Network data analysis method and communication apparatus

Abstract

The present application provides a network data analysis method and a communication apparatus. The method is applied to a recommendation service function. The method comprises: receiving first information from a first network element, the first information comprising a first request and instruction information, the first request being used for requesting data to be recommended, the instruction information being used for instructing to collect first data, and the first data being network data determined by the first network element and associated with the data to be recommended; acquiring a prediction analysis result, the prediction analysis result being obtained on the basis of the first data; on the basis of the prediction analysis result, determining the data to be recommended; and sending the data to be recommended to the first network element. By means of the present application, a globally optimized data recommendation service can be provided when the service logic of each network element does not need to be understood, thereby improving the adaptability and flexibility of the data recommendation service.

Description

A method and communication device for analyzing network data

[0001] This application claims priority to the Chinese patent application with application number 202410057768.6 filed with the State Intellectual Property Office of China on January 12, 2024, and priority to the Chinese patent application with the invention name “A method and communication device for network data analysis”, all contents of which are incorporated by reference into this application. Technical Field

[0002] The present application relates to the field of communication technology, and in particular to a method and a communication device for analyzing network data. Background Art

[0003] Network data analysis is a service that collects relevant network data from network function elements, application function elements, or network management systems, uses machine learning techniques to perform correlation analysis on the collected data of various dimensions, trains and fits models, and uses the models to provide data output. The network elements that provide network data analysis can be network data analytics function (NWDAF) elements, management data analytics function (MDAF) elements, etc. Taking NWDAF as an example, other network function (NF) elements in the network can consume the data analysis services provided by NWDAF and take actions to adjust network operations based on the output of NWDAF.

[0004] As shown in Figure 1, the central network automation function (NWE) element can obtain the target application quality of experience (QoE) level (i.e., target value) expected in a specified location area and the network operation prediction results (i.e., prediction outputs, such as experience analysis prediction results, user plane congestion prediction results, user plane function (UPF) load prediction results, etc.) output by the NWDAF. It can then determine the recommended actions for each network element within the specified location area (e.g., recommended UPF network element selection, recommended control policy settings for terminals, recommended service message transmission parameters for applications, etc.). The above factors affecting service quality of experience are controlled by multiple network elements, such as the policy control function (PCF), session management function (SMF), and application function (AF). These network elements will make network adjustments based on the recommended actions determined by the central NWE, thereby guiding multiple network elements to cooperate with each other to achieve network automation goals.

[0005] However, in the process of determining the recommended actions for each network element, the central network automation function network element needs to output reasonable recommended actions for each network element based on the knowledge graph of each network element, or deeply analyze the processing logic relationship between each network element within the network. Therefore, it is inevitable to understand the business logic of each network element, and develop different recommendation service functions according to different business logic for different business scenarios. This approach will not only increase the workload, but also reduce the adaptability and flexibility of the recommendation service. Summary of the Invention

[0006] The embodiments of the present application provide a method and a communication device for network data analysis. Based on the method described in the present application, a globally optimized data recommendation service can be provided without the need to understand the business logic of each network element, which is conducive to improving the adaptability and flexibility of the data recommendation service.

[0007] In a first aspect, the present application provides a method for network data analysis, which is applied to a recommendation service function, and the method includes: receiving first information from a first network element, the first information including a first request and indication information; the first request is used to request recommended data, and the indication information is used to indicate the collection of first data, the first data being network data associated with the recommended data determined by the first network element; obtaining a predictive analysis result, which is obtained based on the first data; determining the recommended data based on the predictive analysis result; and sending the recommended data to the first network element.

[0008] Based on the method described in the first aspect, since the first data specified for collection is determined by the first network element (for example, the first network element can determine the first data that needs to be collected based on the business processing logic), the predictive analysis results determined based on the first data can feedback the impact of the network adjustment behavior on the network; and the association relationship between the first data and the recommended data can also be determined by the first network element based on the business processing logic. Therefore, the work process executed by the recommendation service function after receiving the first information is data-driven, and can provide globally optimized data recommendation services without the need to understand the business logic of each network element, thereby reducing workload and helping to improve the adaptability and flexibility of the data recommendation service.

[0009] In one possible implementation, the indication information is also used to indicate the data source of the first data. Based on this method, the source of the first data can be clearly identified, and the data source of the first data is used to indicate that the first data is collected from one or more specified network elements. The first network element can determine, based on the business processing logic, one or more network elements that can best reflect the impact of network adjustment behavior on the network as the source of the first data. When the recommendation service function does not need to understand the business logic of each network element, the efficiency of data collection is improved, the workload is further reduced, and the adaptability and flexibility of the data recommendation service are improved.

[0010] In a possible implementation, the first information further includes an analysis type identifier; or the method further includes: determining the analysis type identifier based on the target parameter and the first request. This approach is conducive to improving the flexibility of determining the analysis type identifier.

[0011] In a possible implementation, the analysis type identifier indicates any one of the following types: service experience analysis, user plane congestion analysis, and network element load analysis.

[0012] In a possible implementation, the target parameter includes any one of the following parameters: quality of experience, user plane congestion level, and load levels of multiple UPFs.

[0013] In one possible implementation, obtaining the predictive analysis results includes: collecting the first data and second data corresponding to the analysis type identifier; determining whether the first data belongs to the second data, or whether the first data does not belong to the second data; and determining the predictive analysis results based on the first data and the second data. This approach is conducive to improving the accuracy of the predictive analysis results, thereby improving the efficiency of the recommendation service.

[0014] In one possible implementation, determining a predictive analysis result based on the first data and the second data includes: increasing the weight of the first data relative to other data in the second data excluding the first data based on the indication information; or increasing the weight of the first data relative to the second data based on the indication information; or determining the predictive analysis result based on the weight of the first data, the weight of the second data, the first data, and the second data. It can be understood that when the first data belongs to the second data, during the process of obtaining the predictive analysis result, the recommendation service function can increase the weight of the first data relative to other data in the second data excluding the first data based on the indication for collecting the first data, so that the predictive analysis result can more accurately reflect the impact of network adjustment behavior on the network; when the first data does not belong to the second data, in addition to obtaining the predictive analysis result based on the second data, it is also necessary to simultaneously obtain the predictive analysis result based on the first data, and the weight of the first data relative to the second data based on the indication for collecting the first data, so that the predictive analysis result can reflect the impact of network adjustment behavior on the network. Based on this approach, the accuracy of the predictive analysis results is improved, thereby further improving the efficiency of the recommendation service.

[0015] In one possible implementation, obtaining a predictive analysis result includes: sending a second request to a predictive analysis function; the second request is used to request the predictive analysis result, and the second request includes an analysis type identifier and the indication information; and receiving the predictive analysis result from the predictive analysis function. The indication information instructs the collection of first data, and optionally also indicates a data source for the first data. When the first data belongs to the second data corresponding to the analysis type identifier, the predictive analysis function may, based on the indication to collect the first data, increase the weight of the first data relative to other data in the second data other than the first data, thereby enabling the predictive analysis result to more accurately reflect the impact of network adjustment behavior on the network. When the first data does not belong to the second data, in addition to collecting the second data, it is necessary to simultaneously obtain a predictive analysis result based on the first data, and the weight of the first data relative to the second data may be increased based on the indication to collect the first data, thereby enabling the predictive analysis result to reflect the impact of network adjustment behavior on the network. When the indication information also includes the data source of the first data, the predictive analysis function collects the first data from one or more network elements specified by the data source. Based on this approach, the predictive analysis result can reflect the impact of network adjustment behavior on the network, which is conducive to improving the accuracy of the predictive analysis result and further improving the efficiency of the recommendation service.

[0016] In one possible implementation, the method further includes determining analysis filtering information based on the first request; the analysis filtering information is used to indicate the network range corresponding to the predictive analysis result. Specifically, the analysis filtering information is determined based on information indicating the collection of first data in the first request; or based on the data source of the first data in the first request. The second request also includes the analysis filtering information. Based on this approach, the efficiency of predictive analysis can be improved.

[0017] In one possible implementation, the analysis type identifier indicates service experience analysis; the first network element includes an application function (AF), a policy control function (PCF), and a session management function (SMF); and the first data includes a peak rate and an average rate of the first application at the current moment and / or at historical moments. Based on this approach, the QoE of the application can be improved.

[0018] In a possible implementation, when the first network element is an AF, the recommendation data includes a recommended server instance of the first application.

[0019] In a possible implementation, when the first network element is a PCF, the recommendation data includes a recommended peak rate and a recommended average rate of the first application.

[0020] In a possible implementation, when the first network element is an SMF, the recommendation data includes a recommended data network access identifier of the first application.

[0021] In one possible implementation, the analysis type identifier indicates user plane congestion analysis; the first request includes an identifier of the second application; the first network element includes a PCF; the first data includes a user plane congestion level of the second application at a current moment and / or at a historical moment; and the recommended data includes a recommended peak throughput and / or a recommended average throughput of the second application. Based on this approach, QoS parameters can be adjusted to avoid user plane congestion.

[0022] In one possible implementation, the analysis type identifier indicates a network element load analysis; the first network element includes at least one SMF; the first data includes the number of N4 interface sessions established by the at least one SMF on multiple UPFs at the current time and / or at historical times; and the recommendation data includes the number of N4 interface sessions recommended by the at least one SMF for establishment on the multiple UPFs. Based on this approach, the selection and distribution ratio of each UPF can be determined to prevent uneven or overloaded UPFs.

[0023] In a possible implementation manner, the first information further includes an expected value of the target parameter.

[0024] In one possible implementation, determining the recommended data based on the prediction analysis results includes: determining a deviation value based on the prediction analysis results and the expected value of the target parameter; adjusting model parameters of a first data optimizer in a direction that reduces the deviation value to obtain a second data optimizer; and determining the recommended data using the second data optimizer. This approach can improve the accuracy of the recommended data.

[0025] In the second aspect, the present application provides a method for network data analysis, which is applied to a predictive analysis function, and the method includes: receiving a second request from a recommendation service function; the second request is used to request a predictive analysis result, and the second request includes an analysis type identifier and indication information, and the indication information is used to indicate the collection of first data, where the first data is network data determined by a first network element and is associated with the recommended data output by the recommendation service function; collecting the first data and second data corresponding to the analysis type identifier; wherein the first data belongs to the second data, or the first data does not belong to the second data; determining a predictive analysis result based on the first data and the second data; and sending the predictive analysis result to the recommendation service function.

[0026] The beneficial effects of the possible implementation of the second aspect can be found in the beneficial effects of the possible implementation of the first aspect, and will not be repeated here.

[0027] In a possible implementation, the indication information is further used to indicate a data source of the first data; and collecting the first data includes: collecting the first data from the data source, wherein the data source may be one or more network elements.

[0028] In one possible implementation, determining the prediction analysis result based on the first data and the second data includes: increasing the weight of the first data relative to other data in the second data except the first data according to the indication information; or increasing the weight of the first data relative to the second data according to the indication information; determining the prediction analysis result based on the weight of the first data, the weight of the second data, the first data, and the second data. It can be understood that when the first data belongs to the second data, in the process of obtaining the prediction analysis result, the prediction analysis function can increase the weight of the first data relative to other data in the second data except the first data according to the indication of collecting the first data, so that the prediction analysis result can more accurately reflect the impact of the network adjustment behavior on the network; when the first data does not belong to the second data, in addition to being based on the second data, it is also necessary to obtain the prediction analysis result based on the first data at the same time, and increase the weight of the first data relative to the second data according to the indication of collecting the first data, so that the prediction analysis result can reflect the impact of the network adjustment behavior on the network.

[0029] In one possible implementation, the analysis type identifier indicates service experience analysis; the first network element includes AF, PCF and SMF; the first data includes the peak rate and average rate of the first application at the current moment and / or historical moment; the first data is collected through UPF; the second data includes second information collected from AF, and third information collected from UPF and / or SMF; the second information includes the experience quality of the first application at the current moment and / or historical moment, the server instance of the first application and the identifier of the first application; the third information includes the data network access identifier, user data rate, user message delay and service quality service flow identifier corresponding to the application message of the first application at the current moment and / or historical moment.

[0030] In a possible implementation, the second request further includes analysis filtering information, where the analysis filtering information is used to indicate a network range corresponding to the analysis prediction result.

[0031] In one possible implementation, the analysis type identifier indicates user plane congestion analysis; the first network element includes PCF; the first data includes the user plane congestion level of the second application at the current moment and / or historical moment; the first data is collected through UPF; the second data includes fourth information collected from UPF and / or SMF; the fourth information includes the peak throughput of the second application at the current moment and / or historical moment, the average throughput of the second application, the identifier of the second application, and the location information of the terminal device.

[0032] In one possible implementation, the analysis type is identified as network element load analysis; the first network element includes at least one SMF; the first data includes the number of N4 interface sessions established by the at least one SMF on multiple UPFs at the current moment and / or historical moments; the first data is collected through the multiple UPFs; the second data includes fifth information collected from the multiple UPFs; the fifth information includes load level data of the multiple UPFs at the current moment and / or historical moments.

[0033] In a third aspect, the present application provides a method for network data analysis, which is applied to a first network element, and the method includes: sending first information to a recommendation service function, the first information including a first request and indication information; the first request is used to request recommended data, and the indication information is used to indicate the collection of first data, the first data being network data associated with the recommended data determined by the first network element; receiving recommended data from the recommendation service function; and determining a first parameter based on the recommended data.

[0034] The beneficial effects of possible implementations of the third aspect can be found in the beneficial effects of possible implementations of the first aspect, and will not be repeated here.

[0035] In a possible implementation manner, the method further includes: determining, based on a service processing logic of the first network element, that the first data is associated with the recommended data.

[0036] In a possible implementation, the first data is a network operation indicator affected by the first parameter.

[0037] In a possible implementation, the first information further includes an analysis type identifier. Optionally, the analysis type identifier indicates any one of the following types: service experience analysis, user plane congestion analysis, and network element load analysis.

[0038] In a possible implementation, the method further includes: determining an analysis type identifier based on a service processing logic of the first network element; and determining the first request and the indication information based on the analysis type identifier.

[0039] In a possible implementation, the analysis type identifier indicates service experience analysis; the first network element includes AF, PCF, and SMF; and the first data includes a peak rate and an average rate of the first application at a current moment and / or a historical moment.

[0040] In a possible implementation, when the first network element is an AF, the recommendation data is a recommended server instance of the first application, and the first parameter is a server instance selected by the first application.

[0041] In one possible implementation, when the first network element is a PCF, the recommended data includes a recommended peak rate of the first application and a recommended average rate of the first application; the first parameter includes a maximum stream bit rate or an aggregate maximum rate, and the first parameter also includes a guaranteed bit rate; determining the first parameter based on the recommended data includes: determining the maximum stream bit rate or the aggregate maximum rate based on the recommended peak rate of the first application, and determining the guaranteed bit rate based on the recommended average rate of the first application.

[0042] In a possible implementation, when the first network element is an SMF, the recommended data is a recommended data network access identifier of the first application, and the first parameter is a user plane path selected by the first application.

[0043] In one possible implementation, the analysis type identifier indicates user plane congestion analysis; the first request includes an identifier of the second application; the first network element includes a PCF; the first parameter is a session service quality parameter; the first data includes a user plane congestion level of the second application at a current moment and / or a historical moment; and the recommended data includes a recommended peak throughput and / or a recommended average throughput of the second application.

[0044] In one possible implementation, the analysis type identifier indicates network element load analysis; the first network element includes at least one SMF; the first parameter includes the selection ratio of the at least one SMF for multiple UPFs; the first data includes the number of N4 interface sessions established by the at least one SMF on the multiple UPFs at the current moment and / or historical moment; the recommended data includes the number of N4 interface sessions recommended by the at least one SMF to be established on the multiple UPFs.

[0045] In a possible implementation manner, the indication information is also used to indicate the data source of the first data.

[0046] In a possible implementation, the first information further includes an expected value of a target parameter. Optionally, the target parameter includes any one of the following parameters: quality of experience, user plane congestion level, and load levels of multiple UPFs.

[0047] In a fourth aspect, the present application provides a communication device, which includes a processor. When the processor calls a computer program in a memory, the methods described in the first to third aspects are executed.

[0048] In a fifth aspect, the present application provides a communication device, which includes a processor and a memory, and the processor and the memory are coupled; the processor is used to implement the methods described in the first to third aspects.

[0049] In a sixth aspect, the present application provides a communication device, which includes a processor, a memory, and a transceiver, wherein the processor and the memory are coupled; the transceiver is used to send and receive data, and the processor is used to implement the methods described in the first to third aspects.

[0050] In the seventh aspect, the present application provides a chip, which includes a processor and an interface, and the processor and the interface are coupled; the interface is used to receive or output signals, and the processor is used to execute code instructions so that the methods described in the first to third aspects are executed.

[0051] In an eighth aspect, the present application provides a computer-readable storage medium, in which a computer program or instruction is stored. When the computer program or instruction is executed by a communication device, the method described in the first to third aspects is implemented.

[0052] In a ninth aspect, the present application provides a communication system, which includes a recommendation service function and a first network element. The recommendation service function is used to execute the method described in the first aspect, and the first network element is used to execute the method described in the third aspect.

[0053] In the tenth aspect, the present application provides a communication system, which includes a recommendation service function, a prediction and analysis function and a first network element. The recommendation service function is used to execute the method described in the first aspect, the prediction and analysis function is used to execute the method described in the second aspect, and the first network element is used to execute the method described in the third aspect.

[0054] In an eleventh aspect, the present application provides a computer program product comprising instructions, which, when read and executed by a computer, enables the computer to execute the method as described in the first to third aspects. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] FIG1 is a schematic diagram of a basic architecture of a recommendation service provided in an embodiment of the present application;

[0056] FIG2 is a schematic diagram of a network system architecture provided in an embodiment of the present application;

[0057] FIG3 is a schematic diagram of a CU-DU architecture provided in an embodiment of the present application;

[0058] FIG4 is a schematic diagram of a 5G network architecture provided in an embodiment of the present application;

[0059] FIG5 is a schematic diagram of the basic architecture of a new recommendation service provided in an embodiment of the present application;

[0060] FIG6 is a flow chart of a method for analyzing network data according to an embodiment of the present application;

[0061] FIG7A is a schematic diagram of a process for a second network element to obtain prediction analysis results according to an embodiment of the present application;

[0062] FIG7B is a schematic diagram of another process of obtaining prediction analysis results by a second network element according to an embodiment of the present application;

[0063] FIG7C is a schematic diagram of a process for a second network element to determine recommended data based on the prediction analysis result provided by an embodiment of the present application;

[0064] FIG8 is a flow chart of another method for analyzing network data provided in an embodiment of the present application;

[0065] FIG9A is a schematic diagram of another process of obtaining prediction analysis results by a second network element according to an embodiment of the present application;

[0066] FIG9B is a schematic diagram of another process of obtaining prediction analysis results by a second network element according to an embodiment of the present application;

[0067] FIG10 is a flow chart of another method for analyzing network data provided in an embodiment of the present application;

[0068] FIG11 is a flow chart of another method for analyzing network data provided in an embodiment of the present application;

[0069] FIG12 is a schematic structural diagram of a communication device provided in an embodiment of the present application;

[0070] FIG13 is a schematic structural diagram of another communication device provided in an embodiment of the present application;

[0071] FIG14 is a schematic structural diagram of a chip provided in an embodiment of the present application. DETAILED DESCRIPTION

[0072] The terms "first" and "second" and the like in the specification, claims, and drawings of this application are used to distinguish between different objects, not to describe a particular order. Furthermore, the terms "including" and "having," and any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or apparatus comprising a series of steps or elements is not limited to the listed steps or elements, but may optionally include steps or elements not listed, or may optionally include other steps or elements inherent to the process, method, product, or apparatus.

[0073] References herein to "embodiments" mean that a particular feature, structure, or characteristic described in connection with the embodiments may be included in at least one embodiment of the present application. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it constitute an independent or alternative embodiment that is mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0074] In this application, "at least one (item)" refers to one or more, "more than one" refers to two or more, "at least two (items)" refers to two or three and more than three, and "and / or" is used to describe the corresponding relationship of associated objects, indicating that there can be three relationships. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the previous and next associated objects are in an "or" relationship. "At least one of the following items" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0075] To better understand the embodiments of the present application, the following first introduces the system architecture involved in the embodiments of the present application:

[0076] The technical solutions of the embodiments of the present application can be applied to various communication systems, such as satellite communication systems and traditional mobile communication systems. Among them, the satellite communication system can be integrated with the traditional mobile communication system (i.e., the ground communication system). Mobile communication systems include, for example, wireless local area network (WLAN) communication systems, wireless fidelity (Wi-Fi) systems, long term evolution (LTE) systems, LTE frequency division duplex (FDD) systems, LTE time division duplex (TDD) systems, fifth generation (5G) systems or new radio (NR), and other future communication systems, such as sixth generation (6G) systems, etc., and also support communication systems that integrate multiple wireless technologies. For example, it can also be applied to systems that integrate non-terrestrial networks (NTN) such as drones, satellite communication systems, and high altitude platform stations (HAPS) communications with ground mobile communication networks. It is understandable that the system architecture described in the embodiments of the present application is intended to more clearly illustrate the technical solutions of the embodiments of the present application, and does not constitute a limitation on the technical solutions provided in the embodiments of the present application.

[0077] Please refer to Figure 2, which is a schematic diagram of a network system architecture provided by an embodiment of the present application. As shown in Figure 2, the next-generation mobile communication network architecture formulated by the 3rd Generation Partnership Project (3GPP) standard is called a 5G network architecture. Terminal devices can access a wireless network to obtain services from an external network (such as a data network (DN)) through the wireless network, or communicate with other devices through the wireless network, such as communicating with other terminal devices. The wireless network includes a (radio) access network ((R)AN) and a core network (CN), wherein the (R)AN (hereinafter described as RAN) is used to access the terminal device to the wireless network, and the CN is used to manage the terminal device and provide a gateway for communicating with the DN. The terminal device, RAN, CN and DN involved in the network architecture in Figure 2 are described in detail below.

[0078] 1. Terminal Equipment

[0079] Terminal devices include devices that provide voice and / or data connectivity to users. For example, a terminal device is a device with wireless transceiver capabilities and can be deployed on land, including indoors or outdoors, handheld, wearable, or vehicle-mounted; on water (such as ships); or in the air (such as aircraft, balloons, and satellites). Terminal devices can specifically refer to user equipment (UE), access terminal, subscriber unit, user station, mobile station, remote station, remote terminal, mobile device, user terminal, wireless communication device, user agent, or user device. The terminal device may also be a satellite phone, a cellular phone, a smart phone, a wireless data card, a wireless modem, a machine type communication device, a cordless phone, a session initiation protocol (SIP) phone, a wireless local loop (WLL) station, a PDA, a handheld device with wireless communication capabilities, a computing device or other processing device connected to a wireless modem, an in-vehicle device, a communication device carried on a high-altitude aircraft, a wearable device, a drone, a robot, a terminal in device-to-device (D2D) communication, a terminal in vehicle to everything (V2X), a virtual reality (VR) terminal device, an augmented reality (AR) terminal device, a wireless terminal in industrial control, a wireless terminal in self-driving, a wireless terminal in remote medical care, a wireless terminal in a smart grid, a wireless terminal in transportation safety, a wireless terminal in a smart city, a wireless terminal in a smart home, a wireless terminal in an industrial control system, a wireless terminal in self-driving, a wireless terminal in remote medical care, a wireless terminal in a smart grid, a wireless terminal in transportation safety, a wireless terminal in a smart city, a wireless terminal in a smart home ... The present application does not limit the wireless terminals in the home or terminal devices in future communication networks. The terminal can also be fixed or mobile. In addition, in the present application, when not specifically stated, "terminal device" can refer to the terminal device itself or a component in the terminal device, such as a chip system, SoC, which can be installed in the terminal device. It can be understood that all or part of the functions of the terminal in the present application can also be implemented by software functions running on hardware, or by virtualization functions instantiated on a platform (such as a cloud platform). The terminal device in the present application can be a terminal for 5G or a terminal for 6G, and the present application does not limit this.

[0080] 2. RAN

[0081] The RAN may include one or more RAN devices (or access network devices). The interface between the access network device and the terminal device may be a Uu interface (or air interface). Of course, in communications evolved after 5G, the names of these interfaces may remain unchanged or may be replaced by other names, and this application does not limit this.

[0082] Access network equipment is a node or device that connects a terminal device to a wireless network. Examples of access network equipment include, but are not limited to, next generation node B (gNB), evolved node B (eNB), next generation eNB (ng-eNB), wireless backhaul equipment, radio network controller (RNC), node B (NB), home evolved node B (HeNB) or home node B (HNB), baseband unit (BBU), transmitting and receiving point (TRP), transmitting point (TP), mobile switching center, and equipment that performs base station functions in device-to-device (D2D), vehicle-to-everything (V2X), and machine-to-machine (M2M) communications in 5G communication systems. It may also include centralized units (C-RAN) in cloud radio access network (C-RAN) systems. The RAN in this application may be a 5G RAN or a 6G RAN, and this application does not limit this.

[0083] Taking the RAN including gNB as an example, the RAN can be connected to the core network (for example, the LTE core network or the 5G core network). As shown in Figure 3, the CU and DU can be understood as a logical functional division of a base station (such as a gNB). The CU and DU can be physically separate or deployed together. Multiple DUs can share a single CU. Of course, a single DU can also be connected to multiple CUs (not shown in Figure 2). The CU and DU can be connected via an interface, such as the F1 interface. The CU and DU can be divided based on the protocol layers of the wireless network. For example, one possible division is: the CU is responsible for performing the functions of the radio resource control (RRC) layer, the service data adaptation protocol (SDAP) layer, and the packet data convergence protocol (PDCP) layer, while the DU is responsible for performing the functions of the radio link control (RLC) layer, the media access control (MAC) layer, the physical layer, etc. It should be understood that this division of CU and DU processing functions based on protocol layers is merely an example, and other divisions are also possible. For example, the CU or DU can be divided into functions with more protocol layers. For example, the CU or DU can also be divided into partial processing functions with the protocol layer. In one design, some functions of the RLC layer and the functions of the protocol layers above the RLC layer are set in the CU, and the remaining functions of the RLC layer and the functions of the protocol layers below the RLC layer are set in the DU. In another design, the functions of the CU or DU can also be divided according to the service type or other system requirements. For example, according to the delay, the functions whose processing time needs to meet the delay requirements are set in the DU, and the functions that do not need to meet the delay requirements are set in the CU. In another design, the CU can also have one or more functions of the core network. One or more CUs can be set centrally or separately. For example, the CU can be set on the network side for convenient centralized management. The DU can have multiple radio frequency functions, or the radio frequency functions can be set remotely.

[0084] 3. CN

[0085] The CN may include one or more CN devices (which may be understood as network element devices or network function (NF)). Hereinafter, the CN devices are collectively referred to as core network elements.

[0086] Please refer to Figure 4, which is a schematic diagram of a 5G network architecture provided in this application. The CN shown in Figure 4 includes multiple CN devices: network slice selection function (NSSF), network exposure function (NEF), network function repository function (NRF), policy control function (PCF), unified data management (UDM), unified data repository function (UDR), application function (AF), network slice specific authentication and authorization function (NSSAAF), authentication server function (AUSF), access and mobility management function (AMF), session management function (SMF), user plane function (UPF), service communication proxy (SCP), network slice admission control function (NSACF), data analysis function (such as network data analytics function (NWDAF), management data analytics function (MDAF)), etc. In addition, Figure 4 also includes terminal equipment (UE), access network equipment (RAN), data network (DN) and operation administration and maintenance (OAM) system. The access network equipment can communicate with the OAM system. Among them:

[0087] The AF primarily interacts with the 5G core network to deliver services. For example, it supports the following functions: application impact on service routing, exposure of access network capabilities, and interaction with the policy framework for policy management. Based on the operator's deployment strategy, the trusted AF can directly access internal network element functions within the 5G core network to improve service processing efficiency. Alternatively, it can utilize a general external network element function access framework to exchange information with the corresponding internal network element functions through the NEF.

[0088] The AMF is a control plane function provided by the operator network, responsible for access control and mobility management of terminal devices accessing the operator network, such as mobility status management, allocating temporary user identities, and authenticating and authorizing users.

[0089] SMF is a control plane function provided by the operator network, responsible for managing the protocol data unit (PDU) sessions of terminal devices. A PDU session is a channel for transmitting PDUs. Terminal devices need to transmit PDUs to and from the DN through PDU sessions. The SMF is responsible for establishing, maintaining, and deleting PDU sessions. SMF includes session management (such as session establishment, modification, and release, including tunnel maintenance between the UPF and RAN), UPF selection and control, service and session continuity (SSC) mode selection, roaming, and other session-related functions.

[0090] The PCF is a control plane function provided by the operator, including user subscription data management, policy control, charging policy control, and Quality of Service (QoS) control. It is mainly used to provide PDU session policies to the SMF. Among them, the policies may include charging-related policies, QoS-related policies, and authorization-related policies.

[0091] The UPF is a gateway provided by the operator and serves as the gateway for communication between the operator network and the DN. The UPF includes user-plane-related functions such as packet routing and transmission, packet inspection, QoS processing, uplink packet inspection, and downlink packet storage.

[0092] The UDM is primarily used to manage user subscription and authentication data, as well as perform authentication credit processing, user identity processing, access authorization, registration / mobility management, subscription management, and short message management. In some embodiments, the UDM may also include a unified data repository (UDR). Alternatively, in other embodiments, the 3GPP SBA of the 5G system may also include a UDR. The UDR is used to provide storage and retrieval for PCF policies, storage and retrieval of open structured data, and storage of user information requested by application functions.

[0093] The UDR is primarily used to store and retrieve contract data, policy data, and public architecture data; it provides access to relevant data for the UDM, PCF, and NEF. The UDR must implement different data access authentication mechanisms for different types of data, such as contract data and policy data, to ensure data access security. The UDR must be able to return a failure response with an appropriate reason value for illegal service-based operations or data access requests.

[0094] OAM systems typically categorize network management tasks into three main categories: operations, administration, and maintenance, based on the actual needs of carriers' network operations. Operations primarily involve analyzing, forecasting, planning, and configuring daily network and service operations; maintenance primarily involves day-to-day operational activities such as testing and troubleshooting the network and its services.

[0095] The data analysis function may include a network function that provides predictive analysis services (i.e., a predictive analysis function), which can collect data from various NFs (e.g., AF, AMF, SMF, PCF, NEF, etc.) or from the OAM system and perform analysis and prediction. In the 5G communication system, the data analysis function may specifically be NWDAF, MDAF, etc. In the embodiment of the present application, the data analysis function is described as NWDAF, but the specific form and implementation form are not limited.

[0096] It should be noted that in the embodiments of the present application, a new data-driven network function, namely a recommendation service function, can be added to perform the recommendation service. The specific content of the recommendation service function will be described later and will not be described here. Among them, the recommendation service function and the prediction analysis function can be concentrated in the same network element (that is, the recommendation service function and the prediction analysis function are the same data analysis function). For example, the recommendation service function can be a unit within the NWDAF. In this case, the NWDAF network element is called an NWDAF network element that supports recommendation services. The recommendation service function can also be an independent network element, which is not limited here.

[0097] In addition, the above-mentioned CN devices may also be referred to as network elements or functional network elements. In a 5G communication system, each functional network element may be the name of each functional network element shown in FIG4 . In a communication system evolved after 5G (such as a 6G communication system), each functional network element may still be the name of each functional network element shown in FIG4 , or may have other names. For example, in a 5G communication system, the user plane function may be a UPF. In a communication system evolved after 5G (such as a 6G communication system), the user plane function may still be a UPF, or may have other names, which is not limited in this application.

[0098] It should also be noted that in the 5G communication system, the functions implemented by each functional network element can be independent as shown in Figure 4. In the communication system evolved after 5G (such as the 6G communication system), each functional network element can still be in an independent state as shown in Figure 4, or the functions of multiple functional network elements in Figure 4 can be implemented by an integrated functional network element. For example, in the 5G communication system, the user plane related functions are implemented by the UPF, and the access and mobility management related functions are implemented by the AMF. In the communication system evolved after 5G (such as the 6G communication system), the user plane related functions can still be implemented by the UPF, and the access and mobility management related functions can still be implemented by the AMF, or the user plane related functions and the access and mobility management related functions can be implemented by an integrated functional network element at the same time, which is not limited in this application.

[0099] In Figure 4 , Nnssf, Nnef, Nnrf, Npcf, Nudm, Naf, Nudr, Nnssaaf, Nausf, Namf, Nsmf, Nnsacf, N1, N2, N3, N4, and N6 are interface serial numbers. The meanings of these interface serial numbers are defined in relevant standard protocols and are not limited here.

[0100] 4. DN

[0101] DN, also known as packet data network (PDN), is a network located outside the operator network. The operator network can access multiple DNs. Application servers corresponding to various services can be deployed in the DN to provide a variety of possible services for terminal devices.

[0102] To facilitate understanding of the solutions provided by the embodiments of this application, the following describes the recommended services involved in the embodiments of this application:

[0103] Taking NWDAF as an example, other NFs in the network can consume the predictive analysis services provided by NWDAF and take actions to adjust network operations based on the prediction output of NWDAF.

[0104] For example, SMF can select or reselect UPF for a protocol data unit session (PDU Session) that is being established or has been established based on the load prediction analysis results of UPF output by NWDAF, so as to achieve load balancing of multiple UPF network elements and prevent a certain UPF from being overloaded. The SMF network element can also select a user plane path (UP path) with better experience for the session that transmits the application service message based on the service experience analysis and prediction of a certain application output by NWDAF, where the UP path includes UPF, and / or Data Network Access Identifier (DNAI), and / or application server instance; where DNAI is used to identify the mobile edge computing (MEC) node.

[0105] For example, PCF can analyze and predict the service experience of an application output by NWDAF, determine the diversion control strategy, and allow the application's service messages to be processed at the nearest MEC node.

[0106] For example, AF can analyze and predict the service experience of the application output by NWDAF, deploy the application server in the MEC with better service experience, and select the appropriate application server instance for the user through application load balancing.

[0107] For another example, the PCF network element can adjust the quality of service (QoS) parameters of several applications with the largest traffic based on the user plane congestion analysis and prediction output by the NWDAF, limit the maximum data rate of these applications, and avoid congestion in the user plane.

[0108] However, when multiple NFs use NWDAF's prediction output to make their own decisions about the actions they take, there are scenarios where there is a lack of coordination. For example, when both the SMF and AF simultaneously make decisions based on the NWDAF's service experience prediction, the UPF and DNAI selected by the SMF may not match the application server instance selected by the AF (i.e., they are not located on the same MEC node), thus failing to achieve the optimal service experience. For another example, when multiple SMFs simultaneously select a UPF based on the UPF's load prediction, they may simultaneously allocate a large number of new PDU sessions to the UPF with the lowest load, causing the load on this UPF to suddenly increase, resulting in UPF load fluctuations.

[0109] Therefore, a global recommendation service output is needed to guide multiple NFs to coordinate and take actions that optimize the overall network. As shown in Figure 1, a basic architecture for a recommendation service has been proposed, which adds a central network automation function element. This element uses the central network automation function element to obtain the target application quality of experience level (i.e., target value) and the network operation forecast results (i.e., forecast output, such as experience analysis forecast results, user plane congestion forecast results, and UPF load forecast results) output by the NWDAF. This element then determines the recommended actions for each NF within the specified location area (e.g., recommended UPF NF selection, recommended control policy settings for terminals, recommended service message transmission parameters for applications, etc.). The above factors affecting service quality of experience are controlled by multiple NFs, such as the PCF, SMF, and AF. These NFs adjust the network based on the recommended actions determined by the central network automation function element, thereby guiding multiple NFs to coordinate and achieve network automation goals.

[0110] However, in the process of determining the recommended actions for each network element, the central network automation function network element needs to output reasonable recommended actions for each network element based on the knowledge graph of each network element, or deeply analyze the processing logic relationship between each network element within the network. Therefore, it is inevitable to understand the business logic of each network element, and develop different recommendation service functions according to different business logic for different business scenarios. This approach will not only increase the workload, but also reduce the adaptability and flexibility of the recommendation service.

[0111] Therefore, in order to provide a globally optimized data recommendation service without having to understand the business logic of each network element, thereby improving the adaptability and flexibility of the data recommendation service, the present embodiment provides a new basic architecture for the recommendation service. As shown in Figure 5, the basic architecture of the recommendation service includes a first network element (i.e., a recommendation service consumer), a recommendation service function, and a predictive analysis function.

[0112] The first network element refers to the network element that consumes the recommendation service. For example, it can be a functional network element such as the SMF, PCF, or AF that makes up the network, or it can be the network's OAM system. The recommendation service function is a newly added data-driven network function that receives recommendation service requests or recommendation subscription requests from other NF network elements, obtains network data output by the predictive analysis function for network operation at a certain point in the future, and combines the predicted network data with network operation goals to determine the network data output for the recommendation service. The recommendation service function specifically includes a data optimizer unit and a data deviation monitoring unit. The data deviation monitoring unit can determine the deviation between the output of the prediction model and the optimization target data; the data optimizer determines the recommended data output that currently approaches the optimization target based on the deviation from the target data. The predictive analysis function can be the NWDAF mentioned above. It should be noted that the recommendation service function and the predictive analysis function can be integrated into the same network element (i.e., the recommendation service function and the predictive analysis function are the same data analysis function). For example, the recommendation service function can be a unit within the NWDAF. In this case, the NWDAF network element is referred to as an NWDAF network element that supports recommendation services. This means that the NWDAF network element that supports recommendation services includes both predictive analysis and recommendation service functions. Of course, the recommendation service function can also be an independent network element.

[0113] Specifically, the first network element can send a recommendation service request or subscription to the recommendation service function, specifying the data to be collected in the recommendation service request or subscription; the recommendation service function sends the specified data to be collected to the prediction analysis function through the predictive analysis subscription; the prediction analysis function can collect the specified data and network data from each network element (NFs), analyze and predict the collected data using the prediction model, obtain the prediction analysis results, and send the prediction analysis function to the recommendation service function; the recommendation service function uses the data deviation monitoring unit and the data optimizer for processing to obtain the recommended data, and sends the recommended data to the first network element, so that the first network element can adjust the network according to the recommended data, and realize the mutual cooperation of multiple network elements to achieve the optimization goal. Among them, the recommendation service function and the prediction analysis function can be concentrated in the same network element, or they can be independent network elements, which is not limited here.

[0114] Since the designated collected data is determined by the first network element based on the business processing logic, the prediction output of the predictive analysis function can feedback the impact of network adjustment behavior on the network; at the same time, there is a correlation between the recommended data and the data designated to be collected by the first network element. This correlation depends on the determination of the first network element based on the business processing logic and does not need to be mastered by the recommendation service function. Therefore, the entire working process of the recommendation service function is data-driven. Without the need to understand the business logic of each network element, it can also provide globally optimized data recommendation services, reducing workload and helping to improve the adaptability and flexibility of the data recommendation service.

[0115] The network data analysis method and communication device provided in the embodiments of the present application are further described in detail below.

[0116] Figure 6 is a flow chart of a method for network data analysis provided in an embodiment of the present application. As shown in Figure 6, the method for network data analysis includes the following steps S601 to S605. The execution subject of the method shown in Figure 6 can be the first network element and the second network element. Alternatively, the execution subject of the method shown in Figure 6 can be the chip in the first network element and the chip in the second network element, which is not limited in the embodiment of the present application. Figure 6 takes the first network element and the second network element as the execution subject of the method as an example. Among them, the second network element includes a recommendation service function, for example, the second network element can be a recommendation service function (the recommendation service function can be considered to be an independent network element); the second network element can also include both a recommendation service function and a predictive analysis function (the recommendation service function and the predictive analysis function can be considered to be concentrated in the same network element, that is, the recommendation service function and the predictive analysis function are the same data analysis function, such as the NWDAF network element that supports the recommendation service).

[0117] S601: A first network element sends first information to a second network element. The first information includes a first request and instruction information. The first request is for requesting recommended data, and the instruction information is for instructing the collection of first data. The first data is network data associated with the recommended data determined by the first network element. In response, the second network element receives the first information from the first network element.

[0118] In the embodiment of the present application, the first request here can be considered as a recommendation service subscription request, which is used to request recommended data. The indication information can specify the first data that needs to be collected. Optionally, the indication information can also indicate the data source of the first data, and the data source of the first data is used to indicate that the first data is collected from one or more specified network elements or a specified network range. Optionally, after accepting the recommendation service subscription request from the first network element, the second network element can send a response message of successful subscription to the first network element. The indication information can also be carried in the first request, which is not limited here.

[0119] In one possible implementation, the method further includes: the first network element determining, based on the service processing logic of the first network element, that the first data is associated with the recommended data. The so-called service processing logic refers to the rules and processes that a network element has when providing services to other network elements or consuming services provided by other network elements.

[0120] It can be understood that the first data specified for collection is determined by the first network element according to the business processing logic, and the subsequent predictive analysis results determined based on the first data can naturally also feedback the impact of the network adjustment behavior on the network; and the association between the first data and the recommended data is also determined by the first network element based on the business processing logic. Therefore, the work process executed by the second network element after receiving the first information is data-driven. Without the need to understand the business logic of each network element, it can also provide globally optimized data recommendation services, which reduces the workload and is conducive to improving the adaptability and flexibility of the data recommendation service.

[0121] In one possible implementation, the first information also includes an expected value of a target parameter. For example, the target parameter may be quality of experience, user plane congestion level, or load level of multiple UPFs. Of course, the target parameter may also be other parameters, which are not limited here. The expected value of the target parameter here can be considered as an optimization target. It should be noted that if the first information sent by the first network element does not carry the expected value of the target parameter, the second network element may also obtain the expected value of the target parameter (i.e., the optimization target) through pre-configuration or other means.

[0122] In one possible implementation, the second network element further needs to determine an analysis type identifier, which indicates the type of predictive analysis to be performed on the collected first data. The analysis type identifier can then be used to determine the analysis direction for the first data. For example, the analysis type identifier can indicate any of the following types: service experience analysis, user plane congestion analysis, and network element load analysis. Of course, the analysis type identifier can also indicate other analysis types, which are not limited here.

[0123] The second network element may determine the analysis type identifier in one of the following two ways, which are described in detail below. This way is conducive to improving the flexibility of determining the analysis type identifier.

[0124] Method 1: The first information also includes an analysis type identifier.

[0125] In a specific implementation, after the first network element determines the analysis type identifier, it may carry the analysis type identifier in the first information and send it to the second network element.

[0126] Optionally, the method further includes: the first network element determining the analysis type identifier based on the service processing logic of the first network element.

[0127] For example, assuming that the first network element includes at least one SMF, the at least one SMF expects the second network element to recommend a selection distribution ratio for each UPF to prevent uneven or overloaded UPFs. In this case, the SMF can determine, based on the service processing logic, that an analysis of the operating load level of each UPF is required, and therefore determines that the analysis type identifier indicates a network element load analysis.

[0128] For another example, assume that the first network element includes an AF, a PCF, and an SMF, and the AF, PCF, and SMF expect the second network element to guide them in improving the quality of experience (QoE) of an application. Based on the service processing logic, the AF, PCF, and SMF may determine that an analysis of the QoE of the application is required, and therefore determine that the analysis type identifier indicates a service experience analysis.

[0129] For another example, assuming the first network element includes a PCF, the PCF expects the second network element to adjust QoS parameters to avoid user plane congestion. The PCF may determine based on service processing logic that user plane congestion analysis is required, and thus determine that the analysis type identifier indicates user plane congestion analysis.

[0130] Optionally, the method further includes: the first network element determining the first request and the indication information based on the analysis type identifier.

[0131] For example, assuming that the analysis type identifier indicates network element load analysis, the first request of at least one SMF can be used to request the recommended number of N4 interface sessions to be established on multiple UPFs; the indication information can indicate the number of N4 interface sessions established by at least one SMF on multiple UPFs at the current moment and / or historical moment.

[0132] For another example, assuming the analysis type identifier indicates service experience analysis, the AF's first request may be used to request a recommended server instance for the first application; the PCF's first request may be used to request the first application's recommended peak rate and recommended average rate; and the SMF's first request may be used to request the first application's recommended DNAI. The indication information may indicate the peak rate and average rate of the first application collected at the current time and / or historical time.

[0133] For another example, assuming that the analysis type identifier indicates user plane congestion analysis, the first request of the PCF can be used to request the recommended peak throughput and / or recommended average throughput of the second application; the indication information can indicate the collection of the user plane congestion level of the second application at the current moment and / or historical moments.

[0134] Method 2: The second network element determines the analysis type identifier based on the target parameter and the first request.

[0135] In a specific implementation, the second network element may determine the analysis type identifier according to the target parameter and the first request.

[0136] For example, assuming that the target parameter is the load level of multiple UPFs, the first request is used to request the recommended number of N4 interface sessions to be established on the multiple UPFs; then it can be determined that the analysis type identifier indicates network element load analysis.

[0137] For another example, assuming that the target parameter is quality of experience, the first request to AF is used to request the recommended server instance of the first application, the first request to PCF is used to request the recommended peak rate of the first application and the recommended average rate of the first application, and the first request to SMF is used to request the recommended DNAI of the first application; then it can be determined that the analysis type identifier indicates service experience analysis.

[0138] For another example, assuming that the target parameter is the user plane congestion level, and the first request to the PCF is for requesting a recommended peak throughput and / or recommended average throughput of the second application, it can be determined that the analysis type identifier indicates user plane congestion analysis.

[0139] S602: The second network element obtains a prediction analysis result, where the prediction analysis result is obtained based on the first data.

[0140] In an embodiment of the present application, after receiving the first information, the second network element further obtains a predictive analysis result based on the first data. The second network element here can be a recommendation service function (the recommendation service function can be considered to be an independent network element); the second network element can also include both a recommendation service function and a predictive analysis function (the recommendation service function and the predictive analysis function can be considered to be concentrated in the same network element, that is, the recommendation service function and the predictive analysis function are the same data analysis function). The following describes the specific implementation methods for the second network element to obtain the predictive analysis result for these two situations.

[0141] Case 1: The second network element is a recommendation service function.

[0142] The specific implementation method for the second network element to obtain the prediction analysis result may include the following steps s11 to s14, as shown in Figure 7A. Based on this method, the prediction analysis result can reflect the impact of the network adjustment behavior on the network, which is conducive to improving the accuracy of the prediction analysis result, thereby further improving the efficiency of the recommendation service.

[0143] s11. The recommendation service function sends a second request to the prediction analysis function; the second request is used to request the prediction analysis result, and the second request includes an analysis type identifier and indication information. Correspondingly, the prediction analysis function receives the second request from the recommendation service function.

[0144] In a specific implementation, the second request here can be considered as a prediction analysis subscription request. When the second network element is a recommendation service function, the recommendation service function can request the prediction analysis function for the prediction analysis result.

[0145] Optionally, the method further includes: the recommendation service function determines analysis filtering information based on the first request; the analysis filtering information is used to indicate a network range corresponding to the predictive analysis result.

[0146] Specifically, the first request includes indication information, and the analysis filtering information can be determined based on the indication information for collecting the first data in the first request. For example, based on the indication information for obtaining the recommended DNAI of the first application, the peak rate and the average rate of the first application in the first request, the analysis filtering information can be determined to be the first application; or, based on the indication information of the data source of the first data in the first request, the analysis filtering information can be determined. For example, based on the fact that one or more UPFs are specified based on the data source of the first data in the first request, the analysis filtering information can be determined to be a list of network element instances containing the specified one or more UPFs. The second request also includes the analysis filtering information. Based on this approach, the efficiency of predictive analysis can be improved.

[0147] It should be noted that the application identifier (App ID) can be used as analysis and filtering information, in which case the service experience of the application corresponding to the application identifier needs to be analyzed and predicted; a certain network area or a certain type of network element function can also be used as analysis and filtering information, in which case the service experience of this network area or this type of network element function needs to be analyzed and predicted. Of course, other forms of expression of analysis and filtering information can also be used, and this is not limited here.

[0148] s12. The predictive analysis function collects the first data and second data corresponding to the analysis type identifier; wherein the first data belongs to the second data, or the first data does not belong to the second data.

[0149] In a specific implementation, in addition to collecting the specified first data, the predictive analysis function also needs to collect the second data corresponding to the analysis type identifier. The second data here can be considered as the network data that needs to be collected for the analysis type identifier. Among them, the first data specified for collection may belong to the network data that needs to be collected by the analysis type identifier, or may not belong to the network data that needs to be collected by the analysis type identifier, and this is not limited here. For example, when the analysis type identifier is network element load analysis, the first data indicates that the number of N4 interface sessions established on the collected UPF does not belong to the second data corresponding to the network element load analysis type identifier; for another example, when the analysis type identifier is service experience analysis, the first data indicates that the collected peak rate and average rate of the first application belong to the second data corresponding to the network element load analysis type identifier.

[0150] Further optionally, the indication information is also used to indicate the data source of the first data; the specific implementation method of the predictive analysis function collecting the first data may be: the predictive analysis function collects the first data from the data source. It can be understood that the predictive analysis function can collect the first data from a specified data source (i.e., one or more specified network elements or a specified network range as the source of network data). Based on this method, the first network element can use one or more network elements that can best reflect the impact of the network adjustment behavior on the network as the source of the first data according to the business processing logic. In the case where the recommendation service function does not need to understand the business logic of each network element, the efficiency of collecting data is improved, the workload is further reduced, and the adaptability and flexibility of the data recommendation service are improved.

[0151] s13. The prediction analysis function determines a prediction analysis result based on the first data and the second data.

[0152] In a specific implementation, the predictive analysis function can use the first data and the second data to perform offline model training to obtain a predictive analysis model. Alternatively, the predictive analysis model can be obtained from a network element such as a model training logic function (MTLF), which is not limited here. The predictive analysis model is then used to process the data collected at the current moment in the first data to obtain a predictive analysis result.

[0153] In one possible implementation, the specific implementation method of the predictive analysis function determining the predictive analysis result based on the first data and the second data may be: increasing the weight of the first data relative to other data in the second data except the first data according to the indication information; or, increasing the weight of the first data relative to the second data according to the indication information; determining the predictive analysis result based on the weight of the first data, the weight of the second data, the first data and the second data.

[0154] It can be understood that when the first data belongs to the second data, in the process of obtaining the predictive analysis results using the predictive analysis model, the predictive analysis function can increase the weight of the first data relative to the other data in the second data except the first data according to the instructions for collecting the first data (it can be understood that the weight of the first data is higher than the weight of the other data in the second data except the first data), so that the predictive analysis results can more accurately reflect the impact of the network adjustment behavior on the network; when the first data does not belong to the second data, in addition to being based on the second data, it is also necessary to obtain the predictive analysis results based on the first data at the same time, and increase the weight of the first data relative to the second data according to the instructions for collecting the first data (it can be understood that the weight of the first data is higher than the weight of the second data), so that the predictive analysis results can reflect the impact of the network adjustment behavior on the network. Based on this method, it is beneficial to improve the accuracy of the predictive analysis results, thereby further improving the efficiency of the recommendation service.

[0155] s14. The prediction analysis function sends the prediction analysis result to the recommendation service function. Correspondingly, the recommendation service function receives the prediction analysis result from the prediction analysis function.

[0156] Case 2: The second network element includes both recommendation service functions and prediction analysis functions.

[0157] The specific implementation method of the second network element obtaining the prediction analysis result may include the following steps s21 and s22, as shown in Figure 7B. Based on this method, it is beneficial to improve the accuracy of the prediction analysis result and improve efficiency.

[0158] s21. The second network element collects the first data and the second data corresponding to the analysis type identifier; wherein the first data belongs to the second data, or the first data does not belong to the second data.

[0159] In a specific implementation, when the second network element includes a recommendation service function and a predictive analysis function, the second network element can collect the specified first data and the second data corresponding to the analysis type identifier. The second data here can be considered as the network data required to be collected for the analysis type identifier. The first data specified for collection may or may not be included in the network data required to be collected for the analysis type identifier, and this is not limited here.

[0160] s22. The second network element determines a prediction analysis result based on the first data and the second data.

[0161] In a specific implementation, the second network element can perform offline model training using the first and second data to obtain a predictive analysis model. Alternatively, the second network element can obtain the predictive analysis model from a network element such as the MTLF, without limitation. The predictive analysis model is then used to process the data collected at the current moment in the first data to obtain a predictive analysis result.

[0162] In one possible implementation, determining a prediction analysis result based on the first data and the second data includes: increasing the weight of the first data relative to other data in the second data except the first data according to the indication information; or, increasing the weight of the first data relative to the second data according to the indication information; and determining the prediction analysis result based on the weight of the first data, the weight of the second data, the first data, and the second data.

[0163] It can be understood that when the first data belongs to the second data, in the process of obtaining the prediction analysis results using the prediction analysis model, the recommendation service function can increase the weight of the first data relative to the other data in the second data except the first data according to the instruction of collecting the first data (it can be understood that the weight of the first data is higher than the weight of the other data in the second data except the first data), so that the prediction analysis results can more accurately reflect the impact of the network adjustment behavior on the network; when the first data does not belong to the second data, in addition to being based on the second data, it is also necessary to obtain the prediction analysis results based on the first data at the same time, and the weight of the first data relative to the second data can be increased according to the instruction of collecting the first data (it can be understood that the weight of the first data is higher than the weight of the second data), so that the prediction analysis results can reflect the impact of the network adjustment behavior on the network. Based on this method, it is beneficial to improve the accuracy of the prediction analysis results, thereby further improving the efficiency of the recommendation service.

[0164] S603: The second network element determines recommended data based on the prediction analysis result.

[0165] In a possible implementation, the second network element determines the recommended data based on the prediction analysis result, which may include the following steps s31 to s33, as shown in Figure 7C. Based on this approach, the accuracy of the recommended data can be improved.

[0166] s31. The second network element determines a deviation value based on the prediction analysis result and the expected value of the target parameter.

[0167] In a specific implementation, the data deviation monitoring unit of the recommendation service function is used to calculate the deviation value between the prediction analysis result and the expected value of the target parameter.

[0168] s32. The second network element adjusts the model parameters of the first data optimizer in a direction of reducing the deviation value to obtain a second data optimizer.

[0169] In a specific implementation, the second network element includes a recommendation service function, and the first data optimizer in the recommendation service function can also be a machine learning model with initially configured model parameters. The initially configured model parameters can be obtained from theoretical calculations, or obtained through online learning during the operation of an experimental network or other network, and are not limited here. The first data optimizer continuously adjusts the machine learning model parameters in the direction of reducing the deviation value according to the deviation value, converges to the operation target more quickly, and obtains the final second data optimizer. It should be noted that the second network element can adjust the model parameters of the first data optimizer once every unit time (for example, 1 second or 0.5 minutes, etc.) to obtain the latest second data optimizer. By adjusting the model parameters through online learning during the operation of the network, the first data optimizer becomes more and more accurate.

[0170] s33. The second network element uses the second data optimizer to determine the recommended data.

[0171] In a specific implementation, the second network element further obtains third data at the current moment, such as bit rate, DNAI, application identifier, user permanent identifier, allocated network slice instance identifier, validity period, spatial effectiveness, and other information, which are not limited here. The second network element uses the second data optimizer to process the third data and output the recommended data.

[0172] S604: The second network element sends the recommendation data to the first network element. Correspondingly, the first network element receives the recommendation data from the second network element.

[0173] S605. The first network element determines a first parameter based on the recommendation data.

[0174] In a possible implementation, the first data is a network operation indicator affected by the first parameter.

[0175] It can be understood that after the first network element obtains the recommendation data, it can perform an action according to the recommendation data, that is, determine the first parameter to achieve the network operation goal.

[0176] For example, assuming that the first network element includes at least one SMF, and the recommended data for the at least one SMF is the number of N4 interface sessions recommended to be established on multiple UPFs, then the at least one SMF can determine the selection distribution ratio (i.e., the first parameter) for the multiple UPFs based on the recommended data.

[0177] For another example, assuming that the first network element includes AF, PCF, and SMF, the recommended data for AF is the recommended server instance of the first application, and AF can select the server instance (i.e., the first parameter) based on the recommended data; the recommended data for PCF is the recommended peak rate of the first application and the recommended average rate of the first application, and PCF can determine the QoS parameters of the application service message (i.e., the first parameter) based on the recommended data, for example, the maximum flow bit rate (MFBR) or the aggregate maximum bit rate (AMBR) based on the recommended peak rate of the first application, and the guaranteed bit rate (GFBR) based on the recommended average rate of the first application; the recommended data for SMF is the recommended DNAI of the first application, and SMF can perform user plane path selection (i.e., the first parameter) based on the recommended data.

[0178] For another example, assuming that the first network element includes a PCF, and the recommended data for the PCF is the recommended peak throughput and / or recommended average throughput of the second application, the PCF can determine the QoS parameters (i.e., first parameters) for several major applications based on the recommended data. For example, the MFBR is set according to the recommended peak throughput of the second application, and the GFBR is set according to the recommended average throughput of the second application. The PCF can also further determine the 5G service quality identifier (5G QoS Identifier, 5QI) used by the application data based on the recommended data.

[0179] It should be noted that steps S602 to S605 are repeatedly executed at intervals of a preset time period (eg, every 30 seconds) until the first network element sends a recommended service unsubscription message to the second network element.

[0180] It can be seen that the method described in FIG6 can provide a globally optimized data recommendation service without having to understand the business logic of each network element, thereby reducing workload and improving the adaptability and flexibility of the data recommendation service.

[0181] The following further describes in detail the network data analysis method and communication device provided by the embodiments of the present application in different scenarios.

[0182] 1. To improve the QoE of the application to the expected value, PCF, SMF, and AF request recommendation data from the recommendation service function.

[0183] Figure 8 is a flow chart of another method for network data analysis provided by an embodiment of the present application. As shown in Figure 8, the method for network data analysis includes the following steps S801 to S811. The execution subject of the method shown in Figure 8 can be the first network element and the second network element. Alternatively, the execution subject of the method shown in Figure 8 can be a chip in the first network element and a chip in the second network element, which is not limited in the embodiment of the present application. Figure 8 takes the first network element and the second network element as the execution subject of the method as an example. Among them, the first network element includes PCF, SMF and AF; the second network element includes a recommendation service function, for example, the second network element can be a recommendation service function (the recommendation service function can be considered to be an independent network element), and the second network element can also include both a recommendation service function and a predictive analysis function (the recommendation service function and the predictive analysis function can be considered to be concentrated in the same network element, that is, the recommendation service function and the predictive analysis function are the same data analysis function, such as the NWDAF network element that supports the recommendation service).

[0184] S801: The AF sends first information corresponding to the AF to a second network element. The first information corresponding to the AF includes a first request and instruction information. The first request is used to request recommended data corresponding to the AF, and the instruction information is used to instruct the collection of first data corresponding to the AF. The first data is network data associated with the recommended data determined by the AF. In response, the second network element receives the first information from the AF.

[0185] In an embodiment of the present application, when the AF sends first information corresponding to the AF to the second network element, the first request is used to request recommended data corresponding to the AF, which is a recommended server instance of the first application, and the indication information is used to indicate that the collected first data corresponding to the AF is the peak rate and average rate of the first application at the current time and / or historical time. The indication information may also be carried in the first request, which is not limited here.

[0186] In one possible implementation, the first information corresponding to the AF also includes an expected value of a target parameter. The target parameter is quality of experience. It should be noted that if the first information sent by the AF does not carry the expected value of the target parameter, the second network element may also obtain the expected value of the target parameter (i.e., the optimization target) through pre-configuration or other means.

[0187] In a possible implementation, the method further includes: the AF determining, based on a business processing logic of the AF, that the first data is associated with the recommended data.

[0188] In one possible implementation, the second network element also needs to determine the analysis type identifier corresponding to the AF, where the analysis type identifier corresponding to the AF indicates service experience analysis. The second network element can determine the analysis type identifier corresponding to the AF in one of the following two ways, which are described in detail below.

[0189] Method 1: The first information corresponding to the AF also includes an analysis type identifier corresponding to the AF.

[0190] Method 2: The second network element determines the analysis type identifier corresponding to the AF based on the target parameter and the first request corresponding to the AF.

[0191] The specific implementation of step S801 may refer to the specific implementation of step S601 described above, with the main difference being that the first network element is AF, which will not be elaborated here.

[0192] S802: The PCF sends first information corresponding to the PCF to the second network element. The first information corresponding to the PCF includes a first request and instruction information. The first request is used to request recommended data corresponding to the PCF, and the instruction information is used to instruct the collection of first data corresponding to the PCF. The first data is network data associated with the recommended data determined by the PCF. In response, the second network element receives the first information from the PCF.

[0193] In an embodiment of the present application, when PCF sends the first information corresponding to PCF to the second network element, the first request is used to request that the recommended data corresponding to PCF is the recommended peak rate and / or recommended average rate of the first application, and the indication information is used to indicate that the collected first data corresponding to PCF is the peak rate and average rate of the first application at the current moment and / or historical moment.

[0194] In one possible implementation, the first information corresponding to the PCF also includes an expected value of a target parameter. The target parameter is quality of experience. It should be noted that if the first information sent by the PCF does not carry the expected value of the target parameter, the second network element may also obtain the expected value of the target parameter (i.e., the optimization target) through pre-configuration or other means.

[0195] In a possible implementation, the method further includes: the PCF determining, based on a business processing logic of the PCF, that the first data is associated with the recommended data.

[0196] In one possible implementation, the second network element also needs to determine the analysis type identifier corresponding to the PCF, where the analysis type identifier corresponding to the PCF indicates service experience analysis. The second network element can determine the analysis type identifier corresponding to the PCF in one of the following two ways, which are described in detail below.

[0197] Method 1: The first information corresponding to the PCF also includes an analysis type identifier corresponding to the PCF.

[0198] Method 2: The second network element determines the analysis type identifier corresponding to the PCF based on the target parameter and the first request corresponding to the PCF.

[0199] The specific implementation of step S802 may refer to the specific implementation of step S601 above, the main difference being that the first network element is a PCF, which will not be elaborated here.

[0200] S803: The SMF sends first information corresponding to the SMF to the second network element. The first information corresponding to the SMF includes a first request and instruction information. The first request is used to request recommended data corresponding to the SMF, and the instruction information is used to instruct the SMF to collect first data corresponding to the SMF. The first data is network data associated with the recommended data determined by the SMF. In response, the second network element receives the first information from the SMF.

[0201] In an embodiment of the present application, when the SMF sends the first information corresponding to the SMF to the second network element, the first request is used to request that the recommended data corresponding to the SMF is the recommended DNAI of the first application, and the indication information is used to indicate that the collected first data corresponding to the SMF is the peak rate and average rate of the first application at the current moment and / or historical moment.

[0202] In one possible implementation, the first information corresponding to the SMF also includes an expected value of a target parameter. The target parameter is quality of experience. It should be noted that if the first information sent by the SMF does not carry the expected value of the target parameter, the second network element may also obtain the expected value of the target parameter (i.e., the optimization target) through pre-configuration or other means.

[0203] In a possible implementation, the method further includes: the SMF determining, based on the business processing logic of the SMF, that the first data is associated with the recommended data.

[0204] In one possible implementation, the second network element also needs to determine the analysis type identifier corresponding to the SMF, where the analysis type identifier corresponding to the SMF indicates service experience analysis. The second network element can determine the analysis type identifier corresponding to the SMF in one of the following two ways, which are described in detail below.

[0205] Method 1: The first information corresponding to the SMF also includes an analysis type identifier corresponding to the SMF.

[0206] Method 2: The second network element determines the analysis type identifier corresponding to the SMF based on the target parameter and the first request corresponding to the SMF.

[0207] Among them, the specific implementation method of step S803 can refer to the specific implementation method of the above-mentioned step S601. The main difference is that the first network element is SMF, which is not repeated here.

[0208] It should be noted that the execution order of the above steps S801, S802 and S803 is not limited.

[0209] S804. The second network element obtains the prediction analysis result corresponding to the AF, the prediction analysis result corresponding to the PCF, and the prediction analysis result corresponding to the SMF.

[0210] In the embodiment of the present application, the prediction analysis result corresponding to the AF, the prediction analysis result corresponding to the PCF, and the prediction analysis result corresponding to the SMF are obtained based on the first data. The prediction analysis result corresponding to the AF, the prediction analysis result corresponding to the PCF, and the prediction analysis result corresponding to the SMF are all the QoE of the first application at the next moment.

[0211] The second network element can be a recommendation service function (the recommendation service function can be considered an independent network element); the second network element can also include both the recommendation service function and the prediction analysis function (the recommendation service function and the prediction analysis function can be considered to be concentrated in the same network element). The following describes the specific implementation methods for the second network element to obtain the prediction analysis results corresponding to the AF, the prediction analysis results corresponding to the PCF, and the prediction analysis results corresponding to the SMF for these two situations.

[0212] Case 1: The second network element is a recommendation service function.

[0213] The specific implementation method of the second network element obtaining the prediction analysis results corresponding to the AF, the prediction analysis results corresponding to the PCF, and the prediction analysis results corresponding to the SMF may include the following steps s41 to s44, as shown in Figure 9A.

[0214] S41: The recommendation service function sends a second request to the prediction and analysis function. The second request is used to request the prediction and analysis results corresponding to the AF, the PCF, and the SMF. The second request includes an analysis type identifier and indication information. In response, the prediction and analysis function receives the second request from the recommendation service function.

[0215] Optionally, the method further includes: the recommendation service function determining analysis filtering information based on the first request; the analysis filtering information is used to indicate the network range corresponding to the predictive analysis result; and the second request also includes the analysis filtering information. The indication information may be included in the first request. Specifically, the analysis filtering information may be determined based on the indication information for collecting the first data in the first request; or, the analysis filtering information may be determined based on the data source of the first data in the first request.

[0216] Of course, the recommendation service function may also send multiple second requests to the prediction analysis function, requesting the prediction analysis results corresponding to AF, the prediction analysis results corresponding to PCF, and the prediction analysis results corresponding to SMF respectively, which is not limited here.

[0217] s42. The predictive analysis function collects the first data and the second data corresponding to the analysis type identifier; wherein the first data belongs to the second data, or the first data does not belong to the second data. The first data collected by the predictive analysis function is the peak rate and average rate of the first application at the current moment and / or the historical moment, and the first data is collected through the UPF. The second data collected by the predictive analysis function may include the second information collected from the AF and the third information collected from the UPF and / or SMF; the second information includes the experience quality of the first application at the current moment and / or the historical moment, the server instance of the first application and the identifier of the first application; the third information includes the DNAI, user data rate (BitRate), user message delay (PktDelay) and the quality of service business flow identifier (QoSflow identifier, QFI) corresponding to the application message of the first application at the current moment and / or the historical moment.

[0218] s43. The prediction analysis function determines the prediction analysis result corresponding to AF, the prediction analysis result corresponding to PCF, and the prediction analysis result corresponding to SMF based on the first data and the second data.

[0219] In one possible implementation, the specific implementation method of the predictive analysis function determining the predictive analysis results corresponding to AF, the predictive analysis results corresponding to PCF, and the predictive analysis results corresponding to SMF based on the first data and the second data may be: increasing the weight of the first data relative to other data in the second data except the first data according to the indication information; or, increasing the weight of the first data relative to the second data according to the indication information; determining the predictive analysis results corresponding to AF, the predictive analysis results corresponding to PCF, and the predictive analysis results corresponding to SMF based on the weight of the first data, the weight of the second data, the first data, and the second data.

[0220] s44. The prediction analysis function sends the prediction analysis results corresponding to the AF, the PCF, and the SMF to the recommendation service function. Correspondingly, the recommendation service function receives the prediction analysis results corresponding to the AF, the PCF, and the SMF from the prediction analysis function.

[0221] Case 2: The second network element includes both recommendation service functions and prediction analysis functions.

[0222] The specific implementation method of the second network element obtaining the prediction analysis results corresponding to the AF, the prediction analysis results corresponding to the PCF, and the prediction analysis results corresponding to the SMF may include the following steps s51 and s52, as shown in Figure 9B.

[0223] s51. The second network element collects the first data and the second data corresponding to the analysis type identifier; wherein the first data belongs to the second data, or the first data does not belong to the second data.

[0224] The first data collected by the second network element is the peak rate and average rate of the first application at the current moment and / or historical moment, and the first data is collected through the UPF. The second data collected by the second network element may include second information collected from the AF and third information collected from the UPF and / or SMF; the second information includes the experience quality of the first application at the current moment and / or historical moment, the server instance of the first application, and the identifier of the first application; the third information includes the DNAI, user data rate, user message delay, and QFI corresponding to the application message of the first application at the current moment and / or historical moment.

[0225] s52. The second network element determines the prediction analysis result corresponding to AF, the prediction analysis result corresponding to PCF, and the prediction analysis result corresponding to SMF based on the first data and the second data.

[0226] In one possible implementation, the specific implementation method of determining the predictive analysis results corresponding to AF, the predictive analysis results corresponding to PCF, and the predictive analysis results corresponding to SMF based on the first data and the second data may be: increasing the weight of the first data relative to other data in the second data except the first data according to the indication information; or, increasing the weight of the first data relative to the second data according to the indication information; determining the predictive analysis results corresponding to AF, the predictive analysis results corresponding to PCF, and the predictive analysis results corresponding to SMF based on the weight of the first data, the weight of the second data, the first data, and the second data.

[0227] It should be noted that the specific implementation method of the second network element obtaining the prediction and analysis results corresponding to AF, the prediction and analysis results corresponding to PCF, and the prediction and analysis results corresponding to SMF can refer to the specific implementation method of the above step S602. The main difference is that the first network element includes AF, PCF and SMF, which will not be repeated here.

[0228] S805. The second network element determines the recommended data corresponding to the AF based on the prediction analysis result corresponding to the AF, determines the recommended data corresponding to the PCF based on the prediction analysis result corresponding to the PCF, and determines the recommended data corresponding to the SMF based on the prediction analysis result corresponding to the SMF.

[0229] Among them, the specific implementation method of the second network element determining the recommended data corresponding to the AF based on the prediction analysis results corresponding to the AF can refer to the specific implementation method of the above-mentioned step S603. The main difference is that the first network element is AF, which is not described in detail here. The specific implementation method of the second network element determining the recommended data corresponding to the PCF based on the prediction analysis results corresponding to the PCF can also refer to the specific implementation method of the above-mentioned step S603. The main difference is that the first network element is PCF, which is not described in detail here. The specific implementation method of the second network element determining the recommended data corresponding to the SMF based on the prediction analysis results corresponding to the SMF can also refer to the specific implementation method of the above-mentioned step S603. The main difference is that the first network element is SMF, which is not described in detail here.

[0230] S806: The second network element sends the recommendation data corresponding to the AF to the AF. Correspondingly, the AF receives the recommendation data corresponding to the AF from the second network element.

[0231] S807 : The AF determines a first parameter corresponding to the AF based on the recommended data corresponding to the AF.

[0232] In a possible implementation, the first data corresponding to the AF is a network operation indicator affected by a first parameter corresponding to the AF.

[0233] In the embodiment of the present application, for the recommendation data corresponding to the AF being a recommended server instance of the first application, the AF may select a server instance (ie, the first parameter) based on the recommendation data corresponding to the AF.

[0234] S808: The second network element sends the recommendation data corresponding to the PCF to the PCF. Correspondingly, the PCF receives the recommendation data corresponding to the PCF from the second network element.

[0235] S809. The PCF determines a first parameter corresponding to the PCF based on the recommendation data corresponding to the PCF.

[0236] In a possible implementation, the first data corresponding to the PCF is a network operation indicator affected by a first parameter corresponding to the PCF.

[0237] In an embodiment of the present application, the recommended data corresponding to the PCF is the recommended peak rate of the first application and the recommended average rate of the first application. The PCF can determine the QoS parameters (i.e., the first parameters) of the application service message based on the recommended data corresponding to the PCF, for example, determining the MFBR or AMBR based on the recommended peak rate of the first application, and determining the GFBR based on the recommended average rate of the first application.

[0238] S810: The second network element sends the recommended data corresponding to the SMF to the SMF. Correspondingly, the SMF receives the recommended data corresponding to the SMF from the second network element.

[0239] It should be noted that the execution order of steps S806, S808 and S810 is not limited.

[0240] S811. The SMF determines a first parameter corresponding to the SMF based on the recommended data corresponding to the SMF.

[0241] In a possible implementation, the first data corresponding to the SMF is a network operation indicator affected by a first parameter corresponding to the SMF.

[0242] In an embodiment of the present application, the recommended data corresponding to the SMF is the recommended DNAI of the first application, and the SMF can perform user plane path selection (ie, the first parameter) based on the recommended data corresponding to the SMF.

[0243] It should be noted that the execution order of steps S806 and S807, S808 and S809, and S810 and S811 is not limited. In addition, steps S804 to S811 are repeated at every preset time period (for example, every 30 seconds) until the AF, PCF, and SMF send a recommended service unsubscription message to the second network element.

[0244] It can be seen that the method described in Figure 8 can provide a globally optimized data recommendation service without having to understand the business logic of each network element, thereby reducing workload and improving the adaptability and flexibility of the data recommendation service.

[0245] Second, in order to adjust QoS parameters and avoid user plane congestion, the PCF requests recommendation data from the recommendation service function.

[0246] Figure 10 is a flow chart of another method for network data analysis provided in an embodiment of the present application. As shown in Figure 10, the method for network data analysis includes the following steps S1001 to S1005. The execution subject of the method shown in Figure 10 can be the first network element and the second network element. Alternatively, the execution subject of the method shown in Figure 10 can be the chip in the first network element and the chip in the second network element, which is not limited in the embodiment of the present application. Figure 10 takes the first network element and the second network element as the execution subject of the method as an example. Among them, the first network element includes PCF; the second network element can be a recommendation service function (the recommendation service function can be considered to be an independent network element), and the second network element can also include both the recommendation service function and the prediction analysis function (the recommendation service function and the prediction analysis function can be considered to be concentrated in the same network element, that is, the recommendation service function and the prediction analysis function are the same data analysis function, such as the NWDAF network element that supports the recommendation service).

[0247] S1001. A PCF sends first information to a second network element. The first information includes a first request and instruction information. The first request is for requesting recommended data, and the instruction information is for instructing the collection of first data. The first data is network data associated with the recommended data determined by the PCF. In response, the second network element receives the first information from the PCF.

[0248] In an embodiment of the present application, the first request is used to request recommended data for the recommended peak throughput and / or recommended average throughput of the second application. The indication information is used to indicate that the collected first data is the user plane congestion level of the second application at the current time and / or historical time. The indication information may also be carried in the first request, which is not limited here. Optionally, the first request includes an identifier of the second application.

[0249] In one possible implementation, the first information also includes an expected value of a target parameter. The target parameter is a user plane congestion level. It should be noted that if the first information sent by the PCF does not carry the expected value of the target parameter, the second network element may also obtain the expected value of the target parameter (i.e., the optimization target) through pre-configuration or other means.

[0250] In a possible implementation, the method further includes: the PCF determining, based on a business processing logic of the PCF, that the first data is associated with the recommended data.

[0251] In a possible implementation, the second network element also needs to determine an analysis type identifier, where the analysis type identifier indicates user plane congestion analysis. The second network element may determine the analysis type identifier in one of the following two ways, which are described in detail below.

[0252] Method 1: The first information also includes an analysis type identifier.

[0253] Method 2: The second network element determines the analysis type identifier based on the target parameter and the first request.

[0254] The specific implementation of step S1001 may refer to the specific implementation of step S601 above, the main difference being that the first network element is a PCF, which will not be elaborated here.

[0255] S1002. The second network element obtains a prediction analysis result, where the prediction analysis result is obtained based on the first data.

[0256] The prediction analysis result includes the user plane congestion level at the next moment, the peak throughput of the second application at the next moment, and the average throughput of the second application at the next moment.

[0257] The second network element can be a recommendation service function (which can be considered an independent network element); the second network element can also include both the recommendation service function and the prediction analysis function (which can be considered to be integrated into the same network element). The following describes the specific implementation methods for the second network element to obtain the prediction analysis results for these two scenarios.

[0258] Case 1: The second network element is a recommendation service function.

[0259] The specific implementation method of the second network element obtaining the prediction analysis result may include the following steps s11 to s14, as shown in FIG7A .

[0260] s11. The recommendation service function sends a second request to the prediction analysis function; the second request is used to request the prediction analysis result, and the second request includes an analysis type identifier and indication information. Correspondingly, the prediction analysis function receives the second request from the recommendation service function.

[0261] Optionally, the method further includes: the recommendation service function determining analysis filtering information based on the first request; the analysis filtering information is used to indicate the network range corresponding to the predictive analysis result; and the second request also includes the analysis filtering information. The indication information may be included in the first request. Specifically, the analysis filtering information may be determined based on the indication information for collecting the first data in the first request; or, the analysis filtering information may be determined based on the data source of the first data in the first request.

[0262] s12. The predictive analysis function collects the first data and second data corresponding to the analysis type identifier; wherein the first data belongs to the second data, or the first data does not belong to the second data. The first data collected by the predictive analysis function is the user plane congestion level of the second application at the current moment and / or historical moments, and the first data is collected through the UPF. The second data collected by the predictive analysis function includes fourth information collected from the UPF and / or SMF, and the fourth information includes the peak throughput of the second application at the current moment and / or historical moments, the average throughput of the second application, the identifier of the second application, and the location information of the terminal device.

[0263] s13. The prediction analysis function determines a prediction analysis result based on the first data and the second data.

[0264] In one possible implementation, the specific implementation method of the predictive analysis function determining the predictive analysis result based on the first data and the second data may be: increasing the weight of the first data relative to other data in the second data except the first data according to the indication information; or, increasing the weight of the first data relative to the second data according to the indication information; determining the predictive analysis result based on the weight of the first data, the weight of the second data, the first data and the second data.

[0265] s14. The prediction analysis function sends the prediction analysis result to the recommendation service function. Correspondingly, the recommendation service function receives the prediction analysis result from the prediction analysis function.

[0266] Case 2: The second network element includes both recommendation service functions and prediction analysis functions.

[0267] The specific implementation method of the second network element obtaining the prediction analysis result may include the following steps s21 and s22, as shown in Figure 7B.

[0268] s21. The second network element collects the first data and the second data corresponding to the analysis type identifier; wherein the first data belongs to the second data, or the first data does not belong to the second data.

[0269] The first data collected by the predictive analysis function is the user plane congestion level of the second application at the current moment and / or historical moments, and the first data is collected by the UPF. The second data collected by the predictive analysis function includes fourth information collected from the UPF and / or SMF, and the fourth information includes the peak throughput of the second application at the current moment and / or historical moments, the average throughput of the second application, the identifier of the second application, and the location information of the terminal device.

[0270] s22. The second network element determines a prediction analysis result based on the first data and the second data.

[0271] In one possible implementation, determining a prediction analysis result based on the first data and the second data includes: increasing the weight of the first data relative to other data in the second data except the first data according to the indication information; or, increasing the weight of the first data relative to the second data according to the indication information; and determining the prediction analysis result based on the weight of the first data, the weight of the second data, the first data, and the second data.

[0272] It should be noted that the specific implementation of step S1002 can refer to the specific implementation of the above step S602. The main difference is that the first network element is PCF, which is not repeated here.

[0273] S1003. The second network element determines recommended data based on the prediction analysis result.

[0274] The specific implementation of step S1003 may refer to the specific implementation of step S603 above, the main difference being that the first network element is a PCF, which will not be elaborated here.

[0275] S1004: The second network element sends the recommendation data to the PCF. Correspondingly, the PCF receives the recommendation data from the second network element.

[0276] S1005. The PCF determines a first parameter based on the recommendation data.

[0277] In a possible implementation, the first data is a network operation indicator affected by the first parameter.

[0278] In this embodiment of the present application, if the recommended data corresponding to the PCF is the recommended peak throughput and / or recommended average throughput of the second application, the PCF can determine the QoS parameters for several major application categories based on the recommended data. For example, the MFBR can be set according to the recommended peak throughput of the second application, and the GFBR can be set according to the recommended average throughput of the second application. The PCF can also further determine the 5QI used by the application data based on the recommended data.

[0279] It should be noted that steps S1002 to S1005 are repeatedly executed at intervals of a preset time period (eg, every 30 seconds) until the PCF sends a recommended service unsubscription message to the second network element.

[0280] It can be seen that the method described in Figure 10 can provide a globally optimized data recommendation service without understanding the business logic of each network element, reduce workload, and help improve the adaptability and flexibility of the data recommendation service.

[0281] 3. In order to confirm the selection distribution ratio of each UPF and prevent the UPF from being unevenly loaded or overloaded, at least one SMF requests recommendation data from the recommendation service function.

[0282] Figure 11 is a flow chart of another method for network data analysis provided in an embodiment of the present application. As shown in Figure 11, the method for network data analysis includes the following steps S1101 to S1105. The execution subject of the method shown in Figure 11 can be the first network element and the second network element. Alternatively, the execution subject of the method shown in Figure 11 can be a chip in the first network element and a chip in the second network element, which is not limited in the embodiment of the present application. Figure 11 is illustrated by taking the first network element and the second network element as the execution subject of the method as an example. Among them, the first network element includes at least one SMF; the second network element includes a recommendation service function, for example, the second network element can be a recommendation service function (the recommendation service function can be considered to be an independent network element), and the second network element can also include both a recommendation service function and a predictive analysis function (the recommendation service function and the predictive analysis function can be considered to be concentrated in the same network element, that is, the recommendation service function and the predictive analysis function are the same data analysis function, such as a NWDAF network element that supports recommendation services).

[0283] S1101. At least one SMF sends first information to a second network element. The first information includes a first request and instruction information. The first request is for requesting recommended data, and the instruction information is for instructing collection of first data. The first data is network data associated with the recommended data determined by the at least one SMF. In response, the second network element receives the first information from the at least one SMF.

[0284] In an embodiment of the present application, the first request is used to request the number of N4 interface sessions recommended by at least one SMF for establishment on the multiple UPFs. The indication information is used to indicate that the collected first data is the number of N4 interface sessions established by the at least one SMF on the multiple UPFs at the current time and / or at a historical time. The indication information may also be carried in the first request, which is not limited here.

[0285] In one possible implementation, the first information also includes an expected value of a target parameter. The target parameter is the load level of multiple UPFs. It should be noted that if the first information sent by the SMF does not carry the expected value of the target parameter, the second network element may also obtain the expected value of the target parameter (i.e., the optimization target) through pre-configuration or other means.

[0286] In a possible implementation, the method further includes: at least one SMF determining, based on a business processing logic of the at least one SMF, that the first data is associated with the recommended data.

[0287] In a possible implementation, the second network element also needs to determine an analysis type identifier, where the analysis type identifier indicates network element load analysis. The second network element can determine the analysis type identifier in one of the following two ways, which are described in detail below.

[0288] Method 1: The first information also includes an analysis type identifier.

[0289] Method 2: The second network element determines the analysis type identifier based on the target parameter and the first request.

[0290] Among them, the specific implementation method of step S1101 can refer to the specific implementation method of the above-mentioned step S601. The main difference is that the first network element is at least one SMF, which is not repeated here.

[0291] S1102. The second network element obtains a prediction analysis result, where the prediction analysis result is obtained based on the first data.

[0292] The prediction analysis result includes the load levels of multiple UPFs at the next moment.

[0293] The second network element can be a recommendation service function (which can be considered an independent network element); the second network element can also include both the recommendation service function and the prediction analysis function (which can be considered to be integrated into the same network element). The following describes the specific implementation methods for the second network element to obtain the prediction analysis results for these two scenarios.

[0294] Case 1: The second network element is a recommendation service function.

[0295] The specific implementation method of the second network element obtaining the prediction analysis result may include the following steps s11 to s14, as shown in FIG7A .

[0296] s11. The recommendation service function sends a second request to the prediction analysis function; the second request is used to request the prediction analysis result, and the second request includes an analysis type identifier and indication information. Correspondingly, the prediction analysis function receives the second request from the recommendation service function.

[0297] Optionally, the method further includes: the recommendation service function determining analysis filtering information based on the first request; the analysis filtering information is used to indicate the network range corresponding to the predictive analysis result; and the second request also includes the analysis filtering information. The indication information may be included in the first request. Specifically, the analysis filtering information may be determined based on the indication information for collecting the first data in the first request; or, the analysis filtering information may be determined based on the data source of the first data in the first request.

[0298] s12. The predictive analysis function collects the first data and second data corresponding to the analysis type identifier; wherein the first data belongs to the second data, or the first data does not belong to the second data.

[0299] The first data collected by the predictive analysis function is the number of N4 interface sessions established by the at least one SMF on multiple UPFs at the current time and / or at a historical time; this first data is collected by the multiple UPFs. The second data collected by the predictive analysis function includes fifth information collected from the multiple UPFs, which includes load level data of the multiple UPFs at the current time and / or at a historical time.

[0300] s13. The prediction analysis function determines a prediction analysis result based on the first data and the second data.

[0301] In one possible implementation, the specific implementation method of the predictive analysis function determining the predictive analysis result based on the first data and the second data may be: increasing the weight of the first data relative to other data in the second data except the first data according to the indication information; or, increasing the weight of the first data relative to the second data according to the indication information; determining the predictive analysis result based on the weight of the first data, the weight of the second data, the first data and the second data.

[0302] s14. The prediction analysis function sends the prediction analysis result to the recommendation service function. Correspondingly, the recommendation service function receives the prediction analysis result from the prediction analysis function.

[0303] Case 2: The second network element includes both recommendation service functions and prediction analysis functions.

[0304] The specific implementation method of the second network element obtaining the prediction analysis result may include the following steps s21 and s22, as shown in Figure 7B.

[0305] s21. The second network element collects the first data and the second data corresponding to the analysis type identifier; wherein the first data belongs to the second data, or the first data does not belong to the second data.

[0306] The first data collected by the predictive analysis function is the number of N4 interface sessions established by the at least one SMF on multiple UPFs at the current time and / or at a historical time; this first data is collected by the multiple UPFs. The second data collected by the predictive analysis function includes fifth information collected from the multiple UPFs, which includes load level data of the multiple UPFs at the current time and / or at a historical time.

[0307] s22. The second network element determines a prediction analysis result based on the first data and the second data.

[0308] In one possible implementation, determining a prediction analysis result based on the first data and the second data includes: increasing the weight of the first data relative to other data in the second data except the first data according to the indication information; or, increasing the weight of the first data relative to the second data according to the indication information; and determining the prediction analysis result based on the weight of the first data, the weight of the second data, the first data, and the second data.

[0309] It should be noted that the specific implementation of step S1102 can refer to the specific implementation of the above-mentioned step S602. The main difference is that the first network element is at least one SMF, which is not repeated here.

[0310] S1103. The second network element determines recommended data based on the prediction analysis result.

[0311] Among them, the specific implementation method of step S1103 can refer to the specific implementation method of the above-mentioned step S603. The main difference is that the first network element is at least one SMF, which is not repeated here.

[0312] S1104: The second network element sends the recommendation data to at least one SMF. Correspondingly, the PCF receives the recommendation data from the second network element.

[0313] S1105. At least one SMF determines a first parameter based on the recommendation data.

[0314] In a possible implementation, the first data is a network operation indicator affected by the first parameter.

[0315] In an embodiment of the present application, the recommended data corresponding to at least one SMF is the number of N4 interface sessions recommended to be established on multiple UPFs, and at least one SMF can determine the selection distribution ratio (i.e., the first parameter) for the multiple UPFs based on the recommended data.

[0316] It should be noted that steps S1102 to S1105 are repeated at every preset time period (for example, every 30 seconds) until the SMF sends a recommended service unsubscribe message to the second network element.

[0317] It can be seen that the method described in Figure 11 can provide a globally optimized data recommendation service without understanding the business logic of each network element, reduce workload, and help improve the adaptability and flexibility of the data recommendation service.

[0318] Please refer to Figure 12, which shows a structural diagram of a communication device 1200 of an embodiment of the present application. The communication device shown in Figure 12 can be a recommendation service function, a prediction analysis function or a first network element, or it can be a device in the recommendation service function, the prediction analysis function or the first network element, or it can be used in combination with the recommendation service function, the prediction analysis function or the first network element. Specifically, as shown in Figure 12, the communication device 1200 may include a communication unit 1201 and a processing unit 1202. Among them, the processing unit 1202 is used for data processing. The communication unit 1201 is used for communication. Optionally, the communication unit 1201 integrates a receiving unit and a sending unit. The communication unit 1201 can also be called a transceiver unit. Alternatively, the communication unit 1201 can also be split into a receiving unit and a sending unit.

[0319] In one embodiment, the communication device 1200 may be a recommendation service function, or a device in the recommendation service function, or a device that can be used in conjunction with the recommendation service function, wherein:

[0320] The communication unit 1201 is configured to receive first information from a first network element, the first information including a first request and instruction information; the first request is for requesting recommended data, and the instruction information is for instructing collection of first data, where the first data is network data associated with the recommended data determined by the first network element;

[0321] A processing unit 1202 is configured to obtain a prediction analysis result, where the prediction analysis result is obtained based on the first data;

[0322] The processing unit 1202 is configured to determine the recommended data based on the prediction analysis result;

[0323] The communication unit 1201 is further configured to send the recommendation data to the first network element.

[0324] In a possible implementation, the indication information is further used to indicate a data source of the first data. Optionally, the data source of the first data is used to indicate that the first data is collected from one or more specified network elements.

[0325] In a possible implementation manner, the first information further includes an analysis type identifier; or the method further includes: determining the analysis type identifier based on target parameters and the first request.

[0326] In a possible implementation, the analysis type identifier indicates any one of the following types: service experience analysis, user plane congestion analysis, and network element load analysis.

[0327] In a possible implementation, the target parameter includes any one of the following parameters: quality of experience, user plane congestion level, and load levels of multiple UPFs.

[0328] In one possible implementation, the processing unit 1202, when obtaining the predictive analysis result, is specifically used to: collect the first data and the second data corresponding to the analysis type identifier; wherein the first data belongs to the second data, or the first data does not belong to the second data; and determine the predictive analysis result based on the first data and the second data.

[0329] In one possible implementation, the processing unit 1202, when determining the prediction analysis result based on the first data and the second data, is specifically used to: increase the weight of the first data relative to other data in the second data except the first data according to the indication information; or, increase the weight of the first data relative to the second data according to the indication information; determine the prediction analysis result based on the weight of the first data, the weight of the second data, the first data and the second data.

[0330] In one possible implementation, when the processing unit 1202 obtains the predictive analysis result, the communication unit 1201 is used to send a second request to the predictive analysis function; the second request is used to request the predictive analysis result, and the second request includes the analysis type identifier and the indication information; the communication unit 1201 is used to receive the predictive analysis result from the predictive analysis function.

[0331] In one possible implementation, processing unit 1202 is further configured to: determine analysis filtering information based on the first request and the analysis type identifier; the analysis filtering information is used to indicate the network range corresponding to the predictive analysis result; and the second request also includes the analysis filtering information. Specifically, processing unit 1202 determines the analysis filtering information based on the instruction information for collecting the first data in the first request; or, processing unit 1202 determines the analysis filtering information based on the data source of the first data in the first request.

[0332] In a possible implementation, the analysis type identifier indicates service experience analysis; the first network element includes AF, PCF, and SMF; and the first data includes a peak rate and an average rate of the first application at a current moment and / or a historical moment.

[0333] In a possible implementation, when the first network element is an AF, the recommendation data includes a recommended server instance of the first application.

[0334] In a possible implementation, when the first network element is a PCF, the recommendation data includes a recommended peak rate and a recommended average rate of the first application.

[0335] In a possible implementation, when the first network element is an SMF, the recommendation data includes a recommended data network access identifier of the first application.

[0336] In one possible implementation, the analysis type identifier indicates user plane congestion analysis; the first request includes an identifier of the second application; the first network element includes a PCF; the first data includes a user plane congestion level of the second application at a current moment and / or a historical moment; and the recommended data includes a recommended peak throughput and / or a recommended average throughput of the second application.

[0337] In one possible implementation, the analysis type identifier indicates network element load analysis; the first network element includes at least one SMF; the first data includes the number of N4 interface sessions established by at least one SMF on multiple UPFs at the current moment and / or historical moment; the recommended data includes the number of N4 interface sessions recommended by at least one SMF to be established on multiple UPFs.

[0338] In a possible implementation manner, the first information further includes an expected value of the target parameter.

[0339] In one possible implementation, the processing unit 1202, when determining the recommended data based on the predictive analysis result, is specifically used to: determine a deviation value based on the predictive analysis result and the expected value of the target parameter; adjust the model parameters of the first data optimizer in the direction of reducing the deviation value to obtain a second data optimizer; and use the second data optimizer to determine the recommended data.

[0340] In one embodiment, the communication device 1200 may be a predictive analysis function, or a device in the predictive analysis function, or a device that can be used in conjunction with the predictive analysis function, wherein:

[0341] Communication unit 1201 is configured to receive a second request from the recommendation service function; the second request is configured to request a prediction analysis result, the second request includes an analysis type identifier and instruction information, the instruction information is configured to instruct collection of first data, the first data being network data determined by the first network element to be associated with recommendation data output by the recommendation service function;

[0342] The processing unit 1202 is configured to collect the first data and second data corresponding to the analysis type identifier; wherein the first data belongs to the second data, or the first data does not belong to the second data;

[0343] The processing unit 1202 is further configured to determine a prediction analysis result based on the first data and the second data;

[0344] The communication unit 1201 is further configured to send the prediction analysis result to the recommendation service function.

[0345] In one possible implementation, the indication information is further used to indicate a data source of the first data; and collecting the first data includes: collecting the first data from the data source. Optionally, the data source of the first data is used to indicate that the first data is collected from one or more specified network elements.

[0346] In one possible implementation, determining a prediction analysis result based on the first data and the second data includes: increasing the weight of the first data relative to other data in the second data except the first data according to the indication information; or, increasing the weight of the first data relative to the second data according to the indication information; and determining the prediction analysis result based on the weight of the first data, the weight of the second data, the first data, and the second data.

[0347] In one possible implementation, the analysis type identifier indicates service experience analysis; the first network element includes AF, PCF and SMF; the first data includes the peak rate and average rate of the first application at the current moment and / or historical moment; the first data is collected through UPF; the second data includes second information collected from AF, and third information collected from UPF and / or SMF; the second information includes the experience quality of the first application at the current moment and / or historical moment, the server instance of the first application and the identifier of the first application; the third information includes the data network access identifier, user data rate, user message delay and service quality service flow identifier corresponding to the application message of the first application at the current moment and / or historical moment.

[0348] In a possible implementation, the second request further includes analysis filtering information, where the analysis filtering information is used to indicate a network range corresponding to the analysis prediction result.

[0349] In one possible implementation, the analysis type identifier indicates user plane congestion analysis; the first network element includes PCF; the first data includes the user plane congestion level of the second application at the current moment and / or historical moment; the first data is collected through UPF; the second data includes fourth information collected from UPF and / or SMF; the fourth information includes the peak throughput of the second application at the current moment and / or historical moment, the average throughput of the second application, the identifier of the second application, and the location information of the terminal device.

[0350] In one possible implementation, the analysis type is identified as network element load analysis; the first network element includes at least one SMF; the first data includes the number of N4 interface sessions established by the at least one SMF on multiple UPFs at the current moment and / or historical moments; the first data is collected through the multiple UPFs; the second data includes fifth information collected from the multiple UPFs; the fifth information includes load level data of the multiple UPFs at the current moment and / or historical moments.

[0351] In one embodiment, the communication device 1200 may be a first network element, or a device in the first network element, or a device that can be used in conjunction with the first network element, wherein:

[0352] The communication unit 1201 is configured to send first information to the recommendation service function, the first information including a first request and instruction information; the first request is for requesting recommended data, and the instruction information is for instructing to collect first data, where the first data is network data associated with the recommended data determined by the first network element;

[0353] The communication unit 1201 is further configured to receive recommendation data from the recommendation service function;

[0354] The processing unit 1202 is configured to determine a first parameter based on the recommendation data.

[0355] In a possible implementation, the processing unit 1202 is further configured to: determine, based on the service processing logic of the first network element, that the first data is associated with the recommended data.

[0356] In a possible implementation, the first data is a network operation indicator affected by the first parameter.

[0357] In a possible implementation, the first information further includes an analysis type identifier. Optionally, the analysis type identifier indicates any one of the following types: service experience analysis, user plane congestion analysis, and network element load analysis.

[0358] In a possible implementation, the processing unit 1202 is further configured to: determine an analysis type identifier based on a service processing logic of the first network element; and determine the first request and the indication information based on the analysis type identifier.

[0359] In a possible implementation, the analysis type identifier indicates service experience analysis; the first network element includes AF, PCF, and SMF; and the first data includes a peak rate and an average rate of the first application at a current moment and / or a historical moment.

[0360] In a possible implementation, when the first network element is an AF, the recommendation data is a recommended server instance of the first application, and the first parameter is a server instance selected by the first application.

[0361] In one possible implementation, when the first network element is a PCF, the recommended data includes a recommended peak rate of the first application and a recommended average rate of the first application; the first parameter includes a maximum stream bit rate or an aggregate maximum rate, and the first parameter also includes a guaranteed bit rate; determining the first parameter based on the recommended data includes: determining the maximum stream bit rate or the aggregate maximum rate based on the recommended peak rate of the first application, and determining the guaranteed bit rate based on the recommended average rate of the first application.

[0362] In a possible implementation, when the first network element is an SMF, the recommended data is a recommended data network access identifier of the first application, and the first parameter is a user plane path selected by the first application.

[0363] In one possible implementation, the analysis type identifier indicates user plane congestion analysis; the first request includes an identifier of the second application; the first network element includes a PCF; the first parameter is a session service quality parameter; the first data includes a user plane congestion level of the second application at a current moment and / or a historical moment; and the recommended data includes a recommended peak throughput and / or a recommended average throughput of the second application.

[0364] In one possible implementation, the analysis type identifier indicates network element load analysis; the first network element includes at least one SMF; the first parameter includes the selection ratio of the at least one SMF for multiple UPFs; the first data includes the number of N4 interface sessions established by the at least one SMF on the multiple UPFs at the current moment and / or historical moment; the recommended data includes the number of N4 interface sessions recommended by the at least one SMF to be established on the multiple UPFs.

[0365] In a possible implementation, the indication information is further used to indicate a data source of the first data. Optionally, the data source of the first data is used to indicate that the first data is collected from one or more specified network elements.

[0366] In a possible implementation, the first information further includes an expected value of a target parameter. Optionally, the target parameter includes any one of the following parameters: quality of experience, user plane congestion level, and load levels of multiple UPFs.

[0367] Figure 13 shows a schematic diagram of the structure of another communication device. The communication device 1300 can be the recommended service function, predictive analysis function, or first network element in the above-mentioned method embodiments, or can also be a chip, chip system, or processor that supports the recommended service function, predictive analysis function, or first network element to implement the above-mentioned method. This communication device can be used to implement the method described in the above-mentioned method embodiments. For details, please refer to the description of the above-mentioned method embodiments.

[0368] The communication device 1300 may include one or more processors 1301. The processor 1301 may be a general-purpose processor or a dedicated processor. For example, it may be a baseband processor or a central processing unit (CPU). The baseband processor may be used to process communication protocols and communication data, while the CPU may be used to control the communication device (e.g., a base station, a baseband chip, a terminal, a terminal chip, a DU or a CU), execute software programs, and process software program data.

[0369] Optionally, the communication device 1300 may include one or more memories 1302, on which instructions 1304 may be stored. The instructions may be executed on the processor 1301, causing the communication device 1300 to perform the method described in the above method embodiment. Optionally, the memory 1302 may also store data. The processor 1301 and memory 1302 may be provided separately or integrated together.

[0370] Optionally, the communication device 1300 may further include a transceiver 1305 and an antenna 1306. The transceiver 1305 may be referred to as a transceiver unit, a transceiver, or a transceiver circuit, and is configured to implement transceiver functions. The transceiver 1305 may include a receiver and a transmitter. The receiver may be referred to as a receiver or a receiving circuit, and is configured to implement a receiving function; the transmitter may be referred to as a transmitter or a transmitting circuit, and is configured to implement a transmitting function. The processing unit 1202 shown in FIG. 12 may be the processor 1301. The communication unit 1201 may be the transceiver 1305.

[0371] In another possible design, processor 1301 may include a transceiver for implementing receiving and transmitting functions. For example, the transceiver may be a transceiver circuit, an interface, or an interface circuit. The transceiver circuit, interface, or interface circuit for implementing the receiving and transmitting functions may be separate or integrated. The transceiver circuit, interface, or interface circuit may be used for reading and writing code / data, or the transceiver circuit, interface, or interface circuit may be used for transmitting or delivering signals.

[0372] In another possible design, processor 1301 may optionally store instructions 1303. Instructions 1303, when executed on processor 1301, may cause communication device 1300 to perform the method described in the above method embodiment. Instructions 1303 may be fixed in processor 1301. In this case, processor 1301 may be implemented by hardware.

[0373] In another possible design, the communication device 1300 may include a circuit that can implement the functions of sending, receiving, or communicating in the aforementioned method embodiments. The processor and transceiver described in the embodiments of the present application can be implemented in an integrated circuit (IC), an analog IC, a radio frequency integrated circuit RFIC, a mixed signal IC, an application specific integrated circuit (ASIC), a printed circuit board (PCB), an electronic device, etc. The processor and transceiver can also be manufactured using various IC process technologies, such as complementary metal oxide semiconductor (CMOS), N-type metal oxide semiconductor (nMetal-oxide-semiconductor, NMOS), P-type metal oxide semiconductor (positive channel metal oxide semiconductor, PMOS), bipolar junction transistor (Bipolar Junction Transistor, BJT), bipolar CMOS (BiCMOS), silicon germanium (SiGe), gallium arsenide (GaAs), etc.

[0374] The communication device described in the above embodiments may be a recommendation service function, a prediction analysis function, or a first network element, but the scope of the communication device described in the embodiments of the present application is not limited thereto, and the structure of the communication device may not be limited to FIG13. The communication device may be an independent device or may be part of a larger device. For example, the communication device may be:

[0375] (1) An independent integrated circuit (IC), or chip, or chip system or subsystem;

[0376] (2) A set of one or more ICs, optionally including a storage component for storing data and instructions;

[0377] (3) ASIC, such as modem (MSM);

[0378] (4) Modules that can be embedded in other devices;

[0379] (5) Receivers, terminals, smart terminals, cellular phones, wireless devices, handheld devices, mobile units, vehicle-mounted devices, network devices, cloud devices, artificial intelligence devices, etc.;

[0380] (6)Others, etc.

[0381] In the case where the communication device can be a chip or a chip system, please refer to the chip structure diagram shown in Figure 14. The chip 1400 shown in Figure 14 includes a processor 1401 and an interface 1402. Optionally, it may also include a memory 1403. The number of processors 1401 can be one or more, and the number of interfaces 1402 can be multiple.

[0382] For the case where the chip is used to implement the recommendation service function, the prediction analysis function, or the first network element in the embodiments of the present application:

[0383] The interface 1402 is used to receive or output signals;

[0384] The processor 1401 is configured to execute a service recommendation function, a prediction analysis function, or a data processing operation of the first network element.

[0385] It is understandable that some optional features in the embodiments of the present application may, in certain scenarios, be implemented independently of other features, such as the solution on which they are currently based, to solve corresponding technical problems and achieve corresponding effects. They may also be combined with other features in certain scenarios as needed. Accordingly, the communication device provided in the embodiments of the present application may also implement these features or functions accordingly, which will not be described in detail here.

[0386] It should be understood that the processor in the embodiment of the present application can be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method embodiment can be completed by an integrated logic circuit of hardware in the processor or instructions in the form of software. The above processor can be a general-purpose processor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic device, discrete hardware component.

[0387] It is understood that the memory in the embodiments of the present application may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), and direct RAM bus RAM (DR RAM). It should be noted that the memory of the systems and methods described herein is intended to include, but is not limited to, these and any other suitable types of memory.

[0388] The present application also provides a computer-readable medium, in which a computer program or instruction is stored. When the computer program or instruction is executed by a communication device, the functions of any of the above method embodiments are implemented.

[0389] The present application also provides a computer program product including instructions, which enables a computer to implement the functions of any of the above method embodiments when the computer reads and executes the computer program product.

[0390] The present application provides a communication system, which includes a recommendation service function and a first network element; wherein the recommendation service function is used to execute the method for executing the recommendation service function in the above embodiment, and the first network element is used to execute the method executed by the first network element in the above embodiment.

[0391] The present application provides a communication system, which includes a recommendation service function, a prediction and analysis function, and a first network element; wherein the recommendation service function is used to execute the method for executing the recommendation service function in the above-mentioned embodiment, the prediction and analysis function is used to execute the method for executing the prediction and analysis function in the above-mentioned embodiment, and the first network element is used to execute the method executed by the first network element in the above-mentioned embodiment.

[0392] In the above embodiments, all or part of the embodiments may be implemented by software, hardware, firmware, or any combination thereof. When implemented using software, all or part of the embodiments may be implemented in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present application are generated. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via a wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) method. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrated therein. The available medium may be a magnetic medium (eg, a floppy disk, a hard disk, a magnetic tape), an optical medium (eg, a high-density digital video disc (DVD)), or a semiconductor medium (eg, a solid state disk (SSD)).

[0393] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. A method for network data analysis, characterized in that, The method is applied to the recommendation service function, and the method includes: Receiving first information from a first network element, where the first information includes a first request and indication information; the first request is used to request recommendation data, and the indication information is used to indicate collecting first data, where the first data is network data determined by the first network element and associated with the recommendation data; Obtaining a predictive analysis result, where the predictive analysis result is obtained based on the first data; Determining the recommendation data based on the predictive analysis result; Sending the recommendation data to the first network element.

2. The method according to claim 1, characterized in that, The indication information is further used to indicate the data source of the first data.

3. The method according to claim 1 or 2, characterized in that, The first information further includes an analysis type identifier; or, The method further includes: Determining the analysis type identifier based on a target parameter and the first request.

4. The method according to any one of claims 1 to 3, characterized in that The obtaining the predictive analysis result includes: Collecting second data corresponding to the first data and the analysis type identifier; where the first data belongs to the second data, or the first data does not belong to the second data; Determining the predictive analysis result based on the first data and the second data.

5. The method according to any one of claims 1 to 3, characterized in that The obtaining the predictive analysis result includes: Sending a second request to a predictive analysis function; the second request is used to request a predictive analysis result, and the second request includes the analysis type identifier and the indication information; Receiving the predictive analysis result from the predictive analysis function.

6. The method according to claim 5, wherein The method further includes: Determining analysis filtering information based on the first request; the analysis filtering information is used to indicate the network scope corresponding to the predictive analysis result; The second request further includes the analysis filtering information.

7. The method according to any one of claims 3 to 6, characterized in that The analysis type identifier indicates service experience analysis; the first network element includes an Application Function (AF), a Policy Control Function (PCF), and a Session Management Function (SMF); the first data includes the peak rate and average rate of a first application at the current moment and / or historical moments.

8. The method according to claim 7, wherein When the first network element is the AF, the recommendation data includes a recommended server instance of the first application.

9. The method according to claim 7, wherein When the first network element is the PCF, the recommendation data includes a recommended peak rate and a recommended average rate of the first application.

10. The method according to claim 7, wherein When the first network element is the SMF, the recommendation data includes a recommended data network access identifier of the first application.

11. The method according to any one of claims 3 to 6, characterized in that The analysis type identifier indicates user plane congestion analysis; the first request includes an identifier of a second application; The first network element includes a PCF; the first data includes the user plane congestion level of the second application at the current moment and / or historical moments; the recommendation data includes a recommended peak throughput and / or a recommended average throughput of the second application.

12. The method according to any one of claims 3 to 6, characterized in that, The analysis type identifier indicates network element load analysis; the first network element includes at least one SMF; The first data includes the number of N4 interface sessions established by the at least one SMF on multiple User Plane Functions (UPFs) at the current moment and / or historical moments; the recommendation data includes the number of N4 interface sessions recommended to be established by the at least one SMF on the multiple UPFs.

13. The method according to any one of claims 1 to 12, characterized in that, The first information further includes an expected value of the target parameter.

14. The method according to claim 13, wherein Determining the recommended data based on the prediction analysis result includes: Determining a deviation value based on the prediction analysis result and the expected value of the target parameter; Adjusting the model parameters of the first data optimizer in the direction of reducing the deviation value to obtain a second data optimizer; Using the second data optimizer to determine the recommended data.

15. A method for network data analysis, characterized in that, The method is applied to a prediction analysis function, and the method includes: Receiving a second request from a recommendation service function; the second request is used to request a prediction analysis result, and the second request includes an analysis type identifier and indication information, and the indication information is used to indicate collecting first data, where the first data is network data determined by a first network element and associated with the recommended data output by the recommendation service function; Collecting the first data and second data corresponding to the analysis type identifier; where the first data belongs to the second data, or the first data does not belong to the second data; Determining a prediction analysis result based on the first data and the second data; Sending the prediction analysis result to the recommendation service function.

16. The method according to claim 15, wherein The indication information is further used to indicate the data source of the first data; The collecting the first data includes: Collecting the first data from the data source.

17. The method according to claim 15 or 16, characterized in that, The determining a prediction analysis result based on the first data and the second data includes: Enhancing the weight of the first data relative to other data in the second data except the first data according to the indication information; or enhancing the weight of the first data relative to the second data according to the indication information; Determining a prediction analysis result based on the weight of the first data, the weight of the second data, the first data, and the second data.

18. A method for network data analysis, characterized in that, The method is applied to a first network element, and the method includes: Sending first information to a recommendation service function, where the first information includes a first request and indication information; the first request is used to request recommended data, and the indication information is used to indicate collecting first data, where the first data is network data determined by the first network element and associated with the recommended data; Receiving the recommended data from the recommendation service function; Determining a first parameter based on the recommended data.

19. The method according to claim 18, characterized in that, The method further includes: Determining that the first data is associated with the recommended data based on the service processing logic of the first network element.

20. The method according to claim 18 or 19, characterized in that, The first data is a network operation index affected by the first parameter.

21. The method according to any one of claims 18 to 20, characterized in that, The first information further includes an analysis type identifier.

22. The method according to claim 21, wherein The method further includes: Determining the analysis type identifier based on the service processing logic of the first network element; Determining the first request and the indication information based on the analysis type identifier.

23. The method according to any one of claims 18 to 22, characterized in that, The indication information is further used to indicate the data source of the first data.

24. A communication system, characterized in that, Including a recommendation service function and a first network element; where the recommendation service function is used to execute the method according to any one of claims 1 to 14, and the first network element is used to execute the method according to any one of claims 18 to 23.

25. A communication system, characterized in that, It includes a recommendation service function, a predictive analysis function, and a first network element; wherein, the recommendation service function is used to execute the method described in any one of claims 1 to 14, the predictive analysis function is used to execute the method described in any one of claims 15 to 17, and the first network element is used to execute the method described in any one of claims 18 to 23.

26. A communication device, characterized in that, It includes a unit for executing the method described in any one of claims 1 to 14, or includes a unit for executing the method described in any one of claims 15 to 17, and includes a unit for executing the method described in any one of claims 18 to 23.

27. A communication device, characterized in that, It includes a processor and a memory, the processor and the memory are coupled, the processor is used to implement the method described in any one of claims 1 to 14, or the processor is used to implement the method described in any one of claims 15 to 17, or the processor is used to implement the method described in any one of claims 18 to 23.

28. A chip, characterized in that, It includes a processor and an interface, the processor and the interface are coupled; the interface is used to receive or output signals, the processor is used to execute code instructions to cause the method described in any one of claims 1 to 14 to be executed, or to cause the method described in any one of claims 15 to 17 to be executed, or the processor is used to implement the method described in any one of claims 18 to 23.

29. A computer-readable storage medium, characterized in that, Computer-executable instructions are stored in the computer-readable storage medium, and when the computer-executable instructions are called by the computer, they cause the computer to execute the method described in any one of claims 1 to 14 above, or cause the computer to execute the method described in any one of claims 15 to 17 above, or cause the computer to execute the method described in any one of claims 18 to 23 above.

Images