optimized for the network.
  • An embodiment of the present application also provides a server, including: one or more processors; a memory, on which one or more programs are stored, and when one or more programs are executed by one or more processors, one or more Each processor implements the above-mentioned first network optimization method in the embodiment of the present application.
  • An embodiment of the present application also provides a network side device, including: one or more processors; a memory, on which one or more programs are stored, and when one or more programs are executed by one or more processors, one Or multiple processors implement the above-mentioned second network optimization method in the embodiment of the present application.
  • An embodiment of the present application also provides a network optimization system, including: a server, which is configured to execute the above-mentioned first network optimization method; and one or more network-side devices, which are configured to execute the above-mentioned second network optimization method.
  • the embodiment of the present application further provides a storage medium, and the storage medium stores a computer program.
  • the computer program is executed by a processor, any network optimization method in the embodiment of the present application is implemented.
  • server, network-side device, system, and storage medium in the embodiments of the present application, it is possible to determine network-side measurement control information based on pre-obtained policy information that needs to be met by the network, and request the network-side device to measure according to the network-side Control information for measurement configuration and measurement configuration for terminal side measurement, and use the received wireless network side and terminal side measurement measurement data for machine learning model training to obtain network optimization operations, so as to use artificial intelligence and machine in the network system Learning to conduct in-depth analysis of the collected data provides a new optimization method and intelligent network optimization process for operators' network optimization.
  • measurement configuration can be performed according to the received network-side measurement control information and the measurement configuration of the connected terminal device can be performed, and the measurement can be obtained by executing the measurement.
  • the measurement report and the received measurement report of the terminal device are sent to the predetermined server.
  • the measurement report of the network device and the measurement report of the terminal device are used in the AI server to perform machine learning optimized for the network, so that the network device and the terminal device
  • the collected data can be deeply analyzed in artificial intelligence and machine learning, providing new optimization methods and intelligent network optimization processes for operators' network optimization.
  • FIG. 1 shows a schematic flowchart of a network optimization method according to an embodiment of the present application
  • Figure 2 shows a flowchart of a network optimization method according to another embodiment of the present application
  • FIG. 3 shows a schematic diagram of a flow sequence of a network optimization method according to an embodiment
  • FIG. 4 shows a schematic diagram of a flow sequence of a network optimization method according to another embodiment
  • FIG. 5 shows a schematic flowchart of establishing a communication interface between a server and a network-side device according to an embodiment
  • FIG. 6 shows a schematic flowchart of establishing a data channel between a predetermined server and a user equipment according to an embodiment
  • FIG. 7 shows a schematic structural diagram of a server provided by an embodiment of the present application.
  • FIG. 8 shows a schematic structural diagram of a network side device provided by an embodiment of the present application.
  • FIG. 9 shows a schematic structural diagram of a network optimization system provided by an embodiment of the present application.
  • FIG. 10 shows a structural diagram of a hardware architecture of a computing device provided by an embodiment of the present application.
  • 5G network brings fundamental changes in performance and flexibility, the complexity of network operations is also increasing. Operators urgently need automation and intelligent means to reduce network deployment and operation and maintenance costs, and improve networks Performance and user experience.
  • Long Term Evolution (LTE) and Self-Organized Network (SON) technology and Minimization of Drive Tests (MDT) in 5G have a great effect on network optimization, but they have not been realized
  • 5G networks face the challenge of joint optimization of many key performance indicators (Key Performance Indicators, KPIs). These indicators include, for example, latency, reliability, connection density, user experience, and so on.
  • KPIs Key Performance Indicators
  • Radio Access Network Radio Access Network
  • the embodiments of the present application provide a network optimization solution that uses artificial intelligence/machine learning in-depth analysis of collected data to provide a new network optimization method for operators' network optimization, and realizes the support of AI-based network intelligent optimization processes in the network architecture.
  • Fig. 1 shows a schematic flowchart of a network optimization method according to an embodiment of the present application.
  • the network optimization method in the embodiment of the present application can be applied to an AI server.
  • the network optimization method includes the following steps.
  • S110 Determine network side measurement control information according to pre-acquired policy information that needs to be met by the network.
  • S120 Send a session establishment request to the designated network side device to request the designated network side device to perform measurement configuration according to the network side measurement control information and to perform measurement configuration on the terminal device connected to the specified network side device.
  • S130 Receive a measurement report of a designated network side device and a measurement report of a terminal device.
  • S140 Based on the measurement data and strategy information in the received measurement report, perform machine learning for network optimization to obtain network optimization operation instructions.
  • the network optimization operation instructions are used to instruct designated network side devices and terminal devices to perform according to the network optimization operation instructions Network Optimization.
  • the AI server can perform measurement configuration on the wireless network side and terminal side measurement according to the policy information that the network needs to meet, and receive the measurement data measured by the wireless network side and the terminal side, and will receive the measurement
  • the data is used for machine learning model training to obtain network optimization operations, so that in-depth analysis of the collected data through artificial intelligence and machine learning in the network system provides new optimization methods for operators' network optimization and can support AI-based networks Intelligent optimization process.
  • the network side device may be, for example, a RAN node device in a 5G network or a beyond 5G (B5G) network.
  • the RAN node device includes, but is not limited to, a base station.
  • the method before step S110, the method further includes: S01, obtaining policy information that the pre-configured network needs to meet; or, S02, in response to the received activation message, activating the machine learning function, and obtaining the activation message The carried network needs to meet the policy information.
  • the policy information that the network needs to meet can be directly configured in the AI server in advance, or it can be obtained from the activation message received from the network operation and maintenance (Operation Administration and Maintenance, OAM) system or the core network.
  • OAM Operaation Administration and Maintenance
  • the strategy information that needs to be satisfied is more flexible and can be set according to actual needs.
  • the received activation message may include an indication indicating whether the machine learning function is activated, and the machine learning function is activated or deactivated according to the indication. After the machine learning function is activated, the policy information that the network needs to satisfy is obtained in the message. (Policy Information).
  • the policy information may include object identification information and communication quality index information.
  • step S110 may include the following steps S111 and S112.
  • S111 According to the entity identified by the object identification information, determine the network side device that needs to perform the measurement as the designated network side device.
  • the policy information that the network needs to satisfy includes one or more of the following object identification information: one or more terminal equipment identities (User Equipment Identity, UE ID), one or more service quality (Quality of Service, QoS) Flow Identity (QFI) identifier, one or more cell identifiers, one or more network slice identifiers, one or more public land mobile network identifiers (Public Land Mobile Network Identity, PLMN ID), one or Multiple private network identities, one or more base station identities, and one or more tracking area code identity (TAC ID).
  • object identification information one or more terminal equipment identities (User Equipment Identity, UE ID), one or more service quality (Quality of Service, QoS) Flow Identity (QFI) identifier, one or more cell identifiers, one or more network slice identifiers, one or more public land mobile network identifiers (Public Land Mobile Network Identity, PLMN ID), one or Multiple private network identities, one or more base station identities, and one or more tracking area code identity (TAC ID).
  • the network side device that needs to perform the measurement may be the network side device connected to the identified terminal device; if the object identification information identifies
  • the entity of is at least one of one or more quality of service flows, one or more cells, one or more network slices, one or more public land mobile networks, and one or more private networks, you need to perform measurement
  • the network-side device at may be a network-side device involved in the communication range of the identified entity.
  • the network side device that needs to perform the measurement can also be flexibly selected according to actual machine learning requirements, which is not limited in the embodiment of the present application.
  • S112 According to the communication quality required by the entity indicated by the communication quality index information, determine the measurement quantity and the measurement report mode that the specified network side device needs to configure, as the network side measurement control information.
  • the communication quality index information may be used to indicate the communication quality required by the entity identified by the corresponding object identification information.
  • the communication quality index information may include one or more of the following index items: network energy saving index, network KPI, service quality (Quality of Service) index, user experience quality (Quality of Experience, QoE) index, Business perception key quality indicators (Key Quality Indicators, KQI) and business traffic control preferences (Traffic Steering Preferences) indicators.
  • the network energy saving index may include, for example, one or more of energy saving efficiency, energy saving percentage index, and energy saving value.
  • network KPIs may include, for example, one or more of handover success rate, dropped call rate, access success rate, user throughput rate, cell throughput rate, cell load, network load, radio resource utilization rate, and network coverage rate. item.
  • the service quality index may include, for example, one or more of: a guaranteed service rate, a service maximum/minimum rate, a service delay, a service priority, a delay jitter, and a packet loss rate.
  • the user experience quality index may include: Mean Opinion Score (MOS) for measuring the voice quality of the communication system, one of the streaming media opening and caching time, the streaming media re-caching time, the number of streaming media re-caching times, or Multiple.
  • MOS Mean Opinion Score
  • the AI server can instruct the network side device to perform measurement configuration and configure the terminal device according to the pre-acquired policy information that the network needs to meet.
  • the network operation, management and maintenance system may be referred to as a network management system for short.
  • the core network may be, for example, the 5th Generation Wireless Systems Core Network (5GC), or communication beyond 5G.
  • the core network of the network may be, for example, the 5th Generation Wireless Systems Core Network (5GC), or communication beyond 5G.
  • the current deployment location of the server includes outside the network side device and inside the network side device; S120 may include the following steps.
  • S122 If the deployment location of the current server is inside the network side device, perform measurement configuration on the network side device where the current server is located according to the network side measurement control information, and send a session establishment request message to the designated network side device, the session establishment request message Including network side measurement control information.
  • the network side device that deploys the AI server also needs to perform measurement configuration according to the network side measurement control information.
  • the deployment location of the AI server is external, the AI server can send a machine learning session setup request (Machine Learning Session Setup Request) message or a session setup request message to the network side device to instruct the network side device to perform according to the network side measurement control information Measurement configuration and measurement configuration of the terminal equipment connected to the designated network side equipment.
  • the session establishment request carries the machine learning session identifier and network-side measurement control information, where each machine learning session identifier is used to uniquely identify a machine learning process, and the network-side measurement control information is used for the unique identification
  • the machine learning process indicates the amount of measurement to be collected and the method of measurement reporting.
  • the session establishment request carries network-side measurement control information, where the carried network-side measurement control information is used to indicate the measurement amount that needs to be collected and the measurement report method for all machine learning processes.
  • the session establishment request carries network-side measurement control information.
  • a machine learning session ID (Machine Learning Session ID) may also be carried.
  • the machine learning session identifier can be used to uniquely identify a machine learning process. If the session establishment request does not include the machine learning session identifier, it indicates that the measurement control information in the session establishment request is for all machine learning (ML) processes.
  • the machine learning process may be used to indicate the machine learning corresponding to the network-side measurement control information carried in the session establishment request.
  • the machine learning process can be determined according to different optimization indicators, for example, different optimization indicators are set for different communication quality indicators, and different optimization indicators correspond to different machine learning processes; or, the machine learning process can be determined by machine learning types
  • the type of machine learning includes but is not limited to any of supervised learning, unsupervised learning, reinforcement learning, deep learning, and transfer learning; or, the machine learning process can be determined by a machine learning model, which includes but It is not limited to: any one of convolutional neural network, cyclic neural network, long and short-term memory network, support vector machine, autoregressive moving average model, and decision tree.
  • the method further includes: in response to the received session establishment response message, determining that both the designated network side device and the terminal device are successfully configured for measurement.
  • the network side device may send a session establishment response message to the AI server, where the session establishment response message carries the successful session establishment Indication; or, it can carry an indication of successful session establishment and a machine learning session ID (ML session ID).
  • ML session ID machine learning session ID
  • the network side device if the measurement configuration on the network side device fails or the measurement configuration on the terminal device fails, the network side device sends a machine learning session establishment response message to the AI server, where the message carries an indication of the session establishment failure.
  • the session establishment response message can carry the session identifier, and the AI server resends the session establishment request after receiving the session establishment failure indication.
  • the session establishment request, the session establishment response message, and the received measurement report include a corresponding machine learning session identifier, and the machine learning session identifier is used to uniquely identify the machine learning process.
  • step S130 may include the following steps.
  • S131 Receive the measurement report of the designated network side device and the measurement report of the terminal device sent by the designated network side device.
  • the AI server is not directly connected to the terminal device. After the terminal device reports the measurement report to the network-side device, the network-side device sends the measurement report of the terminal device to the AI server, and the AI server can communicate with the network-side device through The communication interface between the terminal equipment to obtain the measurement report.
  • the measurement data in the received measurement report includes the collected measurement amount and the machine learning session identifier, where the machine learning session identifier is used to uniquely identify the machine learning process.
  • the measurement report message sent by the network side device to the AI server and the measurement report message sent by the terminal device may carry the measurement quantity; or, the measurement quantity and the session identifier may be carried at the same time.
  • the method further includes: sending a network operation request message to a designated network side device, and the network operation request message includes a network optimization operation instruction and corresponding operation parameters.
  • the AI server can notify the network side device to perform related optimization operations (Action) through a network operation request (Action Request) message, where the optimization operation action can be one or more operation instructions and corresponding operation requirements.
  • RRM Radio Resource Management
  • the current server communicates with the designated network side device through a predetermined interface.
  • the method further includes the following steps.
  • the communication interface between the AI server and the network-side device can be determined in advance according to the current deployment location of the server and before the measurement configuration is performed on the network-side device and the terminal device.
  • the AI server can respond to the interface establishment request of the network-side device to establish a control plane interface with the network-side device; and when the deployment location of the AI server is in the designated network Inside the side device, the existing communication transmission interface between the network side devices can be directly used for communication, without the need to establish an additional communication interface between the AI server and the network side device, saving network resources.
  • the interface between the network side devices As an example, 5G network-side devices (such as base stations) can be connected through a core network, and the base station and the core network can rely on optical fiber transmission for communication. As an example, the 5G network side device may also be connected through wired connection, wireless connection, wireless relay, etc., which is not limited in the embodiment of the present application.
  • control plane interface establishment request message includes one or more of the following information items: measurement supported by the specified network-side device, reporting mode supported by the specified network-side device, and specified network-side device Supported network optimization operations and the data plane channel address of the designated network side device.
  • the AI server when it sends a session establishment request to a specified network-side measurement device, it can obtain network-side measurement control information according to the measurement supported by the specified network-side device and the reporting method supported by the specified network-side device , Making the measurement configuration of the network-side equipment more targeted.
  • the measurement of the network-side equipment is configured according to the supported measurement and the reporting mode supported by the specified network-side equipment, which can improve the processing efficiency and accuracy of the measurement configuration.
  • the AI server sends a network operation request message to a designated network side device.
  • the network operation request message may include network optimization operation instructions and corresponding operation parameters that can be supported by the network side measurement device, thereby improving the designated network optimization operation. The efficiency and accuracy of instructions.
  • step S11 the method further includes:
  • S21 Send a control plane interface establishment response message to the designated network side device to indicate that the control plane interface is successfully established; S22, if the control plane interface establishment request message includes the data plane channel address of the designated network side device, then the control plane is sent When the interface establishment response message is sent to the designated network side device, the control plane interface establishment response message carries the data plane channel address of the current server.
  • the interface establishment response message may carry the data plane channel address of the AI server side for establishing and specifying The data plane channel between the network side devices.
  • the AI server and the network-side device can transmit control-type messages required for machine learning on the control plane channel.
  • the control plane data includes, for example, network-side measurement configuration information, sending session establishment requests, and receiving Session establishment response, sending network operation request, receiving network operation response, etc., can also be used to receive measurement data of network side equipment, measurement data of terminal equipment, etc.
  • the data plane channel established in this embodiment is not limited to transmitting measurement data from designated terminal devices.
  • the AI server can configure and receive the measurement data on the wireless network side and the measurement data on the terminal side, perform machine learning model training, and obtain network optimization operations, thereby adopting artificial intelligence and machine learning
  • In-depth analysis of the collected data and the use of intelligent network optimization methods provide operators with new optimization methods for network optimization and can support AI-based network intelligent optimization processes.
  • Fig. 2 shows a flowchart of a network optimization method according to another embodiment of the present application.
  • the network optimization method is applied to a network side device.
  • the network optimization method may include the following steps.
  • S210 in response to receiving a session establishment request from a predetermined server, perform measurement configuration according to the network side measurement control information in the session establishment request and perform measurement configuration on the terminal device connected to the current network side device.
  • S220 Send the current measurement report of the network side device and the received measurement report of the terminal device to a predetermined server, and the measurement report of the current network side device and the measurement report of the terminal device are used in the predetermined server.
  • Machine learning optimized for the network is used in the predetermined server.
  • the network side device can perform measurement configuration according to the received network side measurement control information and perform measurement configuration on the connected terminal device, and perform the measurement to obtain the measurement report and the received terminal side device’s information
  • the measurement report is sent to the predetermined server.
  • the measurement report of the network-side equipment and the measurement report of the terminal equipment are used in the AI server for network-optimized machine learning, so that the data collected by the network-side equipment and terminal equipment can be used in artificial intelligence and In-depth analysis of machine learning provides new optimization methods and intelligent network optimization processes for operators' network optimization.
  • S210 may include the following steps.
  • S211 In response to the session establishment request, perform measurement configuration according to the network-side measurement control information.
  • the network-side measurement control information is used to indicate the measurement quantity and measurement reporting mode that the network-side device needs to configure at present;
  • S212 determine according to the network-side measurement control information
  • the measurement volume and measurement report mode that the terminal device connected to the current network side device needs to configure are used as the terminal side measurement control information;
  • S213, send a first radio resource control message to the terminal device to instruct the terminal device to perform according to the terminal side measurement control information Measurement configuration.
  • the network side device may perform network side measurement configuration according to the network side measurement control information carried in the session establishment request from the AI server, and instruct the connected terminal device to perform terminal side measurement configuration.
  • a session establishment response message is sent to the predetermined server to feed back to the predetermined server that both the current network-side device and the terminal device measure The configuration is successful.
  • the received session establishment request, the session establishment response message, and the measurement report sent to the predetermined server include the corresponding machine learning session identifier, and the machine learning session identifier is used to uniquely identify the machine learning process.
  • the network-side optimization method further includes: S240, receiving and executing a network optimization operation instruction from a predetermined server; S241, if the network optimization operation instruction involves a terminal device, determining that the network optimization operation instruction is related to the terminal device Operation; S242, sending a second radio resource control message to the terminal device to instruct the terminal device to perform related operations.
  • the network side device executes the received network optimization operation instruction. If the network optimization operation instruction involves a terminal device, the network optimization operation instruction is sent to the terminal device involved, thereby performing optimization on the network side and the terminal side Operation to realize network optimization according to network optimization operation instructions.
  • the current network side device communicates with a predetermined server through a predetermined interface, and before step S210, the method may further include the following steps.
  • S31 If the deployment location of the predetermined server is outside the network side device, send a control plane interface establishment request message to the predetermined server according to the address of the predetermined server acquired in advance to request the predetermined server to establish the current network side device and the predetermined server.
  • the control plane interface is used as a predetermined interface.
  • control plane interface establishment request message includes one or more of the following information items: measurements supported by the current network-side device, reporting methods supported by the current network-side device, and supported by the current network-side device Network optimization operation and the data plane channel address of the current network side device.
  • control plane interface establishment request message can be used to request a predetermined server to send network-side measurements that the current network-side device can support according to the measurement supported by the current network-side device and the reporting method supported by the current network-side device
  • the configuration information can improve the success rate and processing efficiency of the network side measurement configuration.
  • control plane interface establishment request message can also be used to request a predetermined server, according to the network optimization operation supported by the current network side device, to send the network optimization operation instruction that the current network side device can support, so as to respond to the received
  • the network optimization operation instructions that can be supported perform network optimization, which improves the data processing performance of the network side measurement and execution of the network optimization operation instructions.
  • the network optimization method further includes: S250, in response to the received control plane interface establishment response message, determining that the control plane interface between the current network side device and the predetermined server is successfully established; wherein, if the control plane interface is established
  • the request message includes the data plane channel address of the current network side device, and the received control plane interface establishment response message includes the data plane channel address of the predetermined server.
  • the network-side device may send an interface establishment request according to the received address of the AI server to request the establishment of a control plane interface between the network-side device and the AI server.
  • the interface establishment request may carry the following One or more items: measurement supported by the network side device; reporting method of the side quantity supported by the network side device; RAN optimization operation supported by the network side device; data plane channel address of the network side device.
  • the interface establishment request carries the data plane channel address of the network side device
  • the data plane interface between the local network side device and the AI server is established.
  • the control plane interface and/or the data plane interface as needed to perform data transmission between the local network side device and the AI server.
  • the network side device can perform measurement configuration on the network side device and the terminal device under the control of the AI server, and can perform the measurement data obtained by the network side device and the received terminal device
  • the measurement data obtained by the measurement is sent to the AI server for use in machine learning for network optimization on the AI server, so as to receive and execute the network optimization operation instructions obtained by the machine learning to perform network optimization.
  • FIG. 3 shows a schematic diagram of a flow sequence of a network optimization method according to an embodiment.
  • the AI server is arranged outside the network side device, and the network optimization method may include the following steps.
  • S301 The network management system or the core network sends an activation message through the interface with the AI server.
  • the OAM or 5GC sends an activation message to instruct the AI server to activate or use the machine learning function.
  • the message contains an indication indicating whether the ML function is activated; the message contains policy information indicating that the RAN network side needs to meet.
  • the AI server sends a machine learning session establishment request message to the RAN node device for configuring wireless side measurement data required for a machine learning session.
  • the machine learning session establishment request message may include a machine learning session ID to uniquely identify an ML process ID, and the measurement control information is used to indicate the measurement quantity to be measured by the RAN node device and the measurement report method. If the session ID is not included in the machine learning session establishment request message, it indicates that the measurement control in the request message is for all ML processes.
  • the RAN node device configures the measurement and reporting mode that the RAN node device needs to perform according to the measurement control information in the received message.
  • the RAN node device configures the measurement control information on the terminal side according to the measurement control information in the received message, and transmits the measurement control information on the UE side through a radio resource control (Radio Resource Control, RRC) establishment/reconfiguration message, which carries the measurement control information on the UE side.
  • RRC Radio Resource Control
  • One or more UEs on the local RAN node equipment instruct the UE to perform measurements and how to report.
  • the RAN node device sends a machine learning session establishment response message to the AI server, where the message carries a success indication, and the message optionally carries a machine learning session ID (MLsession ID). ID). If the RAN node device measurement or UE side measurement configuration fails, the RAN node device sends a machine learning session establishment response message to the AI server, where the message carries a failure indication, and the message optionally carries a machine learning session ID (ML session ID) .
  • MLsession ID machine learning session ID
  • both the RAN node device measurement and the UE side measurement are configured successfully, both the RAN node device and the UE perform related measurements according to the specified measurement configuration.
  • S307-01 The RAN node device sends a measurement report to the AI server.
  • S307-02 The RAN node device sends the received measurement report from the terminal.
  • the measurement report message may carry the measurement amount, or may carry the measurement amount and the machine learning session ID to which the measurement belongs.
  • the AI server selects a suitable ML algorithm for model training, model prediction and model update according to the measurement data and the configured policy information, and obtains the RAN operation required for optimization.
  • the AI server sends a RAN operation request message to notify the RAN node to perform related optimization operations.
  • S310 The RAN node device performs related optimization operations in the RAN operation request message.
  • the RAN node device sends an RRC reconfiguration message or an RRC release message to the UE connected to the RAN node device to notify the UE to perform related operations.
  • the network optimization method in this embodiment After the machine learning function of the AI server is activated, measurement configuration is performed on the RAN node device, and the measurement configuration of the terminal device connected to the RAN node device is performed through the RAN node device.
  • the RAN node device and The terminal device performs the measurement and sends the measurement report to the AI server; the AI server selects the appropriate ML algorithm for model training, model prediction and model update based on the measurement data and the configured strategy information, and obtains and sends the RAN operation required for optimization , So as to optimize the network by optimizing the required RAN operations.
  • FIG. 4 shows a schematic diagram of a flow sequence of a network optimization method according to another embodiment.
  • the AI server is arranged inside the network side device, and the network optimization method may include the following steps.
  • S401 The network management system or the core network sends an activation message through the interface with the AI server.
  • the AI server configures the measurements required by the RAN node device 1.
  • the RAN node device 1 configures measurement control information on the terminal device side.
  • the RAN node device 1 carries terminal device side measurement control information through an RRC establishment/reconfiguration message, and sends it to one or more terminal devices connected to the RAN node device 1, instructing the terminal device to perform which measurement and how Reported.
  • the AI server sends a machine learning session establishment request message to the RAN node device 2 for configuring wireless side measurement data required for a machine learning session.
  • the RAN node device 2 configures the measurement quantity that the RAN node device 2 needs to measure and the measurement report mode according to the measurement control information in the received message.
  • the RAN node device 2 configures the measurement control information on the terminal device side according to the measurement control information in the received message.
  • the RAN node device 2 can use the RRC establishment/reconfiguration message to carry terminal device side measurement control information, and send it to one or more terminal devices connected to the base station to instruct the terminal device to perform measurements and how to report it. .
  • the base station sends a machine learning session establishment response message to the RAN node device 1.
  • the message carries a success indication
  • the message optionally carries a machine learning session ID (machine learning session ID).
  • the RAN node device 2 sends a machine learning session establishment response message to the RAN node device 1, where the message carries a failure indication, and the message optionally carries a machine learning session ID (machine learning session ID).
  • both the RAN node device 2 side measurement and the terminal device side measurement are configured successfully, both the RAN node device 2 and the terminal device perform related measurements according to the specified measurement configuration. At the same time, the RAN node device 1 also performs related measurements according to the measurements configured by the AI server.
  • S409 The RAN node device 1 receives the measurement report.
  • the measurement report may be directly sent by the terminal device connected to the RAN node device 1 to the RAN node device 1 (corresponding to step S409-02), and the measurement report may be the measurement made on the RAN node device 2 and sent to the RAN
  • the measurement report may be sent by the terminal device connected to RAN node device 2 to RAN node device 2 and forwarded by RAN node device 2 to RAN node device 1 (corresponding to step S409-03 ).
  • the AI server selects a suitable machine learning algorithm to perform model training, model prediction and model update according to the measurement data and the configured policy information, and obtains the RAN operation required for optimization.
  • S411 The AI server notifies the RAN node device 1 node to perform related optimization operations.
  • the RAN node device 1 performs related optimization operations specified by the AI server.
  • the RAN node device 1 sends an RRC reconfiguration message or an RRC release message to the terminal device connected to the base station to notify the terminal device to perform the related operation.
  • the RAN node device 1 sends a RAN operation request message to notify the RAN node device 2 to perform related optimization operations.
  • the optimization operation can be one or more operation instructions and parameters required for corresponding operations, such as terminal device handover instructions, close/open cell instructions, radio resource activation/deactivation instructions, power adjustment instructions, RRM parameter reconfiguration instructions, Shunt operation instructions, protocol layer parameter reconfiguration instructions, etc.).
  • S414 The RAN node device 2 performs related optimization operations in the RAN operation request message.
  • the RAN node device 2 sends an RRC reconfiguration message or an RRC release message to the UE connected to the RAN node device 2 to notify the UE to perform related operations.
  • the RAN node device 2 sends an RRC reconfiguration message or an RRC release message to the terminal device connected to the base station to notify the terminal device to perform the related operation.
  • the network optimization method in Figure 4 is basically the same as the network optimization method in Figure 3.
  • the RAN node device that deploys the AI server also needs to perform measurement configuration and perform measurement; when the AI is deployed
  • the RAN node device where the AI server is located can send an RRC reconfiguration message or an RRC release message to the terminal connected to the RAN node Equipment to instruct the terminal equipment connected to the RAN node to perform the specified optimization operation.
  • the base station where the AI server is deployed can send an ML session establishment request and receive an ML session response message through an interface with another base station to request neighboring base stations to participate in RAN optimization operations, without the need to establish additional AI
  • the communication interface between the server and the network side device saves network resources.
  • Fig. 5 shows a schematic flow chart of establishing a communication interface between an AI server and a RAN node device in an embodiment.
  • the interface establishment process may include the following steps.
  • step S501 if the AI server is deployed inside the RAN node, there is no need to perform step S501.
  • the base station sends a communication interface establishment request according to the configured address of the AI server.
  • the RAN node sends an AI interface establishment request message to the address of the AI server to establish the AI interface.
  • the communication interface establishment request may include: measurement supported by the network side device (for example, a base station); the reporting method of the side quantity supported by the network side device; and the RAN optimization operation supported by the network side device.
  • the network side device for example, a base station
  • the communication interface establishment request may further include: the data plane channel address of the network side device.
  • the AI server sends an interface establishment response message to the base station, which is used to indicate whether the interface establishment is successful.
  • the interface establishment response message may carry the data plane channel address of the AI server side.
  • control plane interface of the AI server for the RAN node can be established.
  • the control plane interface can be used to transmit control messages, and can also be used to transmit data required for machine learning, such as measurement data.
  • S504 Establish a data plane channel according to the data plane channel address configured by the AI server and the data plane channel address on the base station side.
  • the data required for machine learning can be transmitted between the network-side device and the AI server on the data plane channel.
  • some relatively large amount of data can be transmitted on the data plane channel, such as measurement data from the base station.
  • the data plane channel established in this embodiment is not limited to transmitting data of a specified user.
  • Fig. 6 shows a schematic flow chart of establishing a data channel between an AI server and a user equipment according to an embodiment.
  • the interface establishment process may include the following steps.
  • S601 Acquire an established A1 control plane interface between the AI server and the RAN node device (for example, a base station).
  • the AI server can be deployed outside the RAN node, or the AI server can be deployed inside a RAN node.
  • the AI server or the base station where the AI server is located sends a channel establishment request message to the neighboring base station.
  • the channel establishment request message may be, for example, a terminal device context setup request message (UE Context Setup Request), and carries the data plane channel address and the terminal device identifier (UE ID) on the AI server side.
  • UE Context Setup Request a terminal device context setup request message
  • UE ID terminal device identifier
  • the base station sends a channel establishment response message to the AI server or the base station where the AI server is located.
  • the channel establishment response message may be, for example, a terminal device context setup response message (UE Context Setup Response), and carries the data plane channel address on the base station side.
  • UE Context Setup Response UE Context Setup Response
  • S604 According to the data plane channel address configured by the AI server and the data plane channel address on the base station side, establish a data plane channel designated by the UE ID and related to a specific user.
  • the data required by ML related to a specific user can be transmitted between the RAN node and the AI server on the data plane channel.
  • some larger data volume data can be transmitted on the data plane channel, such as the user's measurement data.
  • the data required for machine learning can be transmitted between the RAN node device and the AI server on the data plane channel.
  • some relatively large amount of data can be transmitted on the data plane channel, such as measurement data from the base station.
  • the data plane channel established in this embodiment is not limited to transmitting data of a specified user.
  • FIG. 7 shows a schematic structural diagram of a server provided by an embodiment of the present application.
  • the server may be a server with AI functions, referred to as an AI server, as shown in FIG. 7, the server may include the following modules.
  • the measurement control information determining module 710 is configured to determine the network side measurement control information according to the pre-acquired policy information that the network needs to meet.
  • the measurement configuration request module 720 is used to send a session establishment request to a specified network side device to request the specified network side device to perform measurement configuration according to the network side measurement control information and to perform measurement configuration on the terminal device connected to the specified network side device .
  • the measurement report receiving module 730 receives the measurement report of the designated network side device and the measurement report of the terminal device.
  • the machine learning module 740 is used to perform machine learning for network optimization based on the measurement data and strategy information in the received measurement report to obtain network optimization operation instructions.
  • the network optimization operation instructions are used to instruct designated network side devices and terminal devices according to Network optimization operation instructions for network optimization.
  • the server further includes: a policy information obtaining module, which is used to obtain policy information that the pre-configured network needs to meet; or, in response to the received activation message, activate the machine learning function, and obtain the information carried in the activation message Policy information that the network needs to satisfy.
  • a policy information obtaining module which is used to obtain policy information that the pre-configured network needs to meet; or, in response to the received activation message, activate the machine learning function, and obtain the information carried in the activation message Policy information that the network needs to satisfy.
  • the policy information includes object identification information and communication quality index information.
  • the communication quality index information is used to indicate the communication quality required by the entity identified by the object identification information; in this embodiment, the measurement control information determining module 710 can be used to determine the network side device that needs to perform measurement according to the entity identified by the object identification information as the designated network side device; and, according to the communication quality that the entity indicated by the communication quality index information needs to achieve, determine the designated network side device.
  • the measurement quantity and measurement report mode that the network side device needs to configure are used as the network side measurement control information.
  • the deployment location of the current server includes outside the network-side device and inside the network-side device; in this embodiment, the measurement configuration request module 720 can be used to: if the deployment location of the current server is outside the network-side device , The session establishment request message is sent to the designated network side device, and the session establishment request message includes the network side measurement control information; if the current server deployment location is inside the network side device, the network side measurement control information is used to determine where the current server is located.
  • the network side device performs measurement configuration and sends a session establishment request message to the designated network side device.
  • the session establishment request message includes network side measurement control information.
  • the server may further include: a session establishment response module, configured to determine that both the specified network side device and the terminal device are successfully configured for measurement in response to the received session establishment response message.
  • a session establishment response module configured to determine that both the specified network side device and the terminal device are successfully configured for measurement in response to the received session establishment response message.
  • the session establishment request, the session establishment response message, and the received measurement report include a corresponding machine learning session identifier, and the machine learning session identifier is used to uniquely identify the machine learning process.
  • the measurement report receiving module 730 may be used to receive the measurement report of the designated network side device and the measurement report of the terminal device sent by the designated network side device.
  • the server may further include: a network operation request sending module, configured to send a network operation request message to a designated network side device, and the network operation request message includes a network optimization operation instruction and corresponding operation parameters.
  • a network operation request sending module configured to send a network operation request message to a designated network side device, and the network operation request message includes a network optimization operation instruction and corresponding operation parameters.
  • the current server communicates with a designated network-side device through a predetermined interface
  • the server further includes: a first predetermined interface determining module, configured to respond if the current server is deployed at a location outside the network-side device Based on the received control plane interface establishment request message, the control plane interface between the current server and the designated network side device is established as a predetermined interface; the second predetermined interface determination module is used if the current server is deployed in the network side device Internally, the existing communication transmission interface between the network side device where the current server is located and the designated network side device is obtained as a predetermined interface.
  • control plane interface establishment request message includes one or more of the following information items: measurement supported by the specified network-side device, report mode supported by the specified network-side device, and specified network-side device Supported network optimization operations and the data plane channel address of the designated network side device.
  • the server further includes: a control plane interface establishment response sending module, configured to send a control plane interface establishment response message to a designated network side device to indicate that the control plane interface is successfully established; if the control plane interface establishment request message The data plane channel address of the designated network side device is included in the specified network side device, and when the control plane interface establishment response message is sent to the designated network side device, the data plane channel address of the current server is carried in the control plane interface establishment response message.
  • a control plane interface establishment response sending module configured to send a control plane interface establishment response message to a designated network side device to indicate that the control plane interface is successfully established
  • the server uses the configured and received measurement data on the wireless network side and the measurement data on the terminal side to perform machine learning model training to obtain network optimization operations, thereby performing artificial intelligence and machine learning on the collected data
  • machine learning model training to obtain network optimization operations, thereby performing artificial intelligence and machine learning on the collected data
  • intelligent network optimization methods provide operators with new optimization methods and intelligent network optimization procedures for network optimization.
  • FIG. 8 shows a schematic structural diagram of a network side device provided by an embodiment of the present application.
  • the network side device may include the following modules.
  • the measurement configuration module 810 is configured to, in response to receiving a session establishment request from a predetermined server, perform measurement configuration according to the network side measurement control information in the session establishment request and perform measurement configuration on the terminal device connected to the current network side device; measurement report The sending module 820 is used to send the measured measurement report of the current network side device and the received measurement report of the terminal device to a predetermined server, and the current measurement report of the network side device and the measurement report of the terminal device are in the predetermined server, It is used for machine learning optimized for the network.
  • the measurement configuration module 810 may be used to: in response to a session establishment request, perform measurement configuration according to network-side measurement control information.
  • the network-side measurement control information is used to indicate the current network-side device needs to configure measurement quantities and measurements.
  • Reporting method according to the network side measurement control information, determine the measurement quantity and measurement report mode that the terminal device connected to the current network side device needs to configure, as the terminal side measurement control information; send the first radio resource control message to the terminal device to indicate The terminal device performs measurement configuration according to the measurement control information on the terminal side.
  • the network side device may further include: a session establishment response message sending module, configured to send a session establishment response to a predetermined server if the current network side device measurement configuration is successful and the measurement configuration response message of the terminal device is received Message to feed back to the predetermined server that both the current network-side device and the terminal device have successfully measured and configured.
  • a session establishment response message sending module configured to send a session establishment response to a predetermined server if the current network side device measurement configuration is successful and the measurement configuration response message of the terminal device is received Message to feed back to the predetermined server that both the current network-side device and the terminal device have successfully measured and configured.
  • the received session establishment request, the session establishment response message, and the measurement report sent to the predetermined server include the corresponding machine learning session identifier, and the machine learning session identifier is used to uniquely identify the machine learning process.
  • the network-side device may further include: an operation instruction receiving module for receiving and executing network optimization operation instructions from a predetermined server; an operation instruction execution unit for determining if the network optimization operation instruction involves a terminal device The operation related to the terminal device in the network optimization operation instruction; the related operation sending unit is used to send the second radio resource control message to the terminal device to instruct the terminal device to perform the related operation.
  • the current network-side device communicates with a predetermined server through a predetermined interface
  • the network-side device may further include: a predetermined interface establishment request module, configured to: if the deployment location of the predetermined server is outside the current network-side device, According to the address of the predetermined server acquired in advance, a control plane interface establishment request message is sent to the predetermined server to request the predetermined server to establish the control plane interface between the current network side device and the predetermined server as the predetermined interface; the communication transmission interface acquisition module is used If the deployment location of the predetermined server is inside the current network side device, then the existing communication transmission interface between the current network side device and the network side device where the predetermined server is located is acquired as the predetermined interface.
  • a predetermined interface establishment request module configured to: if the deployment location of the predetermined server is outside the current network-side device, According to the address of the predetermined server acquired in advance, a control plane interface establishment request message is sent to the predetermined server to request the predetermined server to establish the control plane interface between the current network side device
  • control plane interface establishment request message includes one or more of the following information items: measurements supported by the current network-side device, reporting methods supported by the current network-side device, and supported by the current network-side device Network optimization operation and the data plane channel address of the current network side device.
  • the network side device may further include: a control plane interface establishment response receiving module, configured to determine the current control plane interface establishment between the network side device and the predetermined server in response to the received control plane interface establishment response message Success; wherein, if the control plane interface establishment request message includes the data plane channel address of the current network side device, the received control plane interface establishment response message includes the data plane channel address of the predetermined server.
  • a control plane interface establishment response receiving module configured to determine the current control plane interface establishment between the network side device and the predetermined server in response to the received control plane interface establishment response message Success; wherein, if the control plane interface establishment request message includes the data plane channel address of the current network side device, the received control plane interface establishment response message includes the data plane channel address of the predetermined server.
  • the network-side device and the terminal device under the control of the AI server, the network-side device and the terminal device can be configured for measurement, and the measurement data obtained by the measurement performed by the network-side device and the received terminal device can be measured.
  • the data is sent to the AI server to perform machine learning for network optimization on the AI server, so as to receive and execute network optimization operation instructions obtained by the machine learning to perform network optimization.
  • FIG. 9 shows a schematic structural diagram of a network optimization system provided by an embodiment of the present application.
  • the network optimization system may include the following server 910 and one or more network side devices 920.
  • a server 910 which may be used to execute the network optimization method described with reference to FIG. 1 in the foregoing embodiment.
  • One or more network-side devices 920, and the network-side device 920 is configured to execute the network optimization method described in conjunction with FIG. 2 in the foregoing embodiment.
  • the AI server 910 and the AI server described in conjunction with FIG. 7 have the same or equivalent structure, and can execute the network optimization method applied to the AI server described in the foregoing embodiment;
  • the network side device 920 is described in conjunction with FIG. 8
  • the network-side devices of have the same or equivalent structure, and can execute the network optimization method applied to the network-side device described in the foregoing embodiments.
  • FIG. 10 shows a structural diagram of a hardware architecture of a computing device provided by an embodiment of the present application.
  • the computing device 1000 includes an input device 1001, an input interface 1002, a central processing unit 1003, a memory 1004, an output interface 1005, and an output device 1006.
  • the input interface 1002, the central processing unit 1003, the memory 1004, and the output interface 1005 are connected to each other through the bus 1010, and the input device 1001 and the output device 1006 are connected to the bus 1010 through the input interface 1002 and the output interface 1005, respectively, and then connected to the computing device 1000
  • the other components are connected.
  • the input device 1001 receives input information from the outside, and transmits the input information to the central processing unit 1003 through the input interface 1002; the central processing unit 1003 processes the input information based on the computer executable instructions stored in the memory 1004 to generate output information.
  • the output information is temporarily or permanently stored in the memory 1004, and then the output information is transmitted to the output device 1006 through the output interface 1005; the output device 1006 outputs the output information to the outside of the computing device 1000 for use by the user.
  • the computing device shown in FIG. 10 may be implemented as a server.
  • the server may include: a memory configured to store a program; a processor configured to run a program stored in the memory to execute the foregoing The network optimization method applied to the AI server described in the embodiment.
  • the computing device shown in FIG. 10 may be implemented as a network-side device, and the network-side device may include: a memory configured to store a program; a processor configured to run a program stored in the memory , To implement the network optimization method applied to the network side device described in the above embodiment.
  • the various embodiments of the present application can be implemented in hardware or dedicated circuits, software, logic or any combination thereof.
  • some aspects may be implemented in hardware, while other aspects may be implemented in firmware or software that may be executed by a controller, microprocessor, or other computing device, although the application is not limited thereto.
  • the embodiments of the present application may be implemented by executing computer program instructions by a data processor of a mobile device, for example, in a processor entity, or by hardware, or by a combination of software and hardware.
  • Computer program instructions can be assembly instructions, instruction set architecture (Instruction Set Architecture, ISA) instructions, machine instructions, machine-related instructions, microcode, firmware instructions, state setting data, or written in any combination of one or more programming languages Source code or object code.
  • the block diagram of any logic flow in the drawings of the present application may represent program steps, or may represent interconnected logic circuits, modules, and functions, or may represent a combination of program steps and logic circuits, modules, and functions.
  • the computer program can be stored on the memory.
  • the memory can be of any type suitable for the local technical environment and can be implemented using any suitable data storage technology, such as but not limited to read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), optical Memory devices and systems (Digital Video Disc (DVD) or Compact Disk (CD)), etc.
  • Computer-readable media may include non-transitory storage media.
  • the data processor can be any type suitable for the local technical environment, such as but not limited to general-purpose computers, special-purpose computers, microprocessors, digital signal processors (Digital Signal Processing, DSP), application specific integrated circuits (ASICs) ), programmable logic devices (Field-Programmable Gate Array, FPGA), and processors based on multi-core processor architecture.
  • DSP Digital Signal Processing
  • ASICs application specific integrated circuits
  • FPGA Field-Programmable Gate Array
  • FPGA Field-Programmable Gate Array

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Software Systems (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Computation (AREA)
  • Medical Informatics (AREA)
  • Theoretical Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Physics & Mathematics (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Data Mining & Analysis (AREA)
  • Mobile Radio Communication Systems (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)
  • Transition And Organic Metals Composition Catalysts For Addition Polymerization (AREA)

Abstract

本文公开一种网络优化方法、服务器、网络侧设备、***和存储介质。该网络优化方法，应用于服务器，包括：根据预先获取的网络需要满足的策略信息，确定网络侧测量控制信息；发送会话建立请求至指定的网络侧设备，以请求指定的网络侧设备根据网络侧测量控制信息对指定的网络侧设备进行测量配置和对连接于指定的网络侧设备的终端设备进行测量配置；接收指定的网络侧设备的测量报告和终端设备的测量报告；基于所接收测量报告中的测量数据和策略信息，进行针对网络优化的机器学习，得到网络优化操作指令，网络优化操作指令用于指示指定的网络侧设备和终端设备根据网络优化操作指令进行网络优化。

Description

网络优化方法、服务器、网络侧设备、***和存储介质
本申请要求在2020年05月24日提交中国专利局、申请号为202010445463.4的中国专利申请的优先权，该申请的全部内容通过引用结合在本申请中。
技术领域
本申请涉及通信技术领域，例如涉及一种网络优化方法、服务器、网络侧设备、***和存储介质。
背景技术
第五代无线***网络(The 5th Generation Wireless Systems Network，5G)正在部署，并且有望在未来发展超过5G(Beyond 5G，B5G)网络。
在对网络进行优化时，还没有如何对通信网络配置人工智能(ArtificialIntelligence，AI)功能，从而基于AI进行网络智能优化流程的网络智能化方案。
发明内容
本申请提供一种网络优化方法、服务器、网络侧设备、***和存储介质。
本申请实施例提供一种网络优化方法，包括：
根据预先获取的网络需要满足的策略信息，确定网络侧测量控制信息；发送会话建立请求至指定的网络侧设备，以请求指定的网络侧设备根据网络侧测量控制信息进行测量配置和对连接于指定的网络侧设备的终端设备进行测量配置；接收指定的网络侧设备的测量报告和终端设备的测量报告；基于所接收测量报告中的测量数据和策略信息，进行针对网络优化的机器学习，得到网络优化操作指令，网络优化操作指令用于指示指定的网络侧设备和终端设备根据网络优化操作指令进行网络优化。
本申请实施例还提供一种网络优化方法，包括：
响应于接收到来自预定服务器的会话建立请求，根据会话建立请求中的网络侧测量控制信息进行测量配置和对连接于当前网络侧设备的终端设备进行测量配置；将测量得到的当前网络侧设备的测量报告和接收到的终端设备的测量报告，发送至预定服务器，当前网络侧设备的测量报告和终端设备的测量报告在预定服务器中，被用于进行针对网络优化的机器学习。
本申请实施例还提供一种服务器，包括：
测量控制信息确定模块，用于根据预先获取的网络需要满足的策略信息，确定网络侧测量控制信息；测量配置请求模块，用于发送会话建立请求至指定的网络侧设备，以请求指定的网络侧设备根据网络侧测量控制信息进行测量配置和对连接于指定的网络侧设备的终端设备进行测量配置；测量报告接收模块，用于接收指定的网络侧设备的测量报告和终端设备的测量报告；机器学习模块，用于基于所接收测量报告中的测量数据和策略信息，进行针对网络优化的机器学习，得到网络优化操作指令，网络优化操作指令用于指示指定的网络侧设备和终端设备根据网络优化操作指令进行网络优化。
本申请实施例还提供一种网络侧设备，包括：
测量配置模块，用于响应于接收到来自预定服务器的会话建立请求，根据会话建立请求中的网络侧测量控制信息进行测量配置和对连接于当前网络侧设备的终端设备进行测量配置；测量报告发送模块，用于将测量的得到当前网络侧设备的测量报告和接收到的终端设备的测量报告，发送至预定服务器，当前网络侧设备的测量报告和终端设备的测量报告在预定服务器中，被用于进行针对网络优化的机器学习。
本申请实施例还提供一种服务器，包括：一个或多个处理器；存储器，其上存储有一个或多个程序，当一个或多个程序被一个或多个处理器执行，使得一个或多个处理器实现本申请实施例中的上述第一种网络优化方法。
本申请实施例还提供一种网络侧设备，包括：一个或多个处理器；存储器，其上存储有一个或多个程序，当一个或多个程序被一个或多个处理器执行，使得一个或多个处理器实现本申请实施例中的上述第二种网络优化方法。
本申请实施例还提供一种网络优化***，包括：服务器，服务器用于执行上述第一种网络优化方法；一个或多个网络侧设备，网络侧设备用于执行上述第二种网络优化方法。
本申请实施例还提供一种存储介质，存储介质存储有计算机程序，计算机程序被处理器执行时实现本申请实施例中的任意的网络优化方法。
根据本申请实施例中的网络优化方法、服务器、网络侧设备、***和存储介质，可以基于预先获取的网络需要满足的策略信息，确定网络侧测量控制信息，并请求网络侧设备根据网络侧测量控制信息进行测量配置和对终端侧测量进行测量配置，并根据接收的无线网络侧和终端侧测量的测量数据用于机器学习模型训练，得到网络优化操作，从而在网络***中通过人工智能和机器学习对采集的数据进行深度分析，为运营商网络优化提供了新的优化方式和网络智 能优化流程。
根据本申请实施例中的网络优化方法、服务器、网络侧设备、***和存储介质，可以根据接收到的网络侧测量控制信息进行测量配置并对连接的终端设备进行测量配置，并将执行测量得到测量报告和接收的终端侧设备的测量报告至预定服务器，网络侧设备的测量报告和终端设备的测量报告在AI服务器中被用于进行针对网络优化的机器学习，使得网络侧设备和终端设备设备所采集的数据能够在人工智能和机器学习中被深度分析，为运营商网络优化提供新的优化方式和网络智能优化流程。
附图说明
图1示出本申请一实施例的网络优化方法的流程示意图；
图2示出本申请另一实施例的网络优化方法的流程图；
图3示出一实施例的网络优化方法的流程时序示意图；
图4示出另一实施例的网络优化方法的流程时序示意图；
图5示出一个实施例的服务器与网络侧设备之间建立通信接口的流程示意图；
图6示出一个实施例的建立预定服务器与用户设备相关的数据通道的流程示意图；
图7示出本申请一实施例提供的一种服务器的结构示意图；
图8示出本申请一实施例提供的一种网络侧设备的结构示意图；
图9示出本申请一实施例提供的网络优化***的结构示意图；
图10示出本申请实施例提供的一种计算设备的硬件架构的结构图。
具体实施方式
下文中将结合附图对本申请的实施例进行说明。
在本申请实施例中，5G网络在性能和灵活性上带来根本性改变的同时，网络运营的复杂度也在提高，运营商急需自动化和智能化手段降低网络部署和运维成本，提升网络性能和用户体验。长期演进(Long Term Evolution，LTE)和5G中的自组织网络(Self-Organized Network，SON)技术和最小化驱动测试(Minimization of Drive Tests，MDT)对网络优化有很大作用，但并没有实现灵活的智能化网络的目标，5G网络面临着诸多网络关键业绩指标(Key Performance Indicator，KPI)的联合优化问题的挑战。这些指标例如包括时延，可靠性，连 接数密度，用户体验等。
网络优化方法日渐呈现出反应周期长，易出错，成本高等问题。无线接入网(Radio Access Network，RAN)节点设备和终端设备中的测量量，一方面可以用于网管***监测网络KPI，也可以协助网络厂商优化无线资源管理，但是AI功能所在的网元如何收集这些测量量也没有解决方案。本申请实施例提供一种网络优化方案，通过人工智能/机器学习深度分析采集的数据，为运营商网络优化提供了新的网络优化方法，实现在网络架构中支持基于AI的网络智能优化流程。
图1示出本申请一实施例的网络优化方法的流程示意图。在一个实施例中，本申请实施例中的网络优化方法可以应用于AI服务器，如图1所示，该网络优化方法包括以下步骤。
S110，根据预先获取的网络需要满足的策略信息，确定网络侧测量控制信息。
S120，发送会话建立请求至指定的网络侧设备，以请求指定的网络侧设备根据网络侧测量控制信息进行测量配置和对连接于指定的网络侧设备的终端设备进行测量配置。
S130，接收指定的网络侧设备的测量报告和终端设备的测量报告。
S140，基于所接收测量报告中的测量数据和策略信息，进行针对网络优化的机器学习，得到网络优化操作指令，网络优化操作指令用于指示指定的网络侧设备和终端设备根据网络优化操作指令进行网络优化。
根据本申请实施例的网络优化方法，AI服务器可以根据网络需要满足的策略信息，对无线网络侧和终端侧测量进行测量配置，并接收无线网络侧和终端侧测量的测量数据，将接收的测量数据用于机器学习模型训练，得到网络优化操作，从而在网络***中通过人工智能和机器学习对采集的数据进行深度分析，为运营商网络优化提供了新的优化方式和可以支持基于AI的网络智能优化流程。
在本申请实施例的描述中，网络侧设备例如可以是5G网络或超过5G(B5G)网络中的RAN节点设备。作为示例，RAN节点设备包括但不限于是基站。
在一个实施例中，在步骤S110之前，方法还包括：S01，获取预先配置的网络需要满足的策略信息；或者，S02，响应于接收到的激活消息，激活机器学习功能，并获取激活消息中携带的网络需要满足的策略信息。
在该实施例中，网络需要满足的策略信息可以预先直接配置在AI服务器， 也可以从接收到的来自网络操作管理维护(Operation Administration and Maintenance，OAM)***或者核心网的激活消息中获取，网络需要满足的策略信息较为灵活，可以根据实际需要进行设定。
示例性地，接收到的激活消息中可以包含指示机器学习功能是否激活的指示，根据该指示激活或关闭机器学习功能，在机器学习功能激活后，获取该消息中携带的网络需要满足的策略信息(Policy Information)。
在一个实施例中，策略信息可以包含对象标识信息和通信质量指标信息。
在该实施例中，上述步骤S110，可以包括以下步骤S111和S112。
S111，根据对象标识信息所标识的实体，确定需要执行测量的网络侧设备，作为指定的网络侧设备。
在一个实施例中，网络需要满足的策略信息包含以下一项或者多项对象标识信息：一个或者多个终端设备标识(User Equipment Identity，UE ID)，一个或者多个服务质量(Quality of Service，QoS)流标识(QoS Flow Identity，QFI)标识，一个或者多个小区标识，一个或者多个网络切片标识，一个或者多个公共陆地移动网络标识(Public Land Mobile Network Identity，PLMN ID)，一个或者多个私有网络标识，一个或者多个基站标识，以及一个或者多个跟踪区域编码标识(Tracking Area Code Identity，TAC ID)。
在该实施例中，若对象标识信息所标识的实体为一个或者多个终端设备，则需要执行测量的网络侧设备可以是所标识的终端设备所连接的网络侧设备；若对象标识信息所标识的实体为一个或者多个服务质量流、一个或者多个小区、一个或者多个网络切片、一个或者多个公共陆地移动网络，以及一个或者多个私有网络中的至少一种，则需要执行测量的网络侧设备可以是所标识的实体通信范围内所涉及的网络侧设备。
在该实施例中，需要执行测量的网络侧设备也可以根据实际机器学习的要求灵活选定，本申请实施例不做限定。
S112，根据通信质量指标信息所指示的实体所需达到的通信质量，确定指定的网络侧设备需要配置的测量量和测量上报方式，作为网络侧测量控制信息。
在一个实施例中，通信质量指标信息，可以用以指示对应对象标识信息所标识的实体所需达到的通信质量。示例性地，通信质量指标信息可以包括以下指标项中的一项或者多项：网络节能指标、网络KPI、业务服务质量(Quality of Service)指标、用户体验质量(Quality of Experience，QoE)指标、业务感知关键质量指标(Key Quality Indicators，KQI)和业务流量调控首选项(Traffic Steering Preferences)指标。
示例性地，网络节能指标例如可以包括：节能效率，节省能耗百分比指标，节省能耗值中的一项或多项。
作为示例，网络KPI例如可以包括：切换成功率，掉话率，接入成功率，用户吞吐率，小区吞吐率，小区负荷，网络负荷，无线资源利用率，网络覆盖率中的一项或多项。
作为示例，业务服务质量指标例如可以包括：业务保证速率，业务最大/最小速率，业务时延，业务优先级，时延抖动，丢包率中的一项或多项。
作为示例，用户体验质量指标例如可以包括：衡量通信***语音质量的平均意见值(Mean Opinion Score，MOS)，流媒体打开缓存时间，流媒体重新缓存时间，流媒体重新缓存次数中的一项或多项。
通过上述步骤S111-S112，AI服务器可以根据预先获取的网络需要满足的策略信息，指示网络侧设备进行测量配置和对终端设备进行配置。
在本申请施例的描述中，网络操作管理维护***可以简称为网管***，核心网例如可以是第五代移动通信技术核心网(The 5th Generation Wireless Systems Core Network，5GC)，或超过5G的通信网络的核心网。
在一个实施例中，当前服务器的部署位置包括在网络侧设备外部和在网络侧设备内部；S120可以包括以下步骤。
S121，若当前服务器的部署位置在网络侧设备外部，则发送会话建立请求消息至指定的网络侧设备，会话建立请求消息中包括网络侧测量控制信息。
S122，若当前服务器的部署位置在网络侧设备内部，则根据网络侧测量控制信息对当前服务器所在的网络侧设备进行测量配置，并发送会话建立请求消息至指定的网络侧设备，会话建立请求消息中包括网络侧测量控制信息。
在该实施例中，若AI服务器的部署位置在网络侧设备内部，则部署AI服务器的网络侧设备，也需要根据网络侧测量控制信息进行测量配置。而AI服务器的部署位置在外部，则AI服务器可以通过发送机器学习会话建立请求(Machine Learning Session Setup Request)消息或会话建立请求消息至网络侧设备，以指示网络侧设备根据网络侧测量控制信息进行测量配置和对连接于指定的网络侧设备的终端设备进行测量配置。
在一个实施例中，会话建立请求中携带机器学习会话标识和网络侧测量控制信息，其中，每个机器学习会话标识用于唯一标识一个机器学习进程，网络侧测量控制信息用于针对唯一标识的机器学习进程，指示需要采集的测量量和测量上报方式。
在另一个实施例中，会话建立请求中携带网络侧测量控制信息，其中，所携带的网络侧测量控制信息，用于针对全部机器学习进程，指示需要采集的测量量和测量上报方式。
也就是说，会话建立请求中携带网络侧测量控制信息。可选地，也可以携带机器学习会话标识(Machine Learning Session ID)。其中，机器学习会话标识可用来唯一标志一个机器学习进程，如果会话建立请求中不包含机器学习会话标识，表明会话建立请求中的测量控制信息是针对所有的机器学习(Machine Learning，ML)进程。
在本申请实施例中，机器学***均模型、决策树中的任一种。
在一个实施例中，在上述步骤S120之后，方法还包括：响应于接收到的会话建立响应消息，确定指定的网络侧设备和终端设备均测量配置成功。
在该实施例中，如果对网络侧设备的测量配置和对终端设备的测量配置都配置成功，则网络侧设备可以向AI服务器发送会话建立响应消息，其中，会话建立响应消息中携带会话建立成功指示；或者，可以携带会话建立成功指示和机器学习会话标识(ML session ID)。
在一个实施例中，如果对网络侧设备的测量配置失败或者对终端设备的测量配置失败，则网络侧设备向AI服务器发送机器学习会话建立响应消息，其中，消息中携带会话建立失败指示。会话建立响应消息中可以携带会话标识，AI服务器在接收到会话建立失败指示后重新发送会话建立请求。
在一个实施例中，会话建立请求、会话建立响应消息和所接收的测量报告中，包括对应的机器学习会话标识，机器学习会话标识用于唯一标识机器学习进程。
在一个实施例中，步骤S130可以包括如下步骤。
S131，接收指定的网络侧设备的测量报告和由指定的网络侧设备发送的终端设备的测量报告。
在该实施例中，AI服务器并不直接与终端设备连接，终端设备将测量报告 上报到网络侧设备后，网络侧设备将终端设备的测量报告发送至AI服务器，AI服务器可以通过与网络侧设备之间的通讯接口，得到终端设备的测量报告。
在一个实施例中，所接收测量报告中的测量数据，包含所采集的测量量和机器学习会话标识，其中，机器学习会话标识用于唯一标识机器学习进程。
在该实施例中，网络侧设备向AI服务器发送的测量报告消息，以及终端设备发送的测量报告消息中可以携带测量量；或者，可以同时携带测量量和会话标识。
在一个实施例中，在步骤S140之后，方法还包括：发送网络操作请求消息至指定的网络侧设备，网络操作请求消息中包含网络优化操作指令和对应的操作参数。
在该实施例中，AI服务器可以通过网络操作请求(Action Request)消息，通知网络侧设备执行相关优化操作(Action)，其中，优化操作action可以是一项或者多项操作指示和对应操作所需的参数，例如包括但不限于是：UE切换指示，关闭/打开小区指示，无线资源激活/去激活指示，功率调整指示，无线资源管理(Radio Resource Management，RRM)参数重配置指示，分流操作指示，协议层参数重配置指示等。
在一个实施例中，当前服务器是通过预定接口与指定的网络侧设备进行通讯的，在步骤S120之前，方法还包括如下步骤。
S11，若当前服务器的部署位置在网络侧设备外部，则响应于接收到的控制面接口建立请求消息，建立当前服务器与指定的网络侧设备之间的控制面接口，作为预定接口。
S12，若当前服务器的部署位置在网络侧设备内部，则获取当前服务器所在的网络侧设备与指定的网络侧设备之间已有的通讯传输接口，作为预定接口。
在该实施例中，可以根据当前服务器的部署位置，在对网络侧设备和终端设备进行测量配置之前，预先确定AI服务器与网络侧设备之间的通讯接口。
当AI服务器的部署位置在指定的网络侧设备外部，AI服务器可以响应于网络侧设备的接口建立请求，建立与网络侧设备之间的控制面接口；而当AI服务器的部署位置在指定的网络侧设备内部，则可以直接使用网络侧设备之间已有的通讯传输接口进行通讯，无需额外建立AI服务器与网络侧设备之间的通讯接口，节省网络资源。该网络侧设备之间的接口。作为示例，5G网络侧设备(例如基站)之间可以通过核心网进行连接，基站和核心网之间可以依靠光纤传输来进行通讯。作为示例，5G网络侧设备还可以通过有线连接、无线连接、无线中继等方式连接，在本申请实施例中不做限定。
在一个实施例中，控制面接口建立请求消息中包括如下信息项的一项或多项：指定的网络侧设备所支持的测量、指定的网络侧设备所支持的上报方式、指定的网络侧设备所支持的网络优化操作和指定的网络侧设备的数据面通道地址。
在一个实施例中，AI服务器向指定的网络侧测量设备发送会话建立请求时，可以根据指定的网络侧设备所支持的测量和指定的网络侧设备所支持的上报方式，得到网络侧测量控制信息，使得对网络侧设备的测量配置更有针对性，网络侧设备的测量根据所支持的测量和指定的网络侧设备所支持的上报方式进行测量配置，能够提高测量配置的处理效率和准确度。
在一个实施例中，AI服务器发送网络操作请求消息至指定的网络侧设备，网络操作请求消息中可以包含网络侧测量设备能够支持的网络优化操作指令和对应的操作参数，从而提高指定网络优化操作指令的效率和准确率。
在一个实施例中，在步骤S11之后，方法还包括：
S21，发送控制面接口建立响应消息至指定的网络侧设备，以指示控制面接口建立成功；S22，若控制面接口建立请求消息中包括指定的网络侧设备的数据面通道地址，则在发送控制面接口建立响应消息至指定的网络侧设备时，在控制面接口建立响应消息中携带当前服务器的数据面通道地址。
在该实施例中，如果AI服务器接收的接口建立请求中携带了网络侧设备的数据面通道地址，则可以在接口建立响应消息中携带AI服务器侧的数据面通道地址，以用于建立与指定的网络侧设备之间的数据面通道。
在本申请实施例中，AI服务器和网络侧设备之间可以在控制面通道上传输机器学习所需的控制类消息，该控制层面数据例如包括：网络侧测量配置信息、发送会话建立请求、接收会话建立响应、发送网络操作请求、接收网络操作响应等，也可以用于接收网络侧设备的测量数据、终端设备的测量数据等。
为了提高数据传输效率，以及在终端设备的测量数据的数据量较大时，不影响AI服务器和网络侧设备之间控制面通道中对控制类消息的传输，可以在AI服务器和网络侧设备之间，建立数据面通道，使用数据面通道上传输机器学习所需数据。
较大数据量的数据可以在数据面通道上传输，比如基站的测量数据。但是，本实施例中建立的数据面通道不限于传输来自指定终端设备的测量数据。
根据本申请实施例提供的网络优化方法，AI服务器可以配置并接收的无线网络侧的测量数据和终端侧测量的测量数据，进行机器学习模型训练，得到网络优化操作，从而通过人工智能和机器学习对采集的数据进行深度分析，使用 智能化的网络优化方法，为运营商网络优化提供了新的优化方式和可以支持基于AI的网络智能优化流程。
图2示出本申请另一实施例的网络优化方法的流程图。在一个实施例中，该网络优化方法应用于网络侧设备，如图2所示，该网络优化方法可以包括如下步骤。
S210，响应于接收到来自预定服务器的会话建立请求，根据会话建立请求中的网络侧测量控制信息进行测量配置和对连接于当前网络侧设备的终端设备进行测量配置。
S220，将测量得到的当前网络侧设备的测量报告和接收到的终端设备的测量报告，发送至预定服务器，当前网络侧设备的测量报告和终端设备的测量报告在预定服务器中，被用于进行针对网络优化的机器学习。
根据本申请实施例的网络优化方法，网络侧设备可以根据接收到的网络侧测量控制信息进行测量配置并对连接的终端设备进行测量配置，并将执行测量得到测量报告和接收的终端侧设备的测量报告至预定服务器，网络侧设备的测量报告和终端设备的测量报告在AI服务器中被用于进行针对网络优化的机器学习，使得网络侧设备和终端设备设备所采集的数据能够在人工智能和机器学习中被深度分析，为运营商网络优化提供新的优化方式和网络智能优化流程。
在一个实施例中，S210可以包括如下步骤。
S211，响应于会话建立请求，根据网络侧测量控制信息进行测量配置，网络侧测量控制信息用于指示当前网络侧设备需要配置的测量量和测量上报方式；S212，根据网络侧测量控制信息，确定连接于当前网络侧设备的终端设备需要配置的测量量和测量上报方式，作为终端侧测量控制信息；S213，向终端设备发送第一无线资源控制消息，以指示终端设备根据终端侧测量控制信息进行测量配置。
在该实施例中，网络侧设备可以根据来自AI服务器的会话建立请求携带的网络侧测量控制信息进行网络侧的测量配置，并指示所连接的终端设备进行终端侧的测量配置。
在一个实施例中，若当前网络侧设备测量配置成功，且接收到终端设备的测量配置响应消息，则向预定服务器发送会话建立响应消息，以向预定服务器反馈当前网络侧设备和终端设备均测量配置成功。
在一个实施例中，接收到的会话建立请求、会话建立响应消息、以及发送至预定服务器的测量报告中包括对应的机器学习会话标识，机器学习会话标识 用于唯一标识机器学习进程。
在一个实施例中，网络侧优化方法还包括：S240，接收并执行来自预定服务器的网络优化操作指令；S241，若网络优化操作指令涉及终端设备，则确定网络优化操作指令中与终端设备相关的操作；S242，发送第二无线资源控制消息至终端设备，以指示终端设备执行相关的操作。
通过步骤S240至S242，网络侧设备执行接收到的网络优化操作指令，若网络优化操作指令涉及终端设备，则将网络优化操作指令发送至所涉及的终端设备，从而在网络侧和终端侧执行优化操作，实现根据网络优化操作指令进行网络优化。
在一个实施例中，当前网络侧设备是通过预定接口与预定服务器进行通讯的，则步骤S210之前，该方法还可以包括如下步骤。
S31，若预定服务器的部署位置在网络侧设备外部，则根据预先获取的预定服务器的地址，向预定服务器发送控制面接口建立请求消息，以请求预定服务器建立当前网络侧设备与预定服务器之间的控制面接口，作为预定接口。
S41，若预定服务器的部署位置在网络侧设备内部，则获取当前网络侧设备与预定服务器所在的网络侧设备之间已有的通讯传输接口，作为预定接口。
通过上述步骤S31和S41，可以实现当前网络侧设备和预定服务器之间控制面板接口通信接口的建立。
在一个实施例中，控制面接口建立请求消息中包括如下信息项的一项或多项：当前网络侧设备所支持的测量、当前网络侧设备所支持的上报方式、当前网络侧设备所支持的网络优化操作和当前网络侧设备的数据面通道地址。
在该实施例中，控制面接口建立请求消息可以用于请求预定服务器，根据当前网络侧设备所支持的测量和当前网络侧设备所支持的上报方式，发送当前网络侧设备能够支持的网络侧测量配置信息，从而能够提高网络侧测量配置的成功率和处理效率。
在该实施例中，控制面接口建立请求消息还可以用于请求预定服务器，根据当前网络侧设备所支持的网络优化操作，发送当前网络侧设备能够支持的网络优化操作指令，以根据接收到的能够支持的网络优化操作指令进行网络优化，提高了网络侧测量执行网络优化操作指令的数据处理性能。
在一个实施例中，网络优化方法还包括：S250，响应于接收到的控制面接口建立响应消息，确定当前网络侧设备与预定服务器之间的控制面接口建立成功；其中，若控制面接口建立请求消息中包括当前网络侧设备的数据面通道地址，则接收到的控制面接口建立响应消息中包括预定服务器的数据面通道地址。
在该实施例中，网络侧设备可以根据接收到的AI服务器的地址，发送接口建立请求，以用于请求建立本网络侧设备与AI服务器之间的控制面接口，接口建立请求中可以携带以下一项或者多项：网络侧设备所支持的测量；网络侧设备所支持侧量的上报方式；网络侧设备所支持的RAN优化操作；网络侧设备的数据面通道地址。
当接口建立请求中携带网络侧设备的数据面通道地址时，响应于接口建立响应中携带的AI服务器的数据面通道地址，建立本网络侧设备与AI服务器之间的数据面接口。以根据需要，利用控制面接口和/或数据面接口，在本网络侧设备与AI服务器之间进行数据传输。
根据本申请实施例的网络优化方法，网络侧设备可以在AI服务器的控制下，对本网络侧设备和终端设备进行测量配置，并可以将本网络侧设备执行测量得到的测量数据和接收的终端设备进行测量得到的测量数据，发送至AI服务器，以用于在AI服务器进行针对网络优化的机器学习，从而接收并执行机器学习得到的网络优化操作指令以进行网络优化。
为了更好地理解本申请，下面通过图3和图4，详细描述本申请实施例的网络优化方法。图3示出一实施例的网络优化方法的流程时序示意图。如图3所示，AI服务器布置在网络侧设备外部，该网络优化方法可以包括如下步骤。
S301，网管***或者核心网通过和AI服务器的接口发送激活消息。
在该步骤中，OAM或者5GC发送激活消息，用以指示AI服务器激活或使用机器学习功能。其中消息中包含指示ML功能是否激活的指示；消息中包含指示RAN网络侧需要满足的策略信息。
S302，AI服务器发送机器学习会话建立请求消息给RAN节点设备，用于配置一个机器学习会话所需要的无线侧测量数据。
在该步骤中，机器学习会话建立请求消息可以包含机器学习会话ID，以用于唯一标志一个ML进程的ID，测量控制信息用以指示RAN节点设备需要测量的测量量和测量上报方式。如果机器学习会话建立请求消息中不包含会话ID，表明请求消息中的测量控制是针对所有的ML进程。
S303，RAN节点设备根据接收消息中的测量控制信息，配置RAN节点设备需要进行的测量以及上报方式。
S304，RAN节点设备根据接收消息中的测量控制信息，配置终端侧的测量控制信息，通过无线资源控制(Radio Resource Control，RRC)建立/重配置消息，携带UE侧测量控制信息，发送给连接在本RAN节点设备上的一个或者多 个UE，指示UE做哪些测量以及如何上报。
S305，如果RAN节点设备测量和UE侧测量都配置成功，则RAN节点设备向AI服务器发送机器学习会话建立响应消息，其中，消息中携带成功指示，消息中可选的携带机器学习会话ID(MLsession ID)。如果RAN节点设备测量或者UE侧测量配置失败，则RAN节点设备向AI服务器发送机器学习会话建立响应消息，其中，消息中携带失败指示，消息中可选的携带机器学习会话ID(ML session ID)。
S306，如果RAN节点设备测量和UE侧测量都配置成功，RAN节点设备和UE都按照指定的测量配置进行相关测量。
S307-01，RAN节点设备向AI服务器发送测量报告。
S307-02，RAN节点设备发送接收到的来自终端的测量报告。
在该步骤中，测量报告消息中可以携带测量量，或者，可以携带测量量和测量所属的机器学习会话ID。
S308，AI服务器根据测量数据以及所配置的策略信息，选择合适的ML算法进行模型训练以及模型预测以及模型更新，并得出优化所需的RAN操作。
S308，AI服务器发送RAN操作请求消息，以通知RAN节点执行相关优化操作。
S310，RAN节点设备执行RAN操作请求消息中的相关优化操作。
S311，RAN节点设备发送RRC重配置消息或者RRC释放消息给连接在本RAN节点设备的UE，通知UE执行相关操作。
通过该实施例中的网络优化方法，AI服务器的机器学习功能被激活后，对RAN节点设备进行测量配置，并通过RAN节点设备对连接于RAN节点设备的终端设备进行测量配置，RAN节点设备和终端设备执行测量，并将测量报告发送至AI服务器；AI服务器根据测量数据以及所配置的策略信息，选择合适的ML算法进行模型训练以及模型预测以及模型更新，得到并发送优化所需的RAN操作，从而通过优化所需的RAN操作进行网络优化。
图4示出另一实施例的网络优化方法的流程时序示意图。如图4所示，AI服务器布置在网络侧设备内部，该网络优化方法可以包括如下步骤。
S401，网管***或者核心网通过和AI服务器的接口发送激活消息。
S402，AI服务器配置本RAN节点设备1所需要的测量。
S403，RAN节点设备1配置终端设备侧的测量控制信息。
在该步骤中，RAN节点设备1通过RRC建立/重配置消息，携带终端设备侧测量控制信息，发送给连接在RAN节点设备1上的一个或者多个终端设备，指示终端设备做哪些测量以及如何上报。
S404，AI服务器发送机器学习会话建立请求消息消息给RAN节点设备2，用于配置一个机器学习会话所需要的无线侧测量数据。
S405，RAN节点设备2根据接收消息中的测量控制信息，配置RAN节点设备2需要测量的测量量以及测量上报方式。
S406，RAN节点设备2根据接收消息中的测量控制信息，配置终端设备侧的测量控制信息。
在该步骤中，RAN节点设备2可以通过RRC建立/重配置消息，携带终端设备侧测量控制信息，发送给连接在本基站上的一个或者多个终端设备，指示终端设备做哪些测量以及如何上报。
S407，如果RAN节点设备2侧测量和终端设备侧测量都配置成功，则基站向RAN节点设备1发送机器学习会话建立响应消息。其中，消息中携带成功指示，消息中可选的携带机器学习会话ID(机器学习session ID)。如果RAN节点设备2侧测量或者终端设备侧测量配置失败，则RAN节点设备2向RAN节点设备1发送机器学习会话建立响应消息，其中，消息中携带失败指示，消息中可选的携带机器学习会话ID(机器学习session ID)。
S408，如果RAN节点设备2侧测量和终端设备侧测量都配置成功，RAN节点设备2和终端设备都按照指定的测量配置进行相关测量。同时RAN节点设备1也按AI服务器配置的测量执行相关测量。
S409，RAN节点设备1接收测量报告。
该步骤中，测量报告可能是连接在RAN节点设备1上的终端设备直接发送给RAN节点设备1的(对应步骤S409-02)，测量报告可能是RAN节点设备2上所作的测量并发送给RAN节点设备1(对应步骤S409-01)，测量报告可能是连接在RAN节点设备2上的终端设备发送给RAN节点设备2并被RAN节点设备2转发给RAN节点设备1的(对应步骤S409-03)。
S410，AI服务器根据测量数据以及所配置的策略信息，选择合适的机器学习算法进行模型训练以及模型预测以及模型更新，并得出优化所需的RAN操作。
S411，AI服务器通知本RAN节点设备1节点执行相关优化操作。
S412，RAN节点设备1执行AI服务器指定的相关优化操作。
在该步骤中，如果相关优化操作涉及到一个或者多个终端设备，RAN节点设备1发送RRC重配置消息或者RRC释放消息给连接在本基站的终端设备，通知终端设备执行相关操作。
S413，RAN节点设备1发送RAN操作请求消息，通知RAN节点设备2执行相关优化操作。
该优化操作可以是一项或者多项操作指示和对应操作所需的参数，如终端设备切换指示，关闭/打开小区指示，无线资源激活/去激活指示，功率调整指示，RRM参数重配置指示，分流操作指示，协议层参数重配置指示等等)。
S414，RAN节点设备2执行RAN操作请求消息中的相关优化操作。
S415，RAN节点设备2发送RRC重配置消息或者RRC释放消息，至连接在本RAN节点设备2的UE，通知UE执行相关操作。
在该步骤中，如果相关优化操作涉及到一个或者多个终端设备，RAN节点设备2发送RRC重配置消息或者RRC释放消息给连接在本基站的终端设备，通知终端设备执行相关操作。
通过上述内容可知，图4中的网络优化方法和图3中的网络优化方法基本相同，不同之处在于：部署了AI服务器的RAN节点设备本身也需要进行测量配置、执行测量；当部署了AI服务器的基站本身需要执行RAN优化操作时，如果相关优化操作涉及到一个或者多个终端设备，AI服务器所在的RAN节点设备，可以发送RRC重配置消息或者RRC释放消息给连接在本RAN节点的终端设备，以指示连接在本RAN节点的终端设备执行指定的优化操作。
在本申请实施例中，部署了AI服务器的基站可以通过和另一个基站的接口，发送ML会话建立请求和接收ML会话响应消息，用于请求相邻的基站参与RAN优化操作，无需额外建立AI服务器与网络侧设备之间的通讯接口，节省网络资源。
图5示出一个实施例的AI服务器与RAN节点设备之间建立通信接口的流程示意图。如图5所示，在一个实施例中，该接口建立过程可以包括如下步骤。
S501，当AI服务器部署在RAN节点外部，网管通过配置消息，或者直接配置AI服务器的地址给基站。
在一些实施例中，如果AI服务器部署在RAN节点内部，则无需执行步骤S501。
S502，基站根据配置的AI服务器的地址，发送通信接口建立请求。
在该步骤中，若将AI服务器和RAN节点之间的接口称为是AI接口，RAN节点向AI服务器的地址发送AI接口建立请求消息，用于建立AI接口。
在一个实施例中，通信接口建立请求中可以包括：网络侧设备(例如基站)所支持的测量；网络侧设备所支持侧量的上报方式；网络侧设备所支持的RAN优化操作。
在一个实施例中，通信接口建立请求中还可以包括：网络侧设备的数据面通道地址。
S503，AI服务器向基站发送接口建立响应消息，用于指示接口是否建立成功。
在该步骤中，如果S502中，AI服务器接收的通信接口建立请求中携带了网络侧设备的数据面通道地址，则可以在接口建立响应消息中携带AI服务器侧的数据面通道地址。
通过上述步骤S501到S503，可以建立AI服务器用于RAN节点的控制面接口，控制面接口可以用于传输控制消息，也可以用于传输机器学习所需的数据，如测量数据等。
S504，根据AI服务器配置的数据面通道地址以及基站侧的数据面通道地址，建立数据面通道。
在本申请实施中，网络侧设备和AI服务器之间可以在数据面通道上传输机器学习所需数据，例如一些较大数据量的数据可以在数据面通道上传输，比如来自基站的测量数据，以缓解控制面接口对应通信通道的数据传输压力，提高数据传输效率。本实施例中建立的数据面通道不限于传输指定用户的数据。
图6示出一个实施例的建立AI服务器与用户设备相关的数据通道的流程示意图。如图6所示，在一个实施例中，该接口建立过程可以包括如下步骤。
S601，获取AI服务器和RAN节点设备(例如基站)之间已建立的A1控制面接口。
AI服务器可以部署在RAN节点外部，或者，AI服务器可以部署在一个RAN节点内部。
S602，AI服务器或者AI服务器所在的基站，向相邻基站发送通道建立请求消息。
作为示例，通道建立请求消息例如可以是终端设备上下文建立请求消息(UE Context Setup Request)，并携带AI服务器侧的数据面通道地址和终端设备标识 (UE ID)。
S603，基站向AI服务器或者AI服务器所在的基站，发送通道建立响应消息。
作为示例，通道建立响应消息例如可以是终端设备上下文建立响应消息(UE Context Setup Response)，并携带基站侧的数据面通道地址。
S604，根据AI服务器配置的数据面通道地址以及基站侧的数据面通道地址，建立UE ID指定和特定用户相关的数据面通道。
在该示例中，RAN节点和AI服务器之间可以在数据面通道上传输和特定用户相关的ML所需数据，通常一些较大数据量的数据可以在数据面通道上传输，比如该用户的测量数据。
在本申请实施例中，RAN节点设备和AI服务器之间可以在数据面通道上传输机器学习所需数据，例如一些较大数据量的数据可以在数据面通道上传输，比如来自基站的测量数据，以缓解控制面接口对应通信通道的数据传输压力，提高数据传输效率。本实施例中建立的数据面通道不限于传输指定用户的数据。
下面结合附图，介绍本申请实施例的一种服务器。图7示出本申请一实施例提供的一种服务器的结构示意图。在一个实施例中，该服务器可能是具有AI功能的服务器，简称AI服务器，如图7所示，该服务器可以包括如下模块。
测量控制信息确定模块710，用于根据预先获取的网络需要满足的策略信息，确定网络侧测量控制信息。
测量配置请求模块720，用于发送会话建立请求至指定的网络侧设备，以请求指定的网络侧设备根据网络侧测量控制信息进行测量配置和对连接于指定的网络侧设备的终端设备进行测量配置。
测量报告接收模块730，接收指定的网络侧设备的测量报告和终端设备的测量报告。
机器学习模块740，用于基于所接收测量报告中的测量数据和策略信息，进行针对网络优化的机器学习，得到网络优化操作指令，网络优化操作指令用于指示指定的网络侧设备和终端设备根据网络优化操作指令进行网络优化。
在一个实施例中，服务器还包括：策略信息获取模块，用于获取预先配置的网络需要满足的策略信息；或者，响应于接收到的激活消息，激活机器学习功能，并获取激活消息中携带的网络需要满足的策略信息。
在一个实施例中，策略信息包括对象标识信息和通信质量指标信息，通信 质量指标信息用于指示对象标识信息所标识的实体所需达到的通信质量；在该实施例中，测量控制信息确定模块710可以用于根据对象标识信息所标识的实体，确定需要执行测量的网络侧设备，作为指定的网络侧设备；以及，根据通信质量指标信息所指示的实体所需达到的通信质量，确定指定的网络侧设备需要配置的测量量和测量上报方式，作为网络侧测量控制信息。
在一个实施例中，当前服务器的部署位置包括在网络侧设备外部和在网络侧设备内部；在该实施例中，测量配置请求模块720可以用于：若当前服务器的部署位置在网络侧设备外部，则发送会话建立请求消息至指定的网络侧设备，会话建立请求消息中包括网络侧测量控制信息；若当前服务器的部署位置在网络侧设备内部，则根据网络侧测量控制信息对当前服务器所在的网络侧设备进行测量配置，并发送会话建立请求消息至指定的网络侧设备，会话建立请求消息中包括网络侧测量控制信息。
在一个实施例中，该服务器还可以包括：会话建立响应模块，用于响应于接收到的会话建立响应消息，确定指定的网络侧设备和终端设备均测量配置成功。
在一个实施例中，会话建立请求、会话建立响应消息和所接收的测量报告中，包括对应的机器学习会话标识，机器学习会话标识用于唯一标识机器学习进程。
在一个实施例中，测量报告接收模块730可以用于：接收指定的网络侧设备的测量报告和由指定的网络侧设备发送的终端设备的测量报告。
在一个实施例中，该服务器还可以包括：网络操作请求发送模块，用于发送网络操作请求消息至指定的网络侧设备，网络操作请求消息中包含网络优化操作指令和对应的操作参数。
在一个实施例中，当前服务器是通过预定接口与指定的网络侧设备进行通讯的，该服务器还包括：第一预定接口确定模块，用于若当前服务器的部署位置在网络侧设备外部，则响应于接收到的控制面接口建立请求消息，建立当前服务器与指定的网络侧设备之间的控制面接口，作为预定接口；第二预定接口确定模块，用于若当前服务器的部署位置在网络侧设备内部，则获取当前服务器所在的网络侧设备与指定的网络侧设备之间已有的通讯传输接口，作为预定接口。
在一个实施例中，控制面接口建立请求消息中包括如下信息项的一项或多项：指定的网络侧设备所支持的测量、指定的网络侧设备所支持的上报方式、 指定的网络侧设备所支持的网络优化操作和指定的网络侧设备的数据面通道地址。
在一个实施例中，该服务器还包括：控制面接口建立响应发送模块，用于发送控制面接口建立响应消息至指定的网络侧设备，以指示控制面接口建立成功；若控制面接口建立请求消息中包括指定的网络侧设备的数据面通道地址，则在发送控制面接口建立响应消息至指定的网络侧设备时，在控制面接口建立响应消息中携带当前服务器的数据面通道地址。
根据本申请实施例的服务器，利用配置并接收的无线网络侧的测量数据和终端侧测量的测量数据，进行机器学习模型训练，得到网络优化操作，从而通过人工智能和机器学习对采集的数据进行深度分析，使用智能化的网络优化方法，为运营商网络优化提供了新的优化方式和网络智能优化流程。
图8示出本申请一实施例提供的网络侧设备的结构示意图。如图8所示，网络侧设备可以包括如下模块。
测量配置模块810，用于响应于接收到来自预定服务器的会话建立请求，根据会话建立请求中的网络侧测量控制信息进行测量配置和对连接于当前网络侧设备的终端设备进行测量配置；测量报告发送模块820，用于将测量的得到当前网络侧设备的测量报告和接收到的终端设备的测量报告，发送至预定服务器，当前网络侧设备的测量报告和终端设备的测量报告在预定服务器中，被用于进行针对网络优化的机器学习。
在一个实施例中，测量配置模块810，可以用于：响应于会话建立请求，根据网络侧测量控制信息进行测量配置，网络侧测量控制信息用于指示当前网络侧设备需要配置的测量量和测量上报方式；根据网络侧测量控制信息，确定连接于当前网络侧设备的终端设备需要配置的测量量和测量上报方式，作为终端侧测量控制信息；向终端设备发送第一无线资源控制消息，以指示终端设备根据终端侧测量控制信息进行测量配置。
在一个实施例中，网络侧设备还可以包括：会话建立响应消息发送模块，用于若当前网络侧设备测量配置成功，且接收到终端设备的测量配置响应消息，则向预定服务器发送会话建立响应消息，以向预定服务器反馈当前网络侧设备和终端设备均测量配置成功。
在一个实施例中，接收到的会话建立请求、会话建立响应消息、以及发送至预定服务器的测量报告中包括对应的机器学习会话标识，机器学习会话标识用于唯一标识机器学习进程。
在一个实施例中，网络侧设备还可以包括：操作指令接收模块，用于接收并执行来自预定服务器的网络优化操作指令；操作指令执行单元，用于若网络优化操作指令涉及终端设备，则确定网络优化操作指令中与终端设备相关的操作；相关操作发送单元，用于发送第二无线资源控制消息至终端设备，以指示终端设备执行相关的操作。
在一个实施例中，当前网络侧设备是通过预定接口与预定服务器进行通讯的，网络侧设备还可以包括：预定接口建立请求模块，用于若预定服务器的部署位置在当前网络侧设备外部，则根据预先获取的预定服务器的地址，向预定服务器发送控制面接口建立请求消息，以请求预定服务器建立当前网络侧设备与预定服务器之间的控制面接口，作为预定接口；通讯传输接口获取模块，用于若预定服务器的部署位置在当前网络侧设备内部，则获取当前网络侧设备与预定服务器所在的网络侧设备之间已有的通讯传输接口，作为预定接口。
在一个实施例中，控制面接口建立请求消息中包括如下信息项的一项或多项：当前网络侧设备所支持的测量、当前网络侧设备所支持的上报方式、当前网络侧设备所支持的网络优化操作和当前网络侧设备的数据面通道地址。
在一个实施例中，网络侧设备还可以包括：控制面接口建立响应接收模块，用于响应于接收到的控制面接口建立响应消息，确定当前网络侧设备与预定服务器之间的控制面接口建立成功；其中，若控制面接口建立请求消息中包括当前网络侧设备的数据面通道地址，则接收到的控制面接口建立响应消息中包括预定服务器的数据面通道地址。
根据本申请实施例的网络侧设备，可以在AI服务器的控制下，对本网络侧设备和终端设备进行测量配置，将本网络侧设备执行测量得到的测量数据和接收的终端设备进行测量得到的测量数据发送至AI服务器，以用于在AI服务器进行针对网络优化的机器学习，从而接收并执行机器学习得到的网络优化操作指令以进行网络优化。
图9示出本申请一实施例提供的网络优化***的结构示意图。如图9所示，网络优化***可以包括如下服务器910和一个或多个网络侧设备920。
服务器910，该服务器910可以用于执行上述实施例中结合图1描述的网络优化方法。
一个或多个网络侧设备920，网络侧设备920用于执行上述实施例中结合图2描述的网络优化方法。
在该实施例中，AI服务器910与结合图7描述的AI服务器具有相同或等同 的结构，并可以执行上述实施例描述的应用于AI服务器的网络优化方法；网络侧设备920与结合图8描述的网络侧设备具有相同或等同的结构，并可以执行上述实施例描述的应用于网络侧设备的网络优化方法。
本申请并不局限于上文实施例中所描述并在图中示出的特定配置和处理。为了描述的方便和简洁，这里省略了对已知方法的描述，并且上述描述的***、模块和单元的工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。
图10示出本申请实施例提供的一种计算设备的硬件架构的结构图。
如图10所示，计算设备1000包括输入设备1001、输入接口1002、中央处理器1003、存储器1004、输出接口1005、以及输出设备1006。其中，输入接口1002、中央处理器1003、存储器1004、以及输出接口1005通过总线1010相互连接，输入设备1001和输出设备1006分别通过输入接口1002和输出接口1005与总线1010连接，进而与计算设备1000的其他组件连接。
输入设备1001接收来自外部的输入信息，并通过输入接口1002将输入信息传送到中央处理器1003；中央处理器1003基于存储器1004中存储的计算机可执行指令对输入信息进行处理以生成输出信息，将输出信息临时或者永久地存储在存储器1004中，然后通过输出接口1005将输出信息传送到输出设备1006；输出设备1006将输出信息输出到计算设备1000的外部供用户使用。
在一个实施例中，图10所示的计算设备可以被实现为一种服务器，该服务器可以包括：存储器，被配置为存储程序；处理器，被配置为运行存储器中存储的程序，以执行上述实施例描述的应用于AI服务器的网络优化方法。
在一个实施例中，图10所示的计算设备可以被实现为一种网络侧设备，该网络侧设备可以包括：存储器，被配置为存储程序；处理器，被配置为运行存储器中存储的程序，以执行上述实施例描述的应用于网络侧设备的网络优化方法。
一般来说，本申请的多种实施例可以在硬件或专用电路、软件、逻辑或其任何组合中实现。例如，一些方面可以被实现在硬件中，而其它方面可以被实现在可以被控制器、微处理器或其它计算装置执行的固件或软件中，尽管本申请不限于此。
本申请的实施例可以通过移动装置的数据处理器执行计算机程序指令来实现，例如在处理器实体中，或者通过硬件，或者通过软件和硬件的组合。计算机程序指令可以是汇编指令、指令集架构(Instruction Set Architecture，ISA)指 令、机器指令、机器相关指令、微代码、固件指令、状态设置数据、或者以一种或多种编程语言的任意组合编写的源代码或目标代码。
本申请附图中的任何逻辑流程的框图可以表示程序步骤，或者可以表示相互连接的逻辑电路、模块和功能，或者可以表示程序步骤与逻辑电路、模块和功能的组合。计算机程序可以存储在存储器上。存储器可以具有任何适合于本地技术环境的类型并且可以使用任何适合的数据存储技术实现，例如但不限于只读存储器(Read-Only Memory，ROM)、随机访问存储器(Random Access Memory，RAM)、光存储器装置和***(数码多功能光碟(Digital Video Disc，DVD)或光盘(Compact Disk，CD))等。计算机可读介质可以包括非瞬时性存储介质。数据处理器可以是任何适合于本地技术环境的类型，例如但不限于通用计算机、专用计算机、微处理器、数字信号处理器(Digital Signal Processing，DSP)、专用集成电路(Application Specific Integrated Circuit，ASIC)、可编程逻辑器件(Field-Programmable Gate Array，FPGA)以及基于多核处理器架构的处理器。

Claims (25)

  1. 一种网络优化方法，应用于服务器，包括：
    根据预先获取的网络需要满足的策略信息，确定网络侧测量控制信息；
    发送会话建立请求至指定的网络侧设备，以请求所述指定的网络侧设备根据所述网络侧测量控制信息对所述指定的网络侧设备进行测量配置和对连接于所述指定的网络侧设备的终端设备进行测量配置；
    接收所述指定的网络侧设备的测量报告和所述终端设备的测量报告；
    基于所接收测量报告中的测量数据和所述策略信息，进行针对网络优化的机器学习，得到网络优化操作指令，所述网络优化操作指令用于指示所述指定的网络侧设备和所述终端设备根据所述网络优化操作指令进行网络优化。
  2. 根据权利要求1所述的方法，在所述根据预先获取的网络需要满足的策略信息，确定网络侧测量控制信息之前，还包括：
    获取预先配置的所述网络需要满足的策略信息；或者，
    响应于接收到的激活消息，激活机器学习功能，并获取所述激活消息中携带的所述网络需要满足的策略信息。
  3. 根据权利要求1所述的方法，其中，所述策略信息包括对象标识信息和通信质量指标信息，所述通信质量指标信息用于指示所述对象标识信息所标识的实体所需达到的通信质量；
    所述根据预先获取的网络需要满足的策略信息，确定网络侧测量控制信息，包括：
    根据所述对象标识信息所标识的实体，确定需要执行测量的网络侧设备，作为所述指定的网络侧设备；
    根据所述通信质量指标信息所指示的所述实体所需达到的通信质量，确定所述指定的网络侧设备需要配置的测量量和测量上报方式，作为所述网络侧测量控制信息。
  4. 根据权利要求1所述的方法，其中，所述服务器的部署位置包括在网络侧设备外部或在网络侧设备内部；
    所述发送会话建立请求至指定的网络侧设备，包括：
    在所述服务器的部署位置为在网络侧设备外部的情况下，发送所述会话建立请求消息至所述指定的网络侧设备，所述会话建立请求消息中包括所述网络侧测量控制信息；
    在所述服务器的部署位置为在网络侧设备内部的情况下，根据所述网络侧 测量控制信息对所述服务器所在的网络侧设备进行测量配置，并发送所述会话建立请求消息至所述指定的网络侧设备，所述会话建立请求消息中包括所述网络侧测量控制信息。
  5. 根据权利要求1所述的方法，在所述发送会话建立请求至指定的网络侧设备之后，还包括：
    响应于接收到的会话建立响应消息，确定所述指定的网络侧设备和所述终端设备均测量配置成功。
  6. 根据权利要求5所述的方法，其中，
    所述会话建立请求、所述会话建立响应消息和所接收的测量报告中，均包括对应的机器学习会话标识，所述机器学习会话标识用于唯一标识机器学习进程。
  7. 根据权利要求1所述的方法，其中，所述接收所述指定的网络侧设备的测量报告和所述终端设备的测量报告，包括：
    接收所述指定的网络侧设备的测量报告和由所述指定的网络侧设备发送的所述终端设备的测量报告。
  8. 根据权利要求1所述的方法，其中，在所述基于所接收测量报告中的测量数据和所述策略信息，进行针对网络优化的机器学习，得到网络优化操作指令之后，还包括：
    发送网络操作请求消息至所述指定的网络侧设备，所述网络操作请求消息中包含所述网络优化操作指令和操作对应的操作参数。
  9. 根据权利要求1所述的方法，其中，所述服务器是通过预定接口与所述指定的网络侧设备进行通讯的；
    在所述发送会话建立请求至指定的网络侧设备之前，还包括：
    在所述服务器的部署位置在网络侧设备外部的情况下，响应于接收到的控制面接口建立请求消息，建立所述服务器与所述指定的网络侧设备之间的控制面接口，作为所述预定接口；
    在所述服务器的部署位置在网络侧设备内部的情况下，获取所述服务器所在的网络侧设备与所述指定的网络侧设备之间已有的通讯传输接口，作为所述预定接口。
  10. 根据权利要求9所述的方法，其中，
    所述控制面接口建立请求消息中包括如下信息项中的至少一项：所述指定的网络侧设备所支持的测量、所述指定的网络侧设备所支持的上报方式、所述 指定的网络侧设备所支持的网络优化操作和所述指定的网络侧设备的数据面通道地址。
  11. 根据权利要求10所述的方法，在所述响应于接收到的控制面接口建立请求消息，建立所述服务器与所述指定的网络侧设备之间的控制面接口，作为所述预定接口之后，还包括：
    发送控制面接口建立响应消息至所述指定的网络侧设备，以指示所述控制面接口建立成功；
    在所述控制面接口建立请求消息中包括所述指定的网络侧设备的数据面通道地址的情况下，在发送所述控制面接口建立响应消息至所述指定的网络侧设备时，在所述控制面接口建立响应消息中携带所述服务器的数据面通道地址。
  12. 一种网络优化方法，应用于网络侧设备，包括：
    响应于接收到来自预定服务器的会话建立请求，根据所述会话建立请求中的网络侧测量控制信息对所述网络侧设备进行测量配置和对连接于所述网络侧设备的终端设备进行测量配置；
    将测量得到的所述网络侧设备的测量报告和接收到的所述终端设备的测量报告，发送至所述预定服务器，所述网络侧设备的测量报告和所述终端设备的测量报告在所述预定服务器中，被用于进行针对网络优化的机器学习。
  13. 根据权利要求12所述的方法，其中，所述响应于接收到来自预定服务器的会话建立请求，根据所述会话建立请求中的网络侧测量控制信息对所述网络侧设备进行测量配置和对连接于网络侧设备的终端设备进行测量配置，包括：
    响应于所述会话建立请求，根据所述网络侧测量控制信息对所述网络侧设备进行测量配置，所述网络侧测量控制信息用于指示所述网络侧设备需要配置的测量量和测量上报方式；
    根据所述网络侧测量控制信息，确定连接于所述网络侧设备的终端设备需要配置的测量量和测量上报方式，作为终端侧测量控制信息；
    向所述终端设备发送第一无线资源控制消息，以指示所述终端设备根据所述终端侧测量控制信息对所述终端设备进行测量配置。
  14. 根据权利要求12所述的方法，其中，
    在所述网络侧设备测量配置成功，且接收到所述终端设备的测量配置响应消息的情况下，向所述预定服务器发送会话建立响应消息，以向所述预定服务器反馈所述网络侧设备和所述终端设备均测量配置成功。
  15. 根据权利要求14所述的方法，其中，
    所述接收到的会话建立请求、所述会话建立响应消息、以及发送至所述预定服务器的测量报告中均包括对应的机器学习会话标识，所述机器学习会话标识用于唯一标识机器学习进程。
  16. 根据权利要求12所述的方法，还包括：
    接收并执行来自所述预定服务器的网络优化操作指令；
    在所述网络优化操作指令涉及所述终端设备的情况下，确定所述网络优化操作指令中与所述终端设备相关的操作；
    发送第二无线资源控制消息至所述终端设备，以指示所述终端设备执行所述相关的操作。
  17. 根据权利要求12所述的方法，其中，所述网络侧设备是通过预定接口与所述预定服务器进行通讯的；
    在所述响应于接收到来自预定服务器的会话建立请求，根据所述会话建立请求中的网络侧测量控制信息对所述网络侧设备进行测量配置和对连接于所述网络侧设备的终端设备进行测量配置之前，还包括：
    在所述预定服务器的部署位置在网络侧设备外部的情况下，根据预先获取的所述预定服务器的地址，向所述预定服务器发送控制面接口建立请求消息，以请求所述预定服务器建立所述网络侧设备与所述预定服务器之间的控制面接口，作为所述预定接口；
    在所述预定服务器的部署位置在网络侧设备内部的情况下，获取所述网络侧设备与所述预定服务器所在的网络侧设备之间已有的通讯传输接口，作为所述预定接口。
  18. 根据权利要求17所述的方法，其中，
    所述控制面接口建立请求消息中包括如下信息项中的至少一项：所述网络侧设备所支持的测量、所述网络侧设备所支持的上报方式、所述网络侧设备所支持的网络优化操作和所述网络侧设备的数据面通道地址。
  19. 根据权利要求18所述的方法，还包括：
    响应于接收到的控制面接口建立响应消息，确定所述网络侧设备与所述预定服务器之间的控制面接口建立成功；其中，在所述控制面接口建立请求消息中包括所述网络侧设备的数据面通道地址的情况下，所述接收到的控制面接口建立响应消息中包括所述预定服务器的数据面通道地址。
  20. 一种服务器，包括：
    测量控制信息确定模块，设置为根据预先获取的网络需要满足的策略信息， 确定网络侧测量控制信息；
    测量配置请求模块，设置为发送会话建立请求至指定的网络侧设备，以请求所述指定的网络侧设备根据所述网络侧测量控制信息对所述指定的网络侧设备进行测量配置和对连接于所述指定的网络侧设备的终端设备进行测量配置；
    测量报告接收模块，设置为接收所述指定的网络侧设备的测量报告和所述终端设备的测量报告；
    机器学习模块，设置为基于所接收测量报告中的测量数据和所述策略信息，进行针对网络优化的机器学习，得到网络优化操作指令，所述网络优化操作指令用于指示所述指定的网络侧设备和所述终端设备根据所述网络优化操作指令进行网络优化。
  21. 一种网络侧设备，包括：
    测量配置模块，设置为响应于接收到来自预定服务器的会话建立请求，根据所述会话建立请求中的网络侧测量控制信息对所述网络侧设备进行测量配置和对连接于所述网络侧设备的终端设备进行测量配置；
    测量报告发送模块，设置为将测量得到的所述网络侧设备的测量报告和接收到的所述终端设备的测量报告，发送至所述预定服务器，所述网络侧设备的测量报告和所述终端设备的测量报告在所述预定服务器中，被用于进行针对网络优化的机器学习。
  22. 一种服务器，包括：
    至少一个处理器；
    存储器，存储有至少一个程序，当所述至少一个程序被所述至少一个处理器执行，使得所述至少一个处理器实现权利要求1-11中任一项所述的网络优化方法。
  23. 一种网络侧设备，包括：
    至少一个处理器；
    存储器，存储有至少一个程序，当所述至少一个程序被所述至少一个处理器执行，使得所述至少一个处理器实现权利要求12-19中任一项所述的网络优化方法。
  24. 一种网络优化***，包括：
    服务器，所述服务器设置为执行权利要求1-11中任一项所述的网络优化方法；
    至少一个网络侧设备，所述至少一个网络侧设备设置为执行权利要求12-19中任一项所述的网络优化方法。
  25. 一种存储介质，存储有计算机程序，所述计算机程序被处理器执行时实现权利要求1-11或权利要求12-19中任一项所述的网络优化方法。
PCT/CN2021/073544 2020-05-24 2021-01-25 网络优化方法、服务器、网络侧设备、***和存储介质 WO2021238277A1 (zh)

Priority Applications (4)

Application Number Priority Date Filing Date Title
US17/923,098 US20230180025A1 (en) 2020-05-24 2021-01-25 Network optimisation method, system and storage medium
MX2022014678A MX2022014678A (es) 2020-05-24 2021-01-25 Metodo de optimizacion de red, servidor, dispositivo del lado de la red, sistema y medio de almacenamiento.
EP21812406.3A EP4161128A4 (en) 2020-05-24 2021-01-25 NETWORK OPTIMIZATION METHOD, SERVER, NETWORK-SIDE DEVICE, SYSTEM, AND STORAGE MEDIUM
AU2021279510A AU2021279510B2 (en) 2020-05-24 2021-01-25 Network optimisation method, server, network side device, system, and storage medium

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010445463.4A CN112512059A (zh) 2020-05-24 2020-05-24 网络优化方法、服务器、网络侧设备、***和存储介质
CN202010445463.4 2020-05-24

Publications (1)

Publication Number Publication Date
WO2021238277A1 true WO2021238277A1 (zh) 2021-12-02

Family

ID=74953272

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/073544 WO2021238277A1 (zh) 2020-05-24 2021-01-25 网络优化方法、服务器、网络侧设备、***和存储介质

Country Status (6)

Country Link
US (1) US20230180025A1 (zh)
EP (1) EP4161128A4 (zh)
CN (1) CN112512059A (zh)
AU (1) AU2021279510B2 (zh)
MX (1) MX2022014678A (zh)
WO (1) WO2021238277A1 (zh)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115499899A (zh) * 2022-11-21 2022-12-20 国网天津市电力公司电力科学研究院 边缘物联代理装置的通信时延测试方法、装置及存储介质
WO2024072549A1 (en) * 2022-09-26 2024-04-04 Qualcomm Incorporated Machine learning signaling and operations for wireless local area networks (wlans)
WO2024074018A1 (en) * 2023-03-24 2024-04-11 Lenovo (Beijing) Limited Method and apparatus of supporting network optimization

Families Citing this family (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113423113B (zh) * 2021-06-17 2022-06-03 中国联合网络通信集团有限公司 无线参数优化处理方法、装置及服务器
CN115915192A (zh) * 2021-08-20 2023-04-04 中兴通讯股份有限公司 网络优化方法、装置、电子设备和存储介质
CN118140511A (zh) * 2021-10-22 2024-06-04 上海诺基亚贝尔股份有限公司 体验质量测量
CN114071560B (zh) * 2021-11-05 2023-06-16 中国联合网络通信集团有限公司 网络优化方法、装置及存储介质
CN114095956B (zh) * 2021-11-09 2023-08-01 中国联合网络通信集团有限公司 网络优化方法、装置及存储介质
CN115866637A (zh) * 2022-01-17 2023-03-28 中兴通讯股份有限公司 通信方法、设备和存储介质
CN116782262A (zh) * 2022-03-10 2023-09-19 维沃移动通信有限公司 数据采集方法、装置、终端及网络侧设备
WO2024031469A1 (en) * 2022-08-10 2024-02-15 Zte Corporation Method of artificial intelligence-assisted configuration in wireless communication system
WO2024036456A1 (zh) * 2022-08-15 2024-02-22 北京小米移动软件有限公司 一种基于人工智能ai提供服务的方法、装置、设备及存储介质

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140098762A1 (en) * 2012-10-04 2014-04-10 Benu Networks, Inc. Application and content awareness for self optimizing networks
CN106507398A (zh) * 2016-12-28 2017-03-15 南京邮电大学 一种基于持续学习的网络自优化方法
CN107276785A (zh) * 2017-05-03 2017-10-20 北京星网锐捷网络技术有限公司 一种无线局域网优化方法及装置
CN110383877A (zh) * 2017-03-10 2019-10-25 华为技术有限公司 网络策略优化的***和方法

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP3516895B1 (en) * 2016-10-13 2022-12-28 Huawei Technologies Co., Ltd. Method and unit for radio resource management using reinforcement learning

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140098762A1 (en) * 2012-10-04 2014-04-10 Benu Networks, Inc. Application and content awareness for self optimizing networks
CN106507398A (zh) * 2016-12-28 2017-03-15 南京邮电大学 一种基于持续学习的网络自优化方法
CN110383877A (zh) * 2017-03-10 2019-10-25 华为技术有限公司 网络策略优化的***和方法
CN107276785A (zh) * 2017-05-03 2017-10-20 北京星网锐捷网络技术有限公司 一种无线局域网优化方法及装置

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP4161128A4

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2024072549A1 (en) * 2022-09-26 2024-04-04 Qualcomm Incorporated Machine learning signaling and operations for wireless local area networks (wlans)
CN115499899A (zh) * 2022-11-21 2022-12-20 国网天津市电力公司电力科学研究院 边缘物联代理装置的通信时延测试方法、装置及存储介质
CN115499899B (zh) * 2022-11-21 2023-05-26 国网天津市电力公司电力科学研究院 边缘物联代理装置的通信时延测试方法、装置及存储介质
WO2024074018A1 (en) * 2023-03-24 2024-04-11 Lenovo (Beijing) Limited Method and apparatus of supporting network optimization

Also Published As

Publication number Publication date
AU2021279510A1 (en) 2022-12-08
EP4161128A1 (en) 2023-04-05
US20230180025A1 (en) 2023-06-08
MX2022014678A (es) 2023-01-04
EP4161128A4 (en) 2024-06-19
CN112512059A (zh) 2021-03-16
AU2021279510B2 (en) 2023-11-16

Similar Documents

Publication Publication Date Title
WO2021238277A1 (zh) 网络优化方法、服务器、网络侧设备、***和存储介质
WO2021238397A1 (zh) 网络优化方法、服务器、客户端设备、网络侧设备、网络设备、***和介质
US11516090B2 (en) Open network automation platform (ONAP)—fifth generation core (5GC) interaction for analytics
US11202231B2 (en) Management device and method for controlling end-to-end network in wireless communication system
US20230209390A1 (en) Intelligent Radio Access Network
EP4099635A1 (en) Method and device for selecting service in wireless communication system
US9918239B2 (en) Self-optimizing network (SON) system for mobile networks
US8971294B2 (en) Method, apparatus and system for reporting MDT log result
US20230262478A1 (en) Model configuration method and apparatus
WO2023069534A1 (en) Using ai-based models for network energy savings
US20220086257A1 (en) Flexible data analytics processing and exposure in 5gc
US20240057139A1 (en) Optimization of deterministic and non-deterministic traffic in radio-access network (ran)
JP2022105306A (ja) オープン無線アクセスネットワーク（ｏ－ｒａｎ）環境でハンドオーバ（ｈｏ）手順を開始するための方法及び装置
US20240112087A1 (en) Ai/ml operation in single and multi-vendor scenarios
CN107333279B (zh) 最小化路测方法、基站及网络管理***
WO2022209234A1 (ja) Ｒａｎノード、ｕｅ及び方法
JP2022540592A (ja) 無線通信ネットワークの条件付き構成
US20240196252A1 (en) Managing resources in a radio access network
US20230344717A1 (en) Policy conflict management method, apparatus, and system
US20230354147A1 (en) Connectionless mobility management for hybrid networks using relativistic routing protocol
WO2024007156A1 (zh) 一种通信方法和装置
WO2023126468A1 (en) Systems and methods for inter-node verification of aiml models
US20210400558A1 (en) Wireless Device, Management Server and Methods Therein for Determining Transmission of Uplink Data
WO2024027980A1 (en) Configuration of ue context surviving during ai/ml operation
CN117812672A (zh) 网络能耗优化方法及基站

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: 21812406

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2021279510

Country of ref document: AU

Date of ref document: 20210125

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2021812406

Country of ref document: EP

Effective date: 20230102