CN114554516A

CN114554516A - Communication method and device

Info

Publication number: CN114554516A
Application number: CN202011332695.5A
Authority: CN
Inventors: 张四海; 郭俊遥; 朱近康; 倪锐
Original assignee: University of Science and Technology of China USTC; Huawei Technologies Co Ltd
Current assignee: University of Science and Technology of China USTC; Huawei Technologies Co Ltd
Priority date: 2020-11-24
Filing date: 2020-11-24
Publication date: 2022-05-27

Abstract

The embodiment of the application provides a communication method and related equipment, which are used for converting continuously-changed flow data in first data volume information into first service state information, can be used for describing long correlation of a network, and can improve the prediction effect of a prediction model on the flow service of a long-term network. In the method, a first network device receives service record data from a second network device, wherein the service record data comprises first data volume information, and the first data volume information is used for indicating the data volume of a first service occurring in a first time period; then, the first network device carries out quantization processing on the service record data according to the target quantization parameter to obtain first service state information; and then, the first network equipment takes the first service state information as the input of a preset model, and outputs second service state information of the first service in a second time period through the processing of the preset model.

Description

Communication method and device

Technical Field

The present application relates to the field of communications, and in particular, to a communication method and apparatus.

Background

In mobile communication networks, traffic prediction has been an important issue, and in the face of increasing number of users and traffic load, operators must meet quality of service (QOS) requirements of users. Generally speaking, the prediction of traffic can be realized by means of traffic modeling.

At present, in a conventional communication network, because the data transmission amount is small and the application is single, a model of the service of the conventional telecommunication network is usually used for reference, a classical poisson process is used for modeling to describe the network traffic, and the established model is obtained by counting continuously changing traffic data in the current network. Wherein, the service sequence described by the continuously changing traffic data has Short Range Dependency (SRD), and can be applied to the prediction of the traffic service of the short-term network.

However, in the above method, when the number of services in the network increases, since the continuously changing traffic data cannot describe the Long Range Dependency (LRD) of the network, it is easy to cause poor prediction effect of the traffic service in the network for a long time.

Disclosure of Invention

The embodiment of the application provides a communication method and related equipment, which are used for converting continuously-changed flow data in first data volume information into first service state information, can be used for describing long correlation of a network, and can improve the prediction effect of a prediction model on the flow service of a long-term network.

A first aspect of the embodiments of the present application provides a communication method, which may be applied to a first network device, and may also be applied to component execution (e.g., a processor, a chip, or a system-on-chip, etc.) of the first network device. In the method, a first network device receives service record data from a second network device, wherein the service record data comprises first data volume information, and the first data volume information is used for indicating the data volume of a first service occurring in a first time period; then, the first network device carries out quantization processing on the service record data according to the target quantization parameter to obtain first service state information; and then, the first network equipment takes the first service state information as the input of a preset model, and outputs second service state information of the first service in a second time period through the processing of the preset model. The input data input into the prediction model and used as the input of the preset model for predicting to obtain the second service state information of the first service in the second time period comprises the first service state information, and the first service state information is obtained by quantizing the service record data. That is to say, the corresponding continuous value (or a large number of possible discrete values) of the first data volume information included in the service record data is approximated to a finite number (or less) of discrete states, and the obtained first service state information enables continuously-changing traffic data in the first data volume information to be converted into the first service state information, which can be used for describing long correlation of a network and improving the prediction effect of a prediction model on the traffic service of a long-term network.

In a specific implementation manner of the first aspect of the embodiment of the present application, before the first network device performs quantization processing on the service record data according to the target quantization parameter to obtain the first service state information, the method further includes: the first network equipment carries out quantization processing on the service record data according to the initial quantization parameter to obtain initial service state information; the first network equipment determines a predictability upper bound value corresponding to the initial service state information; and the first network equipment updates the initial quantization parameter according to the predictability upper bound value and the initial prediction precision to obtain the target quantization parameter.

In this embodiment, before the first network device performs quantization processing on the service record data according to the target quantization parameter to obtain the first service state information, the first network device may further include an update process of the quantization parameter. The process specifically includes the steps of carrying out quantization processing on the service record data according to preset initial quantization parameters to obtain initial service state information, then determining a predictability upper bound value corresponding to the initial service state information, and further updating the initial quantization parameters according to the predictability upper bound value and initial prediction precision to obtain the target quantization parameters. The upper predictability bound in the flow prediction process is effective measurement, theoretical guidance can be provided for flow prediction, the optimal quantization parameter is determined as a target quantization parameter according to the set updating process, and then the target quantization parameter is used for predicting the flow state so as to ensure the prediction accuracy.

In a specific implementation manner of the first aspect of the embodiment of the present application, the determining, by the first network device, the predictability upper bound value corresponding to the initial service state information includes: after the first network device determines the first entropy corresponding to the initial service state information, the first network device determines the upper predictability threshold according to the first entropy.

In this embodiment, in the process of determining the predictability upper bound value, the predictability upper bound value may be specifically determined by a first entropy value corresponding to the initial service state information. The method comprises the steps of carrying out quantitative analysis on initial service state information, carrying out quantitative analysis on the initial service state information, and carrying out quantitative analysis on the initial service state information.

In a specific implementation manner of the first aspect of the embodiment of the present application, after the first network device uses the first service state information as an input of a preset model, and outputs second service state information of the first service in a second time period through processing of the preset model, the method further includes: and the first network equipment performs inverse quantization processing on the second service state information according to the target quantization parameter to obtain second data volume information, wherein the second data volume information is used for indicating the data volume of the first service in the second time period.

In this embodiment, after the first network device obtains the second service state information through processing output of the preset model, the second service state information may further be subjected to inverse quantization processing according to the target quantization parameter, so as to obtain second data volume information, where the second data volume information is used to indicate a data volume of the first service occurring in the second time period, that is, the output of the model is represented in the second data volume information by continuously changing traffic data, and a specific implementation form of the output of the model is provided, so as to more intuitively represent the output of the model.

In a specific implementation manner of the first aspect of the embodiment of the present application, the method further includes: the first network device sends the second data volume information to the second network device.

In this embodiment, the first network device may further send, to the second network device, the second data volume information used for indicating the data volume of the first service occurring in the second time period, and may be applied to design processes of resource reservation, energy allocation, base station dormancy, and the like of the second network device.

In a specific implementation manner of the first aspect of the embodiment of the present application, the service record data further includes a first identifier of the network device corresponding to the first data volume information; and/or the service record data further comprises a second identifier of the terminal device corresponding to the first data volume information.

In this embodiment, the first data volume information may specifically indicate a data volume of the first service occurring in a first time period in a part of terminal devices (or all of the terminal devices) that provide services by the second network device, and/or the first data volume information may specifically indicate a data volume of the first service occurring in a certain specific terminal device that provides services by the second network device, where the two implementation manners may be specifically distinguished by the first identifier and the second identifier, and multiple implementable manners of the scheme are provided.

In a specific implementation manner of the first aspect of the embodiment of the present application, the first service includes one or more of a voice service, a data traffic service, and a real-time traffic service of application software.

In this embodiment, the first service may specifically include one or more of a voice service, a data traffic service, and a real-time traffic service of application software, and the realizability of the scheme is improved while multiple implementation manners of the first service are provided.

A second aspect of the embodiments of the present application provides a communication method, which may be applied to a first network device, and may also be applied to component execution (e.g., a processor, a chip, or a chip system, etc.) of the first network device. In the method, a second network device sends service record data to a first network device, wherein the service record data comprises first data volume information, and the first data volume information is used for indicating the data volume of a first service occurring in a first time period; then, the second network device receives second data volume information from the first network device, where the second data volume information is used to indicate a data volume of the first service occurring in the second time period, and the second data volume information is obtained by performing quantization processing and inverse quantization processing on the service record data. The second network device receives the obtained second data volume information, which is obtained by performing quantization processing and inverse quantization processing on the service record data, that is, after the first data volume information contained in the service record data and corresponding continuous values (or a large number of possible discrete values) are approximated to a finite number (or less) of discrete states, further inverse quantization processing is performed to obtain the second data volume information, so that continuously changing traffic data in the first data volume information is subjected to quantization processing conversion, which can be used for describing long correlation of a network, and the prediction effect of the second data volume information on the traffic of a long-term network is improved.

In a specific implementation manner of the second aspect of the embodiment of the present application, before the second network device sends the service record data to the first network device, the method further includes: the second network equipment records initial information of the first service in the first time period; and the second network equipment performs target processing on the initial information to obtain the first data volume information, wherein the target processing comprises feature extraction processing and/or data desensitization processing.

In this embodiment, before the second network device sends the service record data to the first network device, the method may further include: the second network device records initial information of the first service occurring in the first time period, and performs target processing on the initial information to obtain the first data volume information, wherein the target processing includes feature extraction processing and/or data desensitization processing. Wherein, the characteristic processing is used for extracting effective information in the initial information, and the data desensitization processing is used for protecting the privacy of the user, thereby further optimizing the scheme.

In a specific implementation manner of the second aspect of the embodiment of the present application, the service record data further includes a first identifier of the network device corresponding to the first data volume information; and/or the service record data further comprises a second identifier of the terminal device corresponding to the first data volume information.

In a specific implementation manner of the second aspect of the embodiment of the present application, the first service includes one or more of a voice service, a data traffic service, and a real-time traffic service of application software.

A third aspect of embodiments of the present application provides a communication apparatus, including:

a transceiving unit, configured to receive service record data from a second network device, where the service record data includes first data volume information, and the first data volume information is used to indicate a data volume of a first service occurring in a first time period;

the processing unit is used for carrying out quantization processing on the service record data according to the target quantization parameter to obtain first service state information;

the processing unit is further configured to use the first service state information as an input of a preset model, and output second service state information of the first service in a second time period through processing of the preset model.

In a specific implementation manner of the third aspect of the embodiment of the present application, the processing unit is further configured to:

carrying out quantization processing on the service record data according to the initial quantization parameter to obtain initial service state information;

determining a predictability upper bound value corresponding to the initial service state information;

and updating the initial quantization parameter according to the predictability upper bound value and the initial prediction precision to obtain the target quantization parameter.

In a specific implementation manner of the third aspect of the embodiment of the present application, the processing unit is specifically configured to:

after determining a first entropy corresponding to the initial service state information, determining the predictability upper bound value according to the first entropy.

In a specific implementation manner of the third aspect of the embodiment of the present application, the processing unit is further configured to perform dequantization processing on the second service state information according to the target quantization parameter to obtain second data volume information, where the second data volume information is used to indicate a data volume of the first service occurring in the second time period.

In a specific implementation manner of the third aspect of the embodiment of the present application, the transceiver unit is further configured to:

and sending the second data volume information to the second network equipment.

In a particular implementation of the third aspect of the embodiments of the present application,

the service record data also comprises a first identifier of the network equipment corresponding to the first data volume information; and/or the presence of a gas in the gas,

the service record data further includes a second identifier of the terminal device corresponding to the first data volume information.

In a specific implementation manner of the third aspect of the embodiment of the present application, the first service includes one or more of a voice service, a data traffic service, and a real-time traffic service of application software.

In the third aspect of the embodiment of the present application, the constituent modules of the communication device may also be configured to execute the steps executed in each possible implementation manner of the first aspect, which may specifically refer to the first aspect, and are not described herein again.

A fourth aspect of the embodiments of the present application provides a communication apparatus, including a processing unit and a transceiver unit:

the processing unit is configured to send service record data to the first network device through the transceiving unit, where the service record data includes first data volume information, and the first data volume information is used to indicate a data volume of the first service occurring in a first time period;

the processing unit is further configured to receive, by the transceiver unit, second data volume information from the first network device, where the second data volume information is used to indicate a data volume of the first service occurring in the second time period, and the second data volume information is obtained by performing quantization processing and dequantization processing on the service record data.

In a specific implementation manner of the fourth aspect of the embodiment of the present application, the processing unit is specifically configured to;

recording initial information of the first service in the first time period;

and performing target processing on the initial information to obtain the first data volume information, wherein the target processing comprises feature extraction processing and/or data desensitization processing.

In a particular implementation of the fourth aspect of the embodiments of the present application,

the service record data also comprises a first identifier of the network equipment corresponding to the first data volume information; and/or the presence of a gas in the atmosphere,

In a specific implementation manner of the fourth aspect of the embodiment of the present application, the first service includes one or more of a voice service, a data traffic service, and a real-time traffic service of application software.

In the fourth aspect of the embodiment of the present application, the constituent modules of the communication device may also be configured to execute the steps executed in each possible implementation manner of the second aspect, and specifically refer to the second aspect, which is not described herein again.

A fifth aspect of embodiments of the present application provides a communication apparatus, where the communication apparatus includes a processor, the processor is coupled with a memory, the memory is used for storing a computer program or instructions, and the processor is used for executing the computer program or instructions in the memory, so that the method described in the foregoing first aspect or any one of the possible implementation manners of the first aspect is executed.

A sixth aspect of embodiments of the present application provides a communication apparatus, where the communication apparatus includes a processor coupled with a memory, the memory being configured to store a computer program or instructions, and the processor being configured to execute the computer program or instructions in the memory, so that the method described in any one of the foregoing second aspect or possible implementation manners of the second aspect is performed.

A seventh aspect of embodiments of the present application provides a computer-readable storage medium storing one or more computer-executable instructions, which, when executed by a processor, performs the method according to the first aspect or any one of the possible implementations of the first aspect, or performs the method according to the second aspect or any one of the possible implementations of the second aspect.

An eighth aspect of the embodiments of the present application provides a computer program product (or computer program) storing one or more computers, where when the computer program product is executed by a processor, the processor executes the first aspect or any one of the possible implementation manners of the first aspect, or the processor executes the method of any one of the possible implementation manners of the second aspect or the second aspect.

A ninth aspect of the present embodiment provides a chip system, where the chip system includes a processor, and is configured to support a communication device to implement the first aspect or any one of the possible implementations of the first aspect, or to support the communication device to implement the function related to the second aspect or any one of the possible implementations of the second aspect. In one possible design, the system-on-chip may further include a memory, storage, and a processor for storing necessary program instructions and data for the access network device. The chip system may be constituted by a chip, or may include a chip and other discrete devices.

A tenth aspect of embodiments of the present application provides a communication system including one or more communication apparatuses shown in the third aspect to the fourth aspect described above, or one or more communication apparatuses shown in the fifth aspect to the sixth aspect described above.

The technical effects brought by the third, fifth, seventh to tenth aspects or any one of the possible implementations of the third, fifth, seventh to tenth aspects can be seen in the technical effects brought by the first aspect or different possible implementations of the first aspect.

The technical effects brought by the fourth, sixth, seventh to tenth aspects or any one of the possible implementations of the fourth, sixth, seventh to tenth aspects can be seen in the technical effects brought by the second aspect or different possible implementations of the second aspect.

Drawings

Fig. 1 is a schematic diagram of a network architecture provided in an embodiment of the present application;

fig. 2 is another schematic diagram of a network architecture provided in an embodiment of the present application;

fig. 3 is another schematic diagram of a network architecture provided by an embodiment of the present application;

fig. 4 is a schematic diagram of a communication method provided in an embodiment of the present application;

fig. 5 is another schematic diagram of a communication method provided in an embodiment of the present application;

fig. 6 is another schematic diagram of a communication method provided in an embodiment of the present application;

fig. 7 is another schematic diagram of a communication method provided in an embodiment of the present application;

fig. 8 is a schematic diagram of a communication device according to an embodiment of the present application;

fig. 9 is another schematic diagram of a communication device according to an embodiment of the present application;

fig. 10 is another schematic diagram of a communication device according to an embodiment of the present application.

Detailed Description

The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application.

First, some terms in the embodiments of the present application are explained so as to be easily understood by those skilled in the art.

(1) The terminal equipment: may be a wireless terminal device capable of receiving network device scheduling and indication information, which may be a device providing voice and/or data connectivity to a user, or a handheld device having wireless connection capability, or other processing device connected to a wireless modem.

The terminal devices, which may be mobile terminal devices such as mobile telephones (or "cellular" telephones), computers, and data cards, for example, mobile devices that may be portable, pocket, hand-held, computer-included, or vehicle-mounted, may communicate with one or more core networks or the internet via a Radio Access Network (RAN). Examples of such devices include Personal Communication Service (PCS) phones, cordless phones, Session Initiation Protocol (SIP) phones, Wireless Local Loop (WLL) stations, Personal Digital Assistants (PDAs), tablet computers (pads), and computers with wireless transceiving functions. A wireless terminal device may also be referred to as a system, a subscriber unit (subscriber unit), a subscriber station (subscriber station), a mobile station (mobile station), a Mobile Station (MS), a remote station (remote station), an Access Point (AP), a remote terminal device (remote terminal), an access terminal device (access terminal), a user terminal device (user terminal), a user agent (user agent), a Subscriber Station (SS), a user terminal device (CPE), a terminal (terminal), a User Equipment (UE), a Mobile Terminal (MT), etc. The terminal device may also be a wearable device and a next generation communication system, for example, a terminal device in a 5G communication system or a terminal device in a Public Land Mobile Network (PLMN) for future evolution, etc.

(2) A network device: may be a device in a wireless network, for example, a network device may be a Radio Access Network (RAN) node (or device) that accesses a terminal device to the wireless network, which may also be referred to as a base station. Currently, some examples of RAN equipment are: a new generation base station (gbodeb), a Transmission Reception Point (TRP), an evolved Node B (eNB), a Radio Network Controller (RNC), a Node B (NB), a Base Station Controller (BSC), a Base Transceiver Station (BTS), a home base station (e.g., a home evolved Node B or a home Node B, HNB), a Base Band Unit (BBU), or a wireless fidelity (Wi-Fi) Access Point (AP) in a 5G communication system. In addition, in one network configuration, the network device may include a Centralized Unit (CU) node, or a Distributed Unit (DU) node, or a RAN device including a CU node and a DU node.

The network device can send configuration information (for example, carried in a scheduling message and/or an indication message) to the terminal device, and the terminal device further performs network configuration according to the configuration information, so that network configuration between the network device and the terminal device is aligned; or, the network configuration between the network device and the terminal device is aligned through the network configuration preset in the network device and the network configuration preset in the terminal device. In particular, "alignment" refers to the fact that when an interactive message exists between a network device and a terminal device, the two devices are consistent in understanding the carrier frequency of interactive messaging, the determination of the type of interactive message, the meaning of the field information carried in the interactive message, or other configurations of the interactive message.

Furthermore, the network device may be other means for providing wireless communication functionality for the terminal device, where possible. The embodiments of the present application do not limit the specific technologies and the specific device forms used by the network devices. For convenience of description, the embodiments of the present application are not limited.

The network device may also include a core network device including, for example, an access and mobility management function (AMF), a User Plane Function (UPF), a Session Management Function (SMF), or the like.

In the embodiment of the present application, the apparatus for implementing the function of the network device may be a network device, or may be an apparatus capable of supporting the network device to implement the function, for example, a system on chip, and the apparatus may be installed in the network device. In the technical solution provided in the embodiment of the present application, a device for implementing a function of a network device is taken as an example of a network device, and the technical solution provided in the embodiment of the present application is described.

(3) Traffic volume (voice traffic) is a short term for traffic volume of telecommunications traffic and represents both the load that a telecommunications device is subjected to and the degree to which a user demands for communications. The amount of traffic is related to the number of users, how often the users communicate, the duration of each user communication, and the duration of interest (one minute, one hour, one day, etc.). Since the occurrence of a user call and the amount of time required to complete a communication are both random and varying, traffic is a random variable that varies with time. Traffic prediction (voice traffic forecast) is a mathematical method to obtain the essential data in a telecommunications network, for long-term development planning or for the near term adjustment of the organisation of individual groups of relay circuits.

(4) Big Data (BD) consists of huge datasets, often exceeding the collection, management and processing capabilities of humans at acceptable times. Large data requires special techniques to efficiently process large amounts of data that are tolerant of elapsed time. The method is suitable for the technology of big data, and comprises a large-scale parallel processing database, data mining, a distributed file system, a distributed database, a cloud computing platform, the Internet and an extensible storage system.

(5) Wireless Big Data (WBD) consists of massive information data sets of radio channels of radio base stations and user terminals, physical layer states and link layer states, and its collection, management and processing capabilities under the constraint time of service response.

(6) The terms "system" and "network" in the embodiments of the present application may be used interchangeably. "at least one" means one or more, "a plurality" means two or more. "and/or" describes the association relationship of the associated objects, meaning that there may be three relationships, e.g., a and/or B, which may mean: a alone, A and B together, and B alone, wherein A and B may be singular or plural. The character "/" generally indicates that the former and latter associated objects are in an "or" relationship. "at least one of the following" or similar expressions refer to any combination of these items, including any combination of the singular or plural items. For example, "at least one of A, B, and C" includes A, B, C, AB, AC, BC, or ABC. And, unless specifically stated otherwise, the embodiments of the present application refer to the ordinal numbers "first", "second", etc., for distinguishing between a plurality of objects, and do not limit the order, sequence, priority, or importance of the plurality of objects.

The technical scheme of the embodiment of the application can be applied to the communication system shown in fig. 1. The communication system comprises a core network element 101, an access network device 111 and an access network device 112. Core network element 101 may be connected with access network device 111 and access network device 112. Terminal 121 may communicate with access network device 111. It should be noted that, the core network elements, the base station and the terminal included in the communication system as shown in fig. 1 are only an example, and the interface connection relationship between the base stations is also only an example, and in the embodiment of the present application, the types and the number of the network elements included in the communication system and the connection relationship between the network elements are not limited thereto.

The communication system may be a communication system supporting a fourth generation (4G) access technology, such as a Long Term Evolution (LTE) access technology; alternatively, the communication system may also be a communication system supporting a fifth generation (5G) access technology, such as a New Radio (NR) access technology; alternatively, the communication system may be a communication system supporting a third generation (3G) access technology, such as a Universal Mobile Telecommunications System (UMTS) access technology; or the communication system may also be a communication system of a second generation (2G) access technology, such as a global system for mobile communications (GSM) access technology; alternatively, the communication system may also be a communication system supporting a plurality of wireless technologies, for example, a communication system supporting an LTE technology and an NR technology. In addition, the communication system may also be adapted for future-oriented communication technologies.

Access network device 111 and access network device 112 in fig. 1 may be devices on an access network side for supporting a terminal to access a communication system, and may be, for example, a Base Transceiver Station (BTS) and a Base Station Controller (BSC) in a 2G access technology communication system, a node b (node b) and a Radio Network Controller (RNC) in a 3G access technology communication system, an evolved node b (eNB) in a 4G access technology communication system, a next generation base station (next generation node b, gNB) in a 5G access technology communication system, a Transmission Reception Point (TRP), a relay node (relay node), an Access Point (AP), and the like.

The core network element 101 in fig. 1 may control one or more access network devices, or perform unified management on resources in the system, or may configure resources for the terminal. For example, the core network element may be a serving General Packet Radio Service (GPRS) support node (SGSN) or a Gateway GPRS Support Node (GGSN) in a 3G Access technology communication system, a Mobility Management Entity (MME) or a Serving Gateway (SGW) in a 4G Access technology communication system, an Access and Mobility Management Function (AMF) or a User Plane capability (UPF) network element in a 5G Access technology communication system, or the like.

The core network element 101 may be specifically an edge computing server, a gateway server, or other device types.

The terminal 121 in fig. 1 may be a device that provides voice or data connectivity to a user, and may also be referred to as User Equipment (UE), mobile station (mobile station), subscriber unit (subscriber unit), station (station), Terminal Equipment (TE), and so on. The terminal may be a cellular phone (cellular phone), a Personal Digital Assistant (PDA), a wireless modem (modem), a handheld device (hand-held), a laptop computer (laptop computer), a cordless phone (cordless phone), a Wireless Local Loop (WLL) station, a tablet (pad), or the like. With the development of wireless communication technology, all devices that can access a communication system, can communicate with a network side of the communication system, or communicate with other objects through the communication system may be terminals in the embodiments of the present application, such as terminals and automobiles in intelligent transportation, home devices in smart homes, power meter reading instruments in smart grid, voltage monitoring instruments, environment monitoring instruments, video monitoring instruments in smart security networks, cash registers, and so on. In the embodiment of the present application, the terminal may communicate with an access network device, such as access network device 111 or access network device 112. Communication may also be performed between multiple terminals. The terminals may be stationary or mobile.

As shown in fig. 1, one exemplary implementation of a network architecture implemented for a communication network. Traffic modeling and prediction has been an important issue in mobile communication networks. In the face of increasing number of users and traffic load, operators must meet Quality of Service (QoS) requirements of users. For example, the spatial distribution density of the user terminals will likely be greater than 10⁶User/squareKilometers; the spatial distribution density of the radio base station will likely be greater than 10⁴One per square kilometer; the radio capacity requirements of the area space will likely be greater than 1Tbps per square kilometer. Although there are currently analyses of the regularity and randomness of the communication patterns in millions of cellular base stations, the prediction and application of traffic is still limited, which will significantly increase the operating costs of the cellular base stations. Particularly, the application focuses on and hopes to solve the problems that the flow prediction lacks theoretical support, the existing prediction model is high in calculation complexity, low in prediction accuracy and precision and the like.

The conventional traffic service prediction methods specifically include the following two ways.

In the first mode, the early network data transmission quantity is small, the application is single, the network flow is described by using the classical poisson process by usually taking the reference of a service model of the traditional telecommunication network, and the method has better performance. With the development of telecommunication services and network structures, the poisson process cannot sufficiently reflect service traffic characteristics, and a Markov (Markov) model, an autoregressive model (a linear model including LR, ARMA, ARIMA, SARIMA), a kalman filter model and other models are gradually introduced to describe network traffic, which is called a traditional network traffic model.

With the aging of Machine learning and data mining algorithms and their powerful performance in various fields, many researchers have used them for predicting network traffic in recent years, such as Support Vector Machines (SVMs) applied to the regression problem of processing time series; BP neural network, Elman neural network, etc. are applied to the prediction of network traffic. In addition, various improved schemes are provided for different data set characteristics, such as a core-based flow prediction algorithm under a classification model and a Boosting-based flow prediction algorithm. The neural network is a nonlinear method, which adopts a nonlinear activation function and has one or more hidden layers, and two adjacent layers are connected through a weight. The BP algorithm adopts a gradient descent method to adjust the weight value so as to minimize the error between the actual output and the expected output of the network. In network traffic prediction, the next time traffic is the desired output of the target, the previous traffic information is the input, the network parameters are adjusted, and the neural network is trained using a large amount of labeled training data. Because the flow data is a time sequence and has dynamic characteristics related to the front and the back, the prior document uses an Elman neural network to replace a BP neural network to obtain better effect. The difference of the Elman neural network compared with BP is that the output of the hidden layer of the Elman network is fed back to the input layer to be used as the input of the next network, and based on the characteristic, the Elman neural network can better capture the dynamic characteristic of the time sequence, and thus can be better adapted to the prediction of the time sequence.

The disadvantages of the flow prediction model in the procedure shown in the first embodiment specifically include the conventional model and the machine learning model. Among them, the conventional network traffic model has disadvantages in that the described service sequence has a Short Range Dependency (SRD), and cannot describe a Long Range Dependency (LRD) of the network, and as the number of services increases, it cannot reflect its burstiness. Currently, the accepted, most important statistical feature of network traffic is self-similarity at large timescales. Self-similarity (self-similarity) means that the local structure has a certain degree of consistency compared to the global structure. The long correlation of network traffic is relative to the short correlation model such as poisson. In a physical sense, the long correlation reflects a persistent phenomenon in the self-similarity process, i.e. a phenomenon in which burst characteristics exist on all time scales. The higher the degree of self-similarity (long correlation), the stronger the burstiness. Although the model after the poisson process introduces correlation in a random process, the burstiness of the service can be captured to a certain extent, but the model only has short correlation, only can predict the recent traffic of the network, and cannot describe the long correlation of the network. And when the number of service sources increases, the aggregated service tends to be smooth and cannot reflect the service burstiness.

Various machine learning algorithms are suitable for describing the instability of network traffic, but enough training data is ensured, the calculation amount is usually large, and problems of overfitting, local optimization and the like can exist in the training process. When a machine learning method is adopted, the cost of precision improvement is relatively limited compared with that of the precision improvement.

At present, relevant methods and conclusions exist about the regularity and the distribution characteristics of the flow pattern, which proves that the flow pattern is predictable. The flow density in the spatial domain may be approximated by a lognormal distribution or a weibull distribution. The dynamics of the flow are shown to follow a trimodal distribution, i.e. a combination of complex exponential, power law and exponential distributions, in both the spatial and temporal dimensions. Through time sequence analysis, the network flow dynamics of the base station can be simulated, and the regularity and the randomness of the flow mode of the base station can be represented.

Prediction theory in another field of mobility prediction will be introduced here. In the mobility prediction, for a time sequence containing position information, three kinds of entropies are derived for measuring the uncertainty of the position sequence.

Three entropies for a user who has visited N places are as follows:

(1) random entropy

Ni is the number of different sites visited by the user i, if the user visits the sites with the same probability, the random entropy can express the predictability of the user position;

(2) time-independent entropy

p_i(X_j) Is based on historical records of user i visiting place X_jThe probability of (d);

(3) true entropy

Described by entropy rate of a random process, which not only depends on access frequency, but also is related to the access sequence of nodes and the time of stay of each place, so that complete space-time information in complete individual movement patterns can be obtained. By T_i＝{X₁,X₂,...,X_nDenotes the sequence of positions visited by user i in successive times of the observation period,

is defined as

Wherein, S (X)₁,X₂,...,X_n) For joint entropy, S (X)_i|T_i-1) The conditional entropy of the next state is obtained for the known history sequence. Computing true entropy directly with definitions

The complexity is high, and an entropy value can be estimated by using an entropy rate estimation method based on Lempel-Ziv data compression.

To characterize the predictability of the user group itself, S is defined for each user_i,

The distribution P (S) thus obtained_i),

A user with entropy S moves among N places, and the average predictability satisfies pi ≦ pi after deducing^max(S, N), wherein Π^maxSatisfying the following equation. II type^maxI.e. an upper bound on the accuracy of the prediction.

S＝H(Π^max)+(1-Π^max)log₂(N-1)

H(Π^max)＝-Π^max log₂(Π^max)-(1-Π^max)log₂(1-Π^max)

Calculating three types of pi of the user according to the three types of entropies of the user, and further obtaining a distribution result P (pi) of the user^max),P(Π^unc),P(Π^rand). Wherein the true entropy

The method has the advantages of having the most research value, containing the most information and the least uncertainty and calculating II^maxAt the maximum, it is an upper bound on the prediction accuracy that can be obtained using the optimal prediction model in the ideal case. For a target object with an upper predictability bound of 0.5, this means that only half the time is expected to correctly predict the user's next step location.

However, in the implementation of the second mode, the flow pattern is proved to be predictable, but the accuracy of the existing flow prediction lacks quantitative explanation. For example, some data sets have a high prediction accuracy and some data sets have a low prediction accuracy. How to determine the degree of predictability of the data set by analyzing the flow patterns, and the like. In addition, there are related theories of true entropy and predictability in the field of mobility prediction, but how to apply to the field of traffic prediction still has some technical difficulties. First, for network traffic that approximates a continuous sequence, how to compute true entropy and an upper predictability bound. Secondly, the position information belongs to the information of an application layer, the position sequence presents short correlation, and whether the measurement standard in the mobility prediction is suitable for the traffic prediction problem with long correlation of a network layer is not verified. Finally, in the case that the predictability upper bound index is proved to be effective, the method can be used for solving what problem of flow prediction and how to implement the flow prediction.

Through the analysis of the first mode, in the existing prediction model: the service sequence described by the traditional flow prediction model has the problems that the flow self-similarity cannot be accurately described, the dynamic characteristics of the flow are captured and the like; the machine learning method needs a large amount of training data and calculated amount to ensure performance, the cost of the brought improvement precision is relatively limited compared with the cost, and the problems of overfitting, local optimization and the like exist during model optimization. Therefore, the problem that the prediction precision and the calculation complexity are difficult to guarantee simultaneously in the current flow prediction method is solved, and the business processing with high prediction performance and low calculation cost is realized by combining the quantization of big data analysis and the predictability index with the classic prediction model.

Through the analysis of the second mode, in the existing flow pattern analysis, only the regularity of the flow pattern can be qualitatively analyzed, and the predictable degree of the flow pattern cannot be quantitatively analyzed. Therefore, the invention aims to solve the problem that theoretical analysis and explanation on the prediction effect are difficult, and a method and a link for analyzing the flow prediction by introducing scientific measurement are provided.

In order to solve the above problems, the present application provides several ideas for optimization, including the following.

1, thinking: a flow quantification method is proposed to discretize a continuously varying flow and to approximate a continuous value of the flow (or a large number of possible discrete values) to a finite number (or less) of discrete states. The complexity of the state space, each state in the state space representing a traffic interval, and the accuracy of describing the traffic characteristics are balanced by selecting different quantization parameter values. And (4) conclusion: the larger the quantization granularity, the fewer the number of states, the lower the model complexity, and the higher the prediction accuracy, but the larger the error margin compared to the actual speech traffic.

And (2) thinking: the basic theory based on the information entropy is continued, the real entropy concept in the time sequence is introduced, the uncertainty of the flow state sequence after the user quantization is represented, the entropy and the upper range of the predictability of the voice flow are obtained, and the accuracy of the voice flow prediction is expected to be explained under the support of a stronger theory. The conclusion is that the predictability upper bound in the traffic prediction is an effective measure and can provide theoretical guidance for voice traffic prediction.

And (3) thinking: a prediction method based on entropy and quantization is provided, and under the condition of a given error range and accuracy requirements, an optimal quantization parameter is obtained, so that the lowest prediction error is realized. The predictive algorithm may employ a multi-order markov model similar to the true entropy mathematical theory. The optimal order of the model can be determined by analyzing the dynamics of the traffic data and the autocorrelation order. The disadvantage of the markov model is that it is not possible to describe the long dependencies of the network, and the dependencies utilized tend to be more long dependencies when using higher order models, meaning that the current state is affected by more previous states. And (4) conclusion: t is adjusted by observing predictability indexes, the realized precision is superior to the optimal performance of the existing model, and the calculation complexity is far lower than that of other models.

Specifically, under the framework of wireless big data, the idea 1 is adopted to construct a flow state space of a prediction object, and the idea 2 measures uncertainty of flow state change of the prediction object. The upper predictability bound obtained by concept 2 can guide the selection of the quantization parameters in concept 1. The prediction method of the idea 3 is based on the cooperation of the idea 1/2, so that the processing complexity of the flow prediction task can be reduced, the processing time can be reduced, and the prediction task can be guaranteed to be completed within a constrained error range. Of course, the above-mentioned concept 1/2/3 is only a simplified introduction for the most intuitive, and the practical method of the present invention also needs to be specifically designed to combine with the entities of the wireless communication system to realize an adaptive task system, which is far from the simplicity described in the concept 1/2/3.

The present application will be described below with different angles.

Please refer to fig. 2, which is a schematic diagram of a network architecture of a communication system according to an embodiment of the present application. In the communication system, the first network device is taken as an edge computing server, and the second network device is taken as a base station for example, it is obvious that, as described above, the first network device and the second network device may also be implemented by other entity devices, and this is only an exemplary description here.

As shown in fig. 2, the entity objects included in the communication system include a user side (100), a base station side (200) and an edge computing server (300), as shown in fig. 2. All base stations under the management range are collectively called as a base station side, and each base station in the base station side provides service for users under the coverage area of the base station. Each user terminal internally comprises a communication module (101), a calculation module (102) and a storage module (103); each base station internally comprises a network service module (201), a data acquisition module (202), a charging system (203), a data processing module (204) and a power control module (205); the edge computing server internally comprises a communication module (301), a computing module (302) and a storage module (303). The base station side and the user side communicate through radio waves and provide services, and the base station side and the edge server directly communicate through optical fibers and transmit data and signaling.

Wherein the edge computing server provides cloud computing capabilities in the mobile network to reduce resource management operations and service delivery latency. The Edge Computing server is deployed by adopting a Multi-Access Edge Computing (MEC) architecture. The MEC system is divided into two levels, a host level and a system level, according to the definition of the European Telecommunications Standards Institute (ETSI). The main machine deployment of the MEC adopts a hierarchical DC computer room layout mode with a central Data Center (DC, Data Center) (large-area central computer room), a regional DC (provincial level computer room), a core DC (local network core computer room), an edge DC (local network convergence computer room), a receiving station DC and a base station computer room as basic architectures. Generally, an MEC host and a Control Unit (CU) share a room, and the MEC host is deployed behind a base station, so that data services are closer to users. The MEC system level network manager needs to coordinate operations (such as selecting hosts, application migration, policy interaction, etc.) between different MEC hosts and between the hosts and 5 GCs (5G C-V2X, 5G Cell-Vehicle to X), and is generally deployed at regional DC (provincial level) or central DC (large district center). The MEC device generally has a Content Delivery Network (CDN) function, and compared with a conventional CDN, the MEC is closer to a radio access Network and is located deeper.

Further, taking the network architecture shown in fig. 3 as an example, after receiving the traffic data from the data processing module (204) of each base station at the base station side (200), each level of MEC host of the edge computing server (300) forwards the traffic data to the MEC system level network manager (management server). After receiving the traffic data of all base stations, the System level network manager stores the data in a storage module (303) of an edge computing server (300), namely a Distributed File System (HDFS) of the data center. Only the modules relevant to the present application are listed here, although in reality the individual entities are more complex internally.

Taking the network architecture of the communication system shown in fig. 2 and fig. 3 as an example, the following will describe a communication method and related devices provided in the embodiments of the present application.

Referring to fig. 4, a schematic diagram of a communication method according to an embodiment of the present application is shown, where the method includes the following steps.

S101, the first network equipment receives service record data from the second network equipment, wherein the service record data comprises first data volume information.

In this embodiment, a first network device receives service record data from a second network device, and correspondingly, the second network device sends the service record data to the second network device in step S101, where the service record data includes first data volume information, and the first data volume information is used to indicate a data volume of a first service occurring in a first time period.

In a specific implementation manner, before the second network device sends the service record data to the first network device in step S101, the method further includes: the second network equipment records initial information of the first service in the first time period; and the second network equipment performs target processing on the initial information to obtain the first data volume information, wherein the target processing comprises feature extraction processing and/or data desensitization processing. Specifically, before the second network device sends the service record data to the first network device, the method may further include: the second network device records initial information of the first service occurring in the first time period, and performs target processing on the initial information to obtain the first data volume information, wherein the target processing includes feature extraction processing and/or data desensitization processing. Wherein, the characteristic processing is used for extracting effective information in the initial information, and the data desensitization processing is used for protecting the privacy of the user, thereby further optimizing the scheme.

In addition, the service record data further includes a first identifier of the network device corresponding to the first data volume information; and/or the service record data further comprises a second identifier of the terminal device corresponding to the first data volume information. That is to say, the first data amount information may specifically indicate a data amount of the first service occurring in the first time period in a part of terminal devices (or all of the terminal devices) provided with the service by the second network device, and/or the first data amount information may specifically indicate a data amount of the first service occurring in the first time period in a specific terminal device provided with the service by the second network device, and these two implementation manners may be specifically distinguished by the first identifier and the second identifier, so as to provide multiple implementable manners of the scheme.

In a specific implementation manner, the first service may specifically include one or more of a voice service, a data traffic service, and a real-time traffic service of application software, and multiple implementation manners of the first service are provided while the implementability of the scheme is improved.

The process of determining the first data amount information by the second network device will be described below by a specific example. Taking the network architecture shown in fig. 2 as an example, the first service is a voice service, and the first data volume is traffic volume data corresponding to the voice service. Wherein each wireless base station has a data acquisition device (202) therein as a medium from the network service (201) to the billing system (203).

The network services (201) module provides services to the user side (100) and confirms these services in Call Detail Records (CDR). The original CDR data also contains the number of calls and accesses and some identification codes. These identification codes can identify the date, start time and duration of the call and the serving base station, call characteristics, and how to get through.

After the data acquisition device (202) identifies and confirms the chargeable call, the corresponding CDR data is submitted to a charging system (203) for storage. CDR data of the base station enters a data processing module (204) from a charging system (203), firstly, feature processing is carried out, and key fields of User identification marks (User ID), starting call Time (Time/Day), call Duration (Duration), base station identification marks (BSID) and the like of each record are reserved; secondly, desensitizing the user ID to protect the privacy of the user; finally, the processed CDR data set is converted into a traffic prediction format, as shown in fig. 5. In the process of (a) in fig. 5, the voice service data form of the base station is grouped according to User ID and converted into the voice call detail form of the User; in the process of fig. 5 (b), the form of processing a single user is expanded, and the voice call records are arranged in time sequence, and the call duration is based on the unit of seconds(s). After processing, the CDR data in the traffic prediction format is transmitted from the data processing module (204) of the base station to the edge computing server (300).

S102, the first network equipment carries out quantization processing on the service record data according to the target quantization parameter to obtain first service state information.

In this embodiment, the first network device performs quantization processing on the service record data received in step S101 according to the target quantization parameter, so as to obtain first service state information.

In a specific implementation manner, before the first network device performs quantization processing on the service record data according to the target quantization parameter to obtain the first service state information in step S102, the method further includes: the first network equipment carries out quantization processing on the service record data according to the initial quantization parameter to obtain initial service state information; the first network equipment determines a predictability upper bound value corresponding to the initial service state information; and the first network equipment updates the initial quantization parameter according to the predictability upper bound value and the initial prediction precision to obtain the target quantization parameter. Specifically, the first network device may further include an updating process of the quantization parameter before performing quantization processing on the service record data according to the target quantization parameter to obtain the first service state information. The process specifically includes the steps of carrying out quantization processing on the service record data according to preset initial quantization parameters to obtain initial service state information, then determining a predictability upper bound value corresponding to the initial service state information, and further updating the initial quantization parameters according to the predictability upper bound value and initial prediction precision to obtain the target quantization parameters. The upper predictability bound in the flow prediction process is effective measurement, theoretical guidance can be provided for flow prediction, the optimal quantization parameter is determined as a target quantization parameter according to the set updating process, and then the target quantization parameter is used for predicting the flow state so as to ensure the prediction accuracy.

Specifically, the determining, by the first network device, the predictability upper bound value corresponding to the initial service state information includes: after the first network device determines the first entropy corresponding to the initial service state information, the first network device determines the upper predictability threshold according to the first entropy. Specifically, in the process of determining the predictability upper bound value, the predictability upper bound value may be specifically determined by a first entropy value corresponding to the initial service state information. The method comprises the steps of carrying out quantitative analysis on initial service state information, carrying out quantitative analysis on the initial service state information, and carrying out quantitative analysis on the initial service state information.

S103, the first network device takes the first service state information as an input of a preset model, and outputs second service state information of the first service in a second time period through processing of the preset model.

In this embodiment, the first network device uses the first service state information obtained by the processing in step S102 as an input of a preset model, and outputs the second service state information of the first service in the second time period through the processing of the preset model.

In a specific implementation manner, in step S103, after the first network device uses the first service state information as an input of a preset model and outputs second service state information of the first service in a second time period through processing of the preset model, the method further includes: and the first network equipment performs inverse quantization processing on the second service state information according to the target quantization parameter to obtain second data volume information, wherein the second data volume information is used for indicating the data volume of the first service in the second time period. Specifically, after the first network device obtains the second service state information through processing output of a preset model, the second service state information may further be subjected to inverse quantization processing according to the target quantization parameter, so as to obtain second data volume information, where the second data volume information is used to indicate a data volume of the first service occurring in the second time period, that is, the output of the model is represented in the second data volume information by continuously changing traffic data, and a specific implementation form of the output of the model is provided, so as to more intuitively represent the output of the model.

In a specific implementation manner, after step S103, the method further includes: the first network device sends the second data volume information to the second network device. Specifically, the first network device may further send, to the second network device, the second data volume information for indicating the data volume occurring in the first service in the second time period, which may be applied to design processes of resource reservation, energy allocation, base station dormancy, and the like of the second network device.

Referring to fig. 6, another implementation diagram of a communication method provided in the embodiment of the present application is shown, and in fig. 6, the implementation process of step S102 and step S103 is still described by taking the first network device as an example of the edge computing server (300) in the network architecture shown in fig. 2 and fig. 3. Step S102 and its optional implementation correspond to

steps

3021 and 3022 in fig. 6, and step S103 and its optional implementation correspond to the remaining steps shown in fig. 6. The workflow of predicting the mobile phone traffic by the edge computing server (300) by using the CDR data in the base station side traffic prediction format in the storage module (303) specifically comprises the following steps.

Step 3021: a calculation module (302) of an edge calculation server (300) reads out flow prediction data of all base stations in a management range from an HDFS file system, and the flow prediction data are classified and aggregated according to UserID and converted into a complete voice call list of a single user.

Step 3022: the voice traffic data for a single user is quantized as shown in fig. 7. Arranging voice call data of a single user in a time sequence; and secondly, integrating according to a fixed time period (such as a day, an hour or other implementations) to obtain the traffic data at a fixed sampling interval. The process of converting a single user's call record into daily voice traffic is illustrated in fig. 7, using day-to-day integration as an example. And finally, generating a state space according to a quantization scale T, wherein T represents a quantization interval, and each state in the state space represents a traffic volume range. The larger T, the fewer the number of states, and the larger the quantization granularity, which results in a higher accuracy of the predicted traffic state, but a larger error margin compared to the actual speech traffic.

Step 3023: after the traffic is quantized and mapped to a specified state space, the user's traffic time series has only a limited number of states. Therefore, the entropy and predictability of the random state of the user traffic when the historical records are known can be calculated according to the basic theory of entropy.

Specifically, if T_i＝(X₁,D₁)→(X₂,D₂)→…→(X_j,D_j) Sequence of transitions representing traffic state during observation period for user i, where X_jDenotes the D th_jThe traffic state of the day, then

Wherein p (T'_i) Is at the track T_iOf (c) to find a specific sequence T'_iThe probability of (c). After obtaining the value of the true entropy, the following equation can be solved to obtain the corresponding upper bound of predictability.

S^real＝H(Π^max)+(1-Π^max)log₂(N-1)

H(Π^max)＝-Π^max log₂(Π^max)-(1-Π^max)log₂(1-Π^max)

Step 3024: the required prediction accuracy is recorded as A, the allowable error range is recorded as E, and the predictability is recorded as pi^max. A and E are values set by human, and

will change as T changes. Step 3023 has already calculated the predictability of all users at the current quantization scale T, and compares it with the set expected accuracy requirement, and adjusts the size of the T value control state space and the complexity of the model. The optimization process can be expressed as minT, objectto

Wherein

And finally, determining the optimal quantization parameter T by representing the average predictability of the user group.

Step 3025: and predicting the quantized user telephone traffic data by using a specific prediction algorithm. Models with better prediction performance can be used, such as differential integrated moving average autoregressive (ARIMA) models, SVMs, and the like. Multi-order Mahalanobis Chains (MC) are recommended because the mathematical theory and entropy concept of Markov models are based on the state transitions in the mining history and the training cost is low. In the MC-based model, the traffic trajectory for each user is modeled as an n-order Markov chain, with future states depending only on the order of the preceding n states, i.e.

P(X^t+1＝x^t+1|X^t＝x^t,...,X¹＝x¹)＝P(X^t+1＝x^t+1|X^t＝x^t,...,X^t-n+1＝x^t-n+1)

Wherein X^tAnd expressing the user traffic state at the time t, then determining the prediction through a transition matrix, and selecting the target state with the highest probability.

Step 3026: and inverse quantization of quantization is realized on the predicted result, namely the predicted result is multiplied by the quantization interval T to obtain the telephone traffic of the user at the future moment, and the predicted result is stored in the storage module (303).

The prediction result output by the edge computing server (300) is fed back to the power control module (205) at the base station side, and the method can be used for the design of resource reservation, energy allocation, base station dormancy and the like of the base station.

Specifically, as an extended application of the foregoing method, the edge computing server provided in this application uses the traffic data at the base station side to predict the work flow of the mobile phone traffic, which can be used for prediction processing of the traffic state of the user. In addition, the present embodiment is not limited to predicting only the traffic volume of the user, and may predict the traffic volume of the radio base station. At this time, the data transmitted to the edge computing server (300) by the base station side data processing module (204) of the original embodiment is changed from the CDR data in the traffic prediction format grouped according to the USERID to the CDR data in the traffic prediction format grouped according to the BSID, and meanwhile, the object of the prediction step (3025) in the quantization step (3024) in the workflow is changed from the traffic of the user to the traffic of the base station, and the implementation process is similar to the implementation process shown in fig. 4 to fig. 7, and is not described here again.

In summary, in the communication method shown in fig. 4 to fig. 7, the deployment edge computing server collects, manages and processes the relevant information of the telecommunication service between the plurality of wireless base stations and the huge number of user terminals, obtains the traffic state with a limited value by using the traffic quantization method to disperse voice traffic, and estimates the degree of predictability of the traffic state by using the upper bound index of predictability. The method comprises the steps of calculating the upper predictability bound of traffic states under different quantizations, determining the optimal quantization parameter according to the set performance index and the designed optimization process, predicting the optimal quantized traffic state, and reducing the error range of predicted traffic under the condition of ensuring the accurate probability of prediction. The process specifically includes the following key implementation points.

Key point 1: an implementation of network layer traffic prediction is provided.

User side (100): predicted object, source: a service request is issued.

Base station side (200): the data processing module (204) processes CDR data acquired in the process of providing service for the user side (100) into a flow prediction format and sends the flow prediction format to the edge computing server (300); and the power control module (205) receives the prediction result output by the edge computing server (300) for resource reservation and energy efficiency optimization.

Edge computing server (300): traffic data is generated (3021), user traffic data is quantized (3022), predictability analysis (3023), quantization parameter tuning (3024), and results are predicted output (3025, 3026).

Illustratively, in the implementation step (3022), a quantization method of user traffic data is proposed and discussed, and the numerical change prediction problem is consolidated into a state transition through quantization. In the implementation step (3023), entropy and predictability are used for measurement, and an upper predictability bound of the traffic data is obtained. The method not only enables the prediction performance of the model to be theoretically explained, but also can predict the prediction accuracy which can be realized by the data set and guide the prediction. Has the advantages that: in step (3024), by comparing the predictability under different quantization scales and adjusting according to the set expected accuracy requirement, the optimal quantization scale T is obtained, and the error range of the predicted flow state is reduced as much as possible.

Key point 2: traffic state prediction data formats based on CDR data are provided.

Specifically, the method comprises the following steps: and a data processing module (204) of the base station performs characteristic processing and User privacy desensitization processing on CDR data recorded by the base station, groups the call data according to User IDs and arranges the call data according to time sequence to generate flow prediction data of a single User. Step (3021, 3022) classifying and aggregating the User traffic prediction data of all base stations according to User IDs, converting the User traffic prediction data into a complete voice call list of a single User on all base stations, integrating call duration according to a fixed time period (e.g., one day, one hour), obtaining traffic data at a fixed sampling interval, generating a state space according to a quantization scale T, and generating a quantized traffic state prediction data format.

Illustratively, the method comprises the steps of calculating a predictability upper bound of a quantized traffic state, adjusting a quantization scale by combining a designed optimization process, determining an optimal quantization parameter, and predicting quantized traffic data by adopting an entropy-based traditional model, wherein the prediction performance is superior to the best performance reported at present and superior to an unexplainable neural network prediction model which needs a large amount of training data and has high computational complexity. Has the advantages that: when the performance is tested, the telephone traffic state of 20 ten thousand mobile phone users is predicted by adopting a Markov model based on quantification and entropy. During the training process, the performance of the low-order model is poor and far reaches the upper range of predictability. But as the order increases, the predictive performance becomes better. When the order reaches 25, the upper predictability bound for optimal quantization can be approached.

Key point 3-a traffic prediction process based on entropy and quantization is provided.

The upper limit value of the predictability of the traffic state of the prediction object can be obtained by calculating the entropy of the quantized traffic state, the predictability results of different quantizations are dynamically adjusted according to the expected precision value set by the task, the optimal quantization parameter is determined, and the traditional model based on the entropy is adopted to predict the traffic state after the optimal quantization.

The prediction process can predict the telephone traffic of the mobile phone of the user and can also predict other wireless state information. Therefore, the aim of reducing the calculation complexity and the processing time of the prediction task is fulfilled under the condition of ensuring certain prediction precision.

In addition, although the method is introduced in the embodiment of the present application with the goal of predicting the traffic volume of the mobile phone user, the method of the present technical solution may also be extended to predict other state information of the base station side and the user terminal, such as predicting the traffic volume of the wireless base station, predicting the data traffic occurring in the network service, and predicting the real-time traffic of the application software in the wireless network scenario. The prediction of a single user can be used for improving the service quality of personalized services and key tasks and can also be combined into global requirements; and aiming at the prediction of the base station, the prediction unit is the whole telephone traffic of the base station, and the global resource scheduling is emphasized. Therefore, the method can be used for solving the problems of planning optimization, fault early warning, resource coordination and allocation, real-time processing of micro base station services and the like of the future network.

In this embodiment, a first network device receives service record data from a second network device, where the service record data includes first data volume information, and the first data volume information is used to indicate a data volume of a first service occurring in a first time period; then, the first network device carries out quantization processing on the service record data according to the target quantization parameter to obtain first service state information; and then, the first network equipment takes the first service state information as the input of a preset model, and outputs second service state information of the first service in a second time period through the processing of the preset model. The input data input into the prediction model and used as the input of the preset model for predicting to obtain the second service state information of the first service in the second time period comprises the first service state information, and the first service state information is obtained by quantizing the service record data. That is to say, the corresponding continuous value (or a large number of possible discrete values) of the first data volume information included in the service record data is approximated to a finite number (or less) of discrete states, and the obtained first service state information enables continuously-changing traffic data in the first data volume information to be converted into the first service state information, which can be used for describing long correlation of a network and improving the prediction effect of a prediction model on the traffic service of a long-term network.

The embodiments of the present application have been described above from the perspective of methods, and the signal detection apparatus in the embodiments of the present application will be described below from the perspective of specific apparatus implementation.

Referring to fig. 8, an embodiment of the present application provides a communication apparatus 800, including:

a transceiving unit 802, configured to receive service record data from a second network device, where the service record data includes first data volume information, and the first data volume information is used to indicate a data volume of a first service occurring in a first time period;

the processing unit 801 is configured to perform quantization processing on the service record data according to the target quantization parameter to obtain first service state information;

the processing unit 801 is further configured to use the first service state information as an input of a preset model, and output second service state information of the first service in a second time period through processing of the preset model.

In a specific implementation, the processing unit 801 is further configured to:

In a specific implementation manner, the processing unit 801 is specifically configured to:

In a specific implementation manner, the processing unit 801 is further configured to perform dequantization processing on the second service status information according to the target quantization parameter to obtain second data size information, where the second data size information is used to indicate a data size of the first service occurring in the second time period

In a specific implementation manner, the transceiving unit 802 is further configured to:

and sending the second data volume information to the second network equipment.

In one particular implementation of the method of the invention,

In a specific implementation manner, the first service includes one or more of voice service, data traffic service, and real-time traffic service of application software.

It should be noted that, for details of the information execution process of the units of the communication apparatus 800, reference may be specifically made to the description of the foregoing method embodiments in the present application, and details are not described here again.

Referring to fig. 9, an embodiment of the present application provides a communication apparatus 900, which includes a processing unit 901 and a transceiver unit 902:

the processing unit 901 is configured to send service record data to the first network device through the transceiving unit 902, where the service record data includes first data volume information, and the first data volume information is used to indicate a data volume of the first service occurring in a first time period;

the processing unit 901 is further configured to receive, through the transceiving unit 902, second data volume information from the first network device, where the second data volume information is used to indicate a data volume of the first service occurring in the second time period, and the second data volume information is obtained by performing quantization processing and dequantization processing on the service record data.

In a specific implementation, the processing unit 901 is specifically configured to;

recording initial information of the first service in the first time period;

In one particular implementation of the method of the invention,

In a specific implementation manner, the first service includes one or more of a voice service, a data traffic service, and a real-time traffic service of application software.

It should be noted that, for details of the information execution process of the units of the communication apparatus 900, reference may be specifically made to the description of the method embodiments described above in this application, and details are not described here again.

Referring to fig. 10, a schematic structural diagram of a communication device according to the foregoing embodiment is provided in an embodiment of the present application, where the communication device may specifically be the communication device according to the foregoing embodiment, and the structure of the communication device may refer to the structure shown in fig. 10.

The communication device includes at least one processor 1011, at least one memory 1012, at least one transceiver 1013, at least one network interface 1014, and one or more antennas 1015. The processor 1011, the memory 1012, the transceiver 1013 and the network interface 1014 are connected, for example, by a bus, and in this embodiment, the connection may include various interfaces, transmission lines or buses, which is not limited in this embodiment. An antenna 1015 is connected to the transceiver 1013. The network interface 1014 is used to enable the communication apparatus to connect with other communication devices via a communication link, for example, the network interface 1014 may include a network interface between the communication apparatus and a core network device, such as an S1 interface, and the network interface may include a network interface between the communication apparatus and other network devices (such as other access network devices or core network devices), such as an X2 or Xn interface.

The processor 1011 is mainly used for processing a communication protocol and communication data, controlling the entire communication apparatus, executing a software program, and processing data of the software program, for example, to support the communication apparatus to perform the actions described in the embodiments. The communication device may include a baseband processor and a central processing unit, the baseband processor is mainly used for processing a communication protocol and communication data, and the central processing unit is mainly used for controlling the entire network device, executing a software program, and processing data of the software program. The processor 1011 in fig. 10 may integrate the functions of a baseband processor and a central processing unit, and those skilled in the art will understand that the baseband processor and the central processing unit may also be independent processors, and are interconnected through a bus or the like. Those skilled in the art will appreciate that a network device may include multiple baseband processors to accommodate different network formats, multiple central processors to enhance its processing capabilities, and various components of the network device may be connected by various buses. The baseband processor can also be expressed as a baseband processing circuit or a baseband processing chip. The central processing unit can also be expressed as a central processing circuit or a central processing chip. The function of processing the communication protocol and the communication data may be built in the processor, or may be stored in the memory in the form of a software program, and the processor executes the software program to realize the baseband processing function.

The memory is used primarily for storing software programs and data. The memory 1012, which may be separate, is coupled to the processor 1011. Alternatively, the memory 1012 may be integrated with the processor 1011, for example, within one chip. The memory 1012 can store program codes for implementing the technical solution of the embodiment of the present application, and the processor 1011 controls the execution of the program codes, and various executed computer program codes can also be regarded as drivers of the processor 1011.

Fig. 10 shows only one memory and one processor. In an actual network device, there may be multiple processors and multiple memories. The memory may also be referred to as a storage medium or a storage device, etc. The memory may be a memory element on the same chip as the processor, that is, an on-chip memory element, or a separate memory element, which is not limited in this embodiment.

The transceiver 1013 may be used to support reception or transmission of radio frequency signals between the communication device and the terminal, and the transceiver 1013 may be connected to the antenna 1015. The transceiver 1013 includes a transmitter Tx and a receiver Rx. Specifically, one or more antennas 1015 may receive radio frequency signals, and the receiver Rx of the transceiver 1013 is configured to receive the radio frequency signals from the antennas, convert the radio frequency signals into digital baseband signals or digital intermediate frequency signals, and provide the digital baseband signals or digital intermediate frequency signals to the processor 1011, so that the processor 1011 performs further processing on the digital baseband signals or digital intermediate frequency signals, such as demodulation processing and decoding processing. In addition, the transmitter Tx in the transceiver 1013 is also configured to receive a modulated digital baseband signal or a digital intermediate frequency signal from the processor 1011, convert the modulated digital baseband signal or the digital intermediate frequency signal into a radio frequency signal, and transmit the radio frequency signal through the one or more antennas 1015. Specifically, the receiver Rx may selectively perform one or more stages of down-mixing and analog-to-digital conversion processes on the rf signal to obtain a digital baseband signal or a digital intermediate frequency signal, wherein the order of the down-mixing and analog-to-digital conversion processes is adjustable. The transmitter Tx may selectively perform one or more stages of up-mixing and digital-to-analog conversion processes on the modulated digital baseband signal or the modulated digital intermediate frequency signal to obtain the rf signal, where the order of the up-mixing and the digital-to-analog conversion processes is adjustable. The digital baseband signal and the digital intermediate frequency signal may be collectively referred to as a digital signal.

A transceiver may also be referred to as a transceiver unit, transceiver, transceiving means, etc. Optionally, a device for implementing a receiving function in the transceiver unit may be regarded as a receiving unit, and a device for implementing a sending function in the transceiver unit may be regarded as a sending unit, that is, the transceiver unit includes a receiving unit and a sending unit, the receiving unit may also be referred to as a receiver, an input port, a receiving circuit, and the like, and the sending unit may be referred to as a transmitter, a sending circuit, and the like.

It should be noted that the communication apparatus shown in fig. 10 may be specifically configured to implement the steps implemented by the communication apparatus in the foregoing method embodiment, and details are not described here again.

The present application further provides a computer-readable storage medium storing one or more computer-executable instructions, which when executed by a processor, perform the method as described in the possible implementation manner of the communication device in the foregoing embodiments, where the communication device may specifically be the communication device in the foregoing embodiments.

The embodiments of the present application also provide a computer program product (or computer program) storing one or more computers, and when the computer program product is executed by the processor, the processor executes the method that may be implemented by the communication apparatus, where the communication apparatus may specifically be the communication apparatus in the foregoing embodiments.

An embodiment of the present application further provides a chip system, where the chip system includes a processor, and is configured to support a communication device to implement functions related to possible implementation manners of the communication device. In one possible design, the system-on-chip may further include a memory, which stores program instructions and data necessary for the communication device. The chip system may be formed by a chip, or may include a chip and other discrete devices, where the communication device may specifically be the signal detection device in the foregoing embodiment.

An embodiment of the present application further provides a network system architecture, where the network system architecture includes the communication device described above, and the communication device may specifically be a communication device in any one of the foregoing embodiments.

In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other manners. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units is only one logical division, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.

The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.

In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.

The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application may be substantially or partially embodied or embodied in a software product stored in a storage medium, which includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to perform all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.

Claims

1. A method of communication, comprising:

the method comprises the steps that a first network device receives service record data from a second network device, wherein the service record data comprise first data volume information, and the first data volume information is used for indicating the data volume of a first service occurring in a first time period;

the first network equipment carries out quantization processing on the service record data according to the target quantization parameter to obtain first service state information;

and the first network equipment takes the first service state information as the input of a preset model, and outputs second service state information of the first service in a second time period through the processing of the preset model.

2. The method according to claim 1, wherein before the first network device quantizes the service record data according to the target quantization parameter to obtain the first service status information, the method further comprises:

the first network equipment carries out quantization processing on the service record data according to the initial quantization parameter to obtain initial service state information;

the first network equipment determines a predictability upper bound value corresponding to the initial service state information;

and the first network equipment updates the initial quantization parameter according to the predictability upper bound value and the initial prediction precision to obtain the target quantization parameter.

3. The method of claim 2, wherein the first network device determining the upper predictability threshold corresponding to the initial traffic state information comprises:

and after the first network equipment determines a first entropy value corresponding to the initial service state information, determining the predictability upper bound value according to the first entropy value.

4. The method according to any one of claims 1 to 3, wherein after the first network device takes the first service status information as an input of a preset model, and outputs second service status information of the first service in a second time period through processing of the preset model, the method further comprises:

and the first network equipment performs inverse quantization processing on the second service state information according to the target quantization parameter to obtain second data volume information, wherein the second data volume information is used for indicating the data volume of the first service in the second time period.

5. The method of claim 4, further comprising:

and the first network equipment sends the second data volume information to the second network equipment.

6. The method according to any one of claims 1 to 5,

the service record data further comprises a first identifier of the network equipment corresponding to the first data volume information; and/or the presence of a gas in the gas,

7. The method according to any of claims 1 to 6, wherein the first service comprises one or more of a voice service, a data traffic service, a real-time traffic service of an application software.

8. A method of communication, comprising:

the method comprises the steps that a second network device sends service record data to a first network device, wherein the service record data comprise first data volume information, and the first data volume information is used for indicating the data volume of a first service in a first time period;

and the second network device receives second data volume information from the first network device, wherein the second data volume information is used for indicating the data volume of the first service in the second time period, and the second data volume information is obtained by performing quantization processing and inverse quantization processing on the service record data.

9. The method of claim 8, wherein before the second network device sends the traffic record data to the first network device, the method further comprises:

the second network equipment records initial information of the first service in the first time period;

and the second network equipment performs target processing on the initial information to obtain the first data volume information, wherein the target processing comprises feature extraction processing and/or data desensitization processing.

10. The method according to claim 8 or 9,

11. The method according to any of claims 8 to 10, wherein the first traffic comprises one or more of voice traffic, data traffic, real-time traffic of application software.

12. A communications apparatus, comprising:

13. The apparatus of claim 12, wherein the processing unit is further configured to:

14. The apparatus according to claim 13, wherein the processing unit is specifically configured to:

and after a first entropy value corresponding to the initial service state information is determined, determining the predictability upper bound value according to the first entropy value.

15. The apparatus according to any one of claims 12 to 14,

the processing unit is further configured to perform inverse quantization processing on the second service state information according to the target quantization parameter to obtain second data volume information, where the second data volume information is used to indicate a data volume of the first service occurring in the second time period.

16. The apparatus of claim 15, wherein the transceiver unit is further configured to:

and sending the second data volume information to the second network equipment.

17. The apparatus of any one of claims 12 to 16,

18. The apparatus according to any of claims 12 to 17, wherein the first service comprises one or more of a voice service, a data traffic service, a real-time traffic service of an application software.

19. A communication apparatus, comprising a processing unit and a transceiver unit:

the processing unit is configured to send service record data to a first network device through the transceiver unit, where the service record data includes first data volume information, and the first data volume information is used to indicate a data volume of a first service occurring in a first time period;

20. The apparatus according to claim 19, wherein the processing unit is specifically configured to;

recording initial information of the first service in the first time period;

21. The apparatus of claim 19 or 20,

22. The apparatus according to any of claims 19 to 20, wherein the first service comprises one or more of a voice service, a data traffic service, a real-time traffic service of an application software.

23. A communications apparatus, comprising: a processor coupled to a memory for storing a computer program or instructions, the processor being operable to execute the computer program or instructions in the memory to cause the communication apparatus to perform the method of any of claims 1 to 7 or to cause the communication apparatus to perform the communication method of any of claims 8 to 11.

24. A chip, wherein the chip comprises a processor and a communication interface;

wherein the communication interface is coupled to the processor for executing a computer program or instructions to implement the method of any one of claims 1 to 7 or to implement the communication method of any one of claims 8 to 11.

25. A computer-readable storage medium for storing a computer program or instructions which, when executed, cause the computer to perform the method of any of claims 1 to 7 or cause the computer to perform the method of any of claims 8 to 11.