Method for determining service experience model and communication device

Abstract

The embodiment of the present application provides a method for determining a service experience model and a communication device. The method includes: a data analysis network element obtains first associated information corresponding to a first data set, the first data set includes data of the service on the core network network element; the data analysis network element obtains a second data set based on the first associated information; the data analysis network element determines a fourth data set based on the first information and the second data set, the fourth data set is a subset or all of the second data set, the first information includes capability information of the first device and / or capability information of the data analysis network element; the data analysis network element determines the service experience model of the service based on the fourth data set. In this solution, the data analysis network element obtains the data of the service of the terminal device on the core network network element, and aligns the data with the first device, and then performs model training based on the aligned data to obtain an accurate service experience model. The data analysis network element does not need to obtain the data of the service of the terminal device on the first device.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of communication technologies, and in particular to a method for determining a service experience model and a communication device. Background Art

[0002] In fifth-generation (5G) network communications, service data of terminal devices is distributed across core network elements, as well as access network devices and / or third-party devices.

[0003] In order to evaluate the service experience, it is necessary to train a model based on the service data of terminal devices distributed on different devices or network elements to obtain a service experience model, and then evaluate the service experience based on the service experience model.

[0004] One implementation method involves having data analysis elements within the core network train models based on service data from terminal devices distributed across different devices or elements. However, due to data confidentiality requirements, data analysis elements cannot access complete data from core network elements, access network devices, and third-party devices, making accurate model training impossible. Summary of the Invention

[0005] The embodiments of the present application provide a method for determining a service experience model and a communication device for accurately determining the service experience model of a service.

[0006] In a first aspect, an embodiment of the present application provides a method for determining a service experience model, including: a data analysis network element obtains first associated information corresponding to a first data set, the first data set including data of the service on a core network network element; the data analysis network element obtains a second data set based on the first associated information, the second data set including data corresponding to the second associated information in the first data set, the second associated information is the intersection of the first associated information and the third associated information, the third associated information corresponds to a third data set, and the third data set includes data of the service on the first device; the data analysis network element determines a fourth data set based on the first information and the second data set, the fourth data set being a subset or all of the second data set, the first information including capability information of the first device and / or capability information of the data analysis network element; the data analysis network element determines the service experience model of the service based on the fourth data set.

[0007] Based on the above solution, the data analysis network element obtains the data of the terminal device's business on the core network network element, aligns the data with the first device, and then performs model training based on the aligned data to obtain an accurate business experience model, and the data analysis network element does not need to obtain the data of the terminal device's business on the first device.

[0008] In a possible implementation method, the fourth data set includes a training set, which corresponds to fourth association information. The data analysis network element determines the service experience model of the service based on the fourth data set, including: the data analysis network element obtains the training set based on the fourth association information and the fourth data set; the data analysis network element determines at least one candidate service experience model based on the training set; the data analysis network element determines the service experience model of the service from the at least one candidate service experience model.

[0009] In a possible implementation method, the fourth data set also includes a verification set, which corresponds to the fifth association information. The data analysis network element determines the business experience model of the business from the at least one candidate business experience model, including: the data analysis network element obtains the verification set based on the fifth association information and the fourth data set; the data analysis network element determines the verification results corresponding to the at least one candidate business experience model based on the verification set; the data analysis network element determines the business experience model of the business based on the verification results corresponding to the at least one candidate business experience model.

[0010] In a possible implementation method, the fourth data set also includes a test set, which corresponds to the sixth association information. The method also includes: the data analysis network element obtains the test set based on the sixth association information and the fourth data set; the data analysis network element determines the test results of the service experience model of the service based on the test set.

[0011] In a possible implementation method, the data analysis network element sends the fourth association information corresponding to the training set to the first device.

[0012] In one possible implementation method, the data analysis network element sends a first request to the second device, where the first request carries identification information of the first device and is used to request the first data set; the data analysis network element receives the first data set from the second device.

[0013] In a possible implementation method, the data analysis network element obtains the second data set based on the first association information, including: the data analysis network element sends the first association information to the first device; the data analysis network element receives the second association information from the first device; the data analysis network element obtains the second data set based on the second association information and the first data set.

[0014] In a possible implementation method, the data analysis network element obtains the second data set based on the first association information, including: the data analysis network element receives the third association information from the first device; the data analysis network element determines the second association information based on the third association information and the first association information; the data analysis network element obtains the second data set based on the second association information and the first data set.

[0015] In a possible implementation method, the data analysis network element obtains the first association information corresponding to the first data set, including: the data analysis network element sends a second request to the second device, the second request carries the identification information of the first device, and the second request is used to request the first association information corresponding to the first data set; the data analysis network element receives the first association information from the second device.

[0016] In a possible implementation method, the data analysis network element obtains the second data set based on the first association information, including: the data analysis network element determines the second association information based on the first association information; the data analysis network element sends a third request to the second device, the third request carries the second association information, and the third request is used to request the second data set; the data analysis network element receives the second data set from the second device.

[0017] Based on this solution, when the data volume of the second data set is less than the data volume of the first data set, the method can reduce the data transmission volume, thereby alleviating the data transmission pressure.

[0018] In a possible implementation method, the data analysis network element sends a fourth request to the network element storage function network element, where the fourth request is used to request the address information of the second device; the data analysis network element receives the address information of the second device from the network element storage function network element.

[0019] In a possible implementation method, the second device is a data analysis network element supporting a data lake function, or a data analysis network element supporting a data collection and coordination function, or a data analysis network element supporting a data collection function.

[0020] In a possible implementation method, the first device is an access network device or a service device.

[0021] In a possible implementation method, the first association information includes the following information: identification information of the first device, identification information allocated by the first device to the terminal device, and a timestamp.

[0022] In a possible implementation method, the identification information allocated by the first device to the terminal device is the identification information allocated by the first device to the terminal device on a first interface, and the first interface is an interface between the first device and the core network element.

[0023] In a possible implementation method, the data analysis network element determines the service experience model of the service by using vertical federated learning based on the fourth data set.

[0024] In a second aspect, embodiments of the present application provide a communications device, which may be a data analysis network element or a chip for a data analysis network element. The device has the function of implementing the first aspect described above or various possible implementation methods based on the first aspect. The function may be implemented in hardware or by hardware executing corresponding software. The hardware or software includes one or more modules corresponding to the above-described functions.

[0025] In a third aspect, embodiments of the present application provide a communication device, comprising a processor coupled to a memory, the memory being configured to store programs or instructions. When the programs or instructions are executed by the processor, the device implements the first aspect or various possible implementation methods based on the first aspect. The memory may be located within or outside the device. The processor may include one or more processors.

[0026] In a fourth aspect, an embodiment of the present application provides a communication device, comprising units or means for executing each step of the above-mentioned first aspect or each possible implementation method based on the first aspect.

[0027] In a fifth aspect, an embodiment of the present application provides a communication device, comprising a processor and an interface circuit, wherein the processor is configured to control the interface circuit to communicate with other devices and execute the above-mentioned first aspect or various possible implementation methods based on the first aspect. The processor includes one or more.

[0028] In a sixth aspect, an embodiment of the present application further provides a computer-readable storage medium comprising instructions, which, when executed on a computer, enables the computer to execute the above-mentioned first aspect or various possible implementation methods based on the first aspect.

[0029] In a seventh aspect, an embodiment of the present application further provides a computer program product, which, when running on a computer, enables the computer to execute the above-mentioned first aspect or various possible implementation methods based on the first aspect.

[0030] In an eighth aspect, an embodiment of the present application further provides a chip system, comprising a processor coupled to a memory, the memory being used to store programs or instructions. When the programs or instructions are executed by the processor, the chip system implements the first aspect or various possible implementation methods based on the first aspect. The memory may be located within or outside the chip system. The processor may include one or more processors. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1A schematic diagram of a 5G network architecture applicable to embodiments of the present application;

[0032] Figure 2 This is a diagram of the service experience evaluation process based on NWDAF;

[0033] Figure 3 Schematic diagram of the training process for vertical federated learning;

[0034] Figure 4 Schematic diagram of data set division;

[0035] Figure 5 This is a diagram of the functional decomposition architecture of NWDAF;

[0036] Figure 6 Schematic diagram of NWDAF obtaining UE data through pairwise association;

[0037] Figure 7 Provided is a method for determining a service experience model for an embodiment of the present application;

[0038] Figure 8 Provided is a method for determining a service experience model for an embodiment of the present application;

[0039] Figure 9 Provided is a method for determining a service experience model for an embodiment of the present application;

[0040] Figure 10 Provided is a method for determining a service experience model for an embodiment of the present application;

[0041] Figure 11 A communication device is provided for an embodiment of the present application;

[0042] Figure 12 Another communication device is provided for an embodiment of the present application. DETAILED DESCRIPTION

[0043] refer to Figure 1 , is a schematic diagram of a 5G network architecture applicable to the embodiments of this application, Figure 1 The 5G network architecture shown in the figure consists of three parts: the terminal device part, the data network (DN) part, and the operator network part. The functions of some of these network elements are briefly described below.

[0044] The operator network may include one or more of the following network elements: an authentication server function (AUSF) network element, a network exposure function (NEF) network element, a policy control function (PCF) network element, a unified data management (UDM) network element, a unified data repository (UDR) network element, a network repository function (NRF) network element, an access and mobility management function (AMF) network element, a session management function (SMF) network element, a radio access network (RAN) user plane function (UPF) network element, and a network data analytics function (NWDAF) network element. In the above-mentioned operator network, the portion other than the radio access network portion may be referred to as the core network portion. In one possible implementation method, the operator network also includes an application function (AF) network element.

[0045] In a specific implementation, the terminal device in the embodiment of the present application may be a device for implementing wireless communication functions. The terminal device may be a user equipment (UE), an access terminal, a terminal unit, a terminal station, a mobile station, a mobile station, a remote station, a remote terminal, a mobile device, a wireless communication device, a terminal agent, or a terminal device in a 5G network or a future evolved public land mobile network (PLMN). An access terminal may be a cellular phone, a cordless phone, a Session Initiation Protocol (SIP) phone, a wireless local loop (WLL) station, a personal digital assistant (PDA), a handheld device with wireless communication capabilities, a computing device or other processing device connected to a wireless modem, an in-vehicle device or a wearable device, 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 smart grids, a wireless terminal in transportation safety, a wireless terminal in smart cities, a wireless terminal in smart homes, etc. The terminal may be mobile or fixed.

[0046] The above-mentioned terminal device can establish a connection with the operator network through the interface provided by the operator network (such as N1, etc.), and use the data and / or voice services provided by the operator network. The terminal device can also access the DN through the operator network, use the operator services deployed on the DN, and / or services provided by a third party. Among them, the above-mentioned third party may be a service provider other than the operator network and the terminal device, and may provide other data and / or voice services to the terminal device. Among them, the specific form of the above-mentioned third party can be determined according to the actual application scenario and is not limited here.

[0047] RAN, as an access network element, is a subnetwork of the operator network and an implementation system between the service nodes and terminal devices in the operator network. To access the operator network, the terminal device must first pass through the RAN, and then connect to the service nodes of the operator network through the RAN. The RAN device in this application is a device that provides wireless communication functions for the terminal device. The RAN device is also called an access network device. The RAN device in this application includes but is not limited to: the next generation base station (gnodeB, gNB) in 5G, evolved node B (evolved node B, eNB), radio network controller (radio network controller, RNC), node B (node ​​B, NB), base station controller (base station controller, BSC), base transceiver station (base transceiver station, BTS), home base station (for example, home evolved nodeB, or homenode B, HNB), baseband unit (baseBand unit, BBU), transmission point (transmitting and receiving point, TRP), transmission point (transmitting point, TP), mobile switching center, etc.

[0048] The AMF network element mainly performs functions such as mobility management and access authentication / authorization. In addition, it is responsible for transmitting user policies between the UE and the PCF.

[0049] The SMF network element mainly performs functions such as session management, execution of control policies issued by the PCF, selection of the UPF, and allocation of UE Internet Protocol (IP) addresses.

[0050] The UPF network element, as the interface UPF with the data network, completes functions such as user plane data forwarding, session / flow-level billing statistics, and bandwidth limitation.

[0051] The UDM network element is mainly responsible for managing contract data, user access authorization and other functions.

[0052] UDR is mainly responsible for the storage and access of contract data, policy data, application data and other types of data.

[0053] NEF network element is mainly used to support the opening of capabilities and events.

[0054] The AF network element primarily communicates application-side requirements to the network, such as Quality of Service (QoS) requirements or user status event subscriptions. The AF can be a third-party functional entity or an application service deployed by an operator, such as the IP Multimedia Subsystem (IMS) voice call service.

[0055] The PCF network element is mainly responsible for policy control functions such as billing at the session and service flow levels, QoS bandwidth guarantee and mobility management, and UE policy decision-making.

[0056] NRF network elements can be used to provide network element discovery capabilities, providing network element information corresponding to the network element type based on requests from other network elements. NRF also provides network element management services such as network element registration, update, and deregistration, as well as network element status subscription and push.

[0057] AUSF network element: Mainly responsible for authenticating users to determine whether users or devices are allowed to access the network.

[0058] The NWDAF network element is mainly used to collect network data (including one or more of terminal device data, RAN device data, core network data, and third-party application data), and provide network data analysis services. It can output data analysis results for use by the network, network management, and application execution policy decisions. NWDAF can use machine learning models for data analysis. In 3GPP Release 17, the functions of NWDAF are decomposed, including data collection function (or data collection logic function), model training function (or machine learning model training logical function), and model reasoning function (or analysis logic function). In scenarios where the data collection function, training function, and reasoning function are separated, the data collection function, training function, and reasoning function of the same model can be deployed separately in different NWDAF instances. An NWDAF deployed with data collection capabilities (which can be called a data collection NWDAF, data lake, or data repository function (DRF)) can be used to collect data from terminal devices, RAN devices, core network devices, and third-party applications. An NWDAF deployed with training capabilities (which can be called a training NWDAF) can train models based on the collected data to obtain trained models. An NWDAF deployed with inference capabilities (referred to as an inference NWDAF) can perform model inference by obtaining the model provided by the training NWDAF and provide data analysis services. The NWDAF can be a separate network element or can be co-located with other network elements, for example, by configuring the NWDAF within the PCF network element.

[0059] A DN is a network located outside of a carrier network. A carrier network can connect to multiple DNs, and a variety of services can be deployed on the DN, providing data and / or voice services to terminal devices. For example, a DN is the private network of a smart factory. Sensors installed in the workshop can be terminal devices. The DN houses a sensor control server, which provides services to the sensors. Sensors can communicate with the control server, receive instructions from the control server, and transmit collected sensor data to the control server based on the instructions. Another example is a DN that is a company's internal office network. An employee's mobile phone or computer can be a terminal device, allowing them to access information and data resources on the company's internal office network.

[0060] Figure 1Among them, Nnwdaf, Nausf, Nnef, Npcf, Nudm, Naf, Namf, Nsmf, N1, N2, N3, N4, and N6 are interface sequence numbers. The meanings of these interface sequence numbers can be found in the 3GPP standard protocol and are not limited here.

[0061] It should be noted that, in the embodiment of the present application, the data analysis network element may be Figure 1 The NWDAF network element shown may also be other network elements in future communication systems that have the functions of the NWDAF network element in this application. The mobility management network element may be Figure 1 The AMF network element shown may also be other network elements in future communication systems that have the functions of the AMF network element in this application. The policy control network element may be Figure 1 The PCF network element shown in the figure may also be other network elements in the future communication system having the functions of the PCF network element in this application. The user plane network element may be Figure 1 The UPF network element shown in the figure may also be other network elements in the future communication system that have the functions of the UPF network element in this application. The application function network element may be Figure 1 The AF network element shown may also be other network elements in future communication systems that have the functions of the AF network element in this application. The access network device may be Figure 1 The RAN device shown may also be other network elements in a future communication system having the functions of the RAN device in this application.

[0062] For ease of explanation, in the embodiments of this application, the data analysis network element is the NWDAF network element, the mobility management network element is the AMF network element, the policy control network element is the PCF network element, the user plane network element is the PCF network element, the application function network element is the AF network element, and the access network device is the RAN device. In addition, the NWDAF network element is further divided into the data lake (also known as the data collection NWDAF), the training NWDAF network element, and the reasoning NWDAF network element. In addition, the terminal device is the UE as an example for explanation.

[0063] To facilitate understanding of the embodiments of the present application, the following first introduces technologies related to the embodiments of the present application.

[0064] 1. Cross-domain data analysis

[0065] Service providers (such as service equipment or AF) are most concerned about the service experience of their services in 5G networks. Accurately understanding their service characteristics allows them to accurately measure the service experience and effectively monitor service quality. However, current 5G networks lack a service experience evaluation mechanism and attempt to guarantee the rich and diverse 5G services through fixed QoS parameters, resulting in an inaccurate match between service experience requirements and network resources.

[0066] like Figure 2 The figure below is a schematic diagram of the service experience evaluation process based on NWDAF. The parameters of the three domains, RAN equipment, core network (CN), and service provider, jointly affect the service experience. NWDAF determines the service experience data analysis results of the service flow and sends them to PCF. PCF determines whether the service experience of the service flow can be met based on the relationship between the service experience data analysis results and the service experience requirements. If not, PCF can re-determine the QoS parameters of the service flow. Based on the new QoS parameters, the network can improve the service experience of the service flow, so that the service experience data analysis results obtained by NWDAF based on the new service experience of the service flow can meet the service experience requirements.

[0067] Table 1 shows examples of parameters that affect service experience.

[0068] Table 1

[0069]

[0070]

[0071] Assuming that all the data in Table 1 can be aggregated into the NWDAF, the NWDAF can train a relationship model between service experience and RAN data, core network (CN) data, and AF data, namely the service experience model. Based on the service experience model, the NWDAF can determine the service experience data analysis results corresponding to the new data from the RAN, core network, and AF. Taking linear regression as an example, the service experience model is as follows:

[0072] h(x)=w0x0+w1x1+w2x2+w3x3+w4x4+w5x5...+w D x D (Formula 1)

[0073] Among them, h(x) represents the service experience. Generally, the larger the value of h(x), the better the service experience. x0 is 1, x i (i=1,2,...,D) represents the data of RAN, CN and AF, D is the dimension of the data, w i (i=0, 1, 2, ..., D) is the weight of each data in affecting the service experience.

[0074] In fact, the private data of RAN, CN, and AF cannot be concentrated in the same NWDAF for centralized training. This is because the private data of RAN, CN, and AF come from different domains. Among them, the private data of RAN comes from the access network domain, the private data of CN comes from the core network domain, and the private data of AF comes from the third-party device domain.

[0075] To train data from different domains, distributed model training can be achieved with the help of Vertical Federated Learning (VFL). As a new machine learning technology, VFL addresses model training and inference challenges when participants are reluctant to share their original data. It is suitable for scenarios where there is significant overlap in the identification (ID) of participants' training samples, but little overlap in their data features. VFL combines the different data features of shared samples from multiple participants for federated learning, meaning that each participant's training data is vertically partitioned, hence the name Vertical Federated Learning.

[0076] The business experience model based on vertical federated learning is as follows:

[0077]

[0078] Among them, x i Represents the i-th sample data, where is the private data distributed on RAN in the i-th sample data, is the private data distributed on CN in the i-th sample data, is the private data distributed on AF in the i-th sample data, is the public data actively reported by RAN, CN and AF in the i-th sample data (such as RSRP / RSRQ / SINR, QoS flow Bit Rate / QoS flow Packet Delay / QoS flow Packet Error Rate, Buffer Size in Table 1), Θ A 、Θ B 、Θ C 、Θ D They are The corresponding model parameters.

[0079] It is worth mentioning that is a data vector consisting of one or more data, correspondingly, Θ A 、Θ B 、Θ C 、Θ D is a parameter vector consisting of one or more model parameters.

[0080] 2. Training Process of Vertical Federated Learning

[0081] Taking the linear regression algorithm as an example, the process of vertical federated learning training is as follows: Figure 3 shown.

[0082] Client A has a dataset Client B has a dataset where y i It is the label data, so the model to be trained is as follows:

[0083]

[0084] Assume that the objective function used for linear regression is as follows,

[0085]

[0086] Where L is the loss function, as follows:

[0087]

[0088] Since the original data D on Client A A and D on Client B B They cannot be aggregated together, so they cannot be trained based on the traditional centralized training method. However, they can be trained based on the vertical federation training method, as follows:

[0089] make Then the L transformation is as follows:

[0090]

[0091] make So

[0092] L=L A +L B +L AB (Formula 7)

[0093] Let the residual Then L is about Θ A and Θ B The gradient is as follows:

[0094]

[0095]

[0096] Accordingly, the model parameters are updated as follows:

[0097]

[0098]

[0099] The training process of vertical federated learning is as follows:

[0100] Step 1: Client A and Client B initialize model parameters Θ respectively Aand Θ B ;

[0101] Step 2, Client A based on Θ A calculate and L A , and then send it to Client B;

[0102] Step 3, Client B based on Θ B calculate Further based on and y i Calculate d i 、L AB 、L B , and finally based on L A 、L AB 、L B Calculate L. Client B will d i Send to Client A;

[0103] Step 4: Client A and Client B each use d i Calculate separately as well as Then based on as well as Update model parameters Θ A and Θ B .

[0104] Among them, steps 2 to 4 are executed in a loop until the model training end condition is met, such as the number of iterations reaches a set threshold (such as 10,000 times) or the value of the loss function L is less than a set threshold (such as 0.001).

[0105] Through the above technology, the interaction of original data between different domains is avoided, and the business experience model can be trained. In the inference phase, Client A and Client B are based on the trained model parameters Θ A and Θ B Compute local inference results as well as Client A then sends the local inference result Sent to Client B, which will finalize the inference result

[0106] 3. Dataset Division and NWDAF Functional Decomposition for Model Training

[0107] refer to Figure 4Figure 2 shows a diagram of dataset partitioning. To prevent overfitting during training, the dataset is divided into a training set, a validation set, and a test set. The training set is used to train different algorithms to obtain corresponding models, while the validation set is used to verify the results of each model. During training, the algorithm can be continuously updated to adjust the model. The best model can be selected based on the validation results. Finally, the test results of the model are determined based on the test set and the best model.

[0108] refer to Figure 5 The figure below shows the functional decomposition architecture of NWDAF. NWDAF can be broken down into the training NWDAF (responsible for model training), the inference NWDAF (responsible for reasoning about data analysis results), and the data lake (responsible for the collection and management of training and inference data).

[0109] 4. NWDAF implements business data analysis through pairwise association

[0110] Typically, the NWDAF performs end-to-end UE-level (per UE) data analysis across the RAN, CN, and AF. It then needs to consider correlating the UE data distributed across the RAN, CN, and AF. The NWDAF can determine UE data by pairwise correlation using correlation information. For example, this embodiment of the present application employs the following approach: RAN-AMF-SMF-UPF-AF.

[0111] refer to Figure 6 , which is a schematic diagram of how the NWDAF obtains UE data through pairwise association. When reporting data, the RAN and AMF both carry a timestamp, the RAN UE NGAP ID assigned to the UE on the N2 interface, and the RAN globally unique identifier (Global RAN Node ID). That is, each piece of data is identified by association information, which includes the Timestamp, RAN UE NGAP ID, and Global RAN Node ID. NGAP is the abbreviation for Next Generation Application Protocol. Therefore, the NWDAF uses the Timestamp, RAN UE NGAP ID, and Global RAN Node ID to associate the UE data on the RAN and AMF.

[0112] Similarly, NWDAF associates the UE data on RAN and UPF through Timestamp and AN Tunnel Info, associates the UE data on SMF and PCF through Timestamp and Subscription Permanent Identifier (SUPI), associates the UE data on AMF and SMF through Timestamp and SUPI, associates the UE data on SMF and UPF through Timestamp and UE IP, and associates the UE data on AF and UPF through Timestamp and Internet Protocol 5-tuple (IP 5-tuple).

[0113] For example, the association information between RAN and AMF is represented by a, the association information between AMF and SMF is represented by b, the association information between SMF and UPF is represented by c, and the association information between UPF and AF is represented by d, then:

[0114] The format of the UE sample data collected by the RAN is (a, UE data);

[0115] The format of the UE sample data collected by the AMF is (a, b, UE data), and then the AMF reports (a, b, UE data) to the NWDAF;

[0116] The format of the UE sample data collected by SMF is (b, c, UE data), and then SMF reports (b, c, UE data) to NWDAF;

[0117] The format of the UE sample data collected by UPF is (c, d, UE data), and then UPF reports (c, d, UE data) to NWDAF;

[0118] The format of the sample data of the UE collected by the AF is (d, UE data), and then the format reported by the AF to the NWDAF is (d, UE data).

[0119] The sample data reported by NWDAF from AMF, SMF, UPF, and AF are: (a, UE data), (a, b, UE data), (b, c, UE data), (c, d, UE data), and (d, UE data). NWDAF can then convert the format of the sample data into (a, UE data) based on the correspondence between a and b, the correspondence between b and c, and the correspondence between c and d.

[0120] An embodiment of the present application provides a method for determining a service experience model, which can be executed by an NWDAF in a core network or a chip used for an NWDAF.

[0121] In this embodiment of the present application, a first data set includes data on services of a terminal device on a core network element. Specifically, the first data set represents a collection of data on services of the terminal device on the core network element. Exemplarily, the core network element may be one or more of a UPF, SMF, AMF, and PCF. The acquired data on the core network element may include the UPF data and AMF data shown in Table 1. First association information is used to associate the first data set with a third data set on a first device (which may be an access network device or a service device (also known as an AF)). The first association information may also be used to identify the data in the first data set. The first association information corresponds to the first data set. Exemplarily, the first association information may include the following information: identification information of the first device, identification information allocated by the first device to the terminal device, and a timestamp. The identification information allocated by the first device to the terminal device may be identification information allocated by the first device to the terminal device on a first interface. The first interface may be an interface between the first device and a core network element. The first device may be an access network device or an AF. For example, if the first device is an access network device, the first association information may include a timestamp, a RAN UE NGAP ID, and a Global RAN Node ID. Among them, Timestamp is a timestamp, RAN UENGAP ID is identification information assigned by the access network device to the terminal device, and Global RAN Node ID is identification information of the access network device.

[0122] In an embodiment of the present application, the third data set includes data on the terminal device's services on the first device (which may be an access network device or an AF), that is, the third data set is used to represent a set of data on the terminal device's services on the first device. Exemplarily, the data obtained on the first device may include the RAN data or AF data shown in Table 1. The third data set corresponds to the same service as the first data set. The third association information is used to associate the third data set with the first data set on the NWDAF. The third association information can also be used to identify the data of the third data set. The third association information corresponds to the third data set. Exemplarily, the third association information may include the following information: identification information of the first device, identification information assigned by the first device to the terminal device, and a timestamp. The identification information assigned by the first device to the terminal device may be identification information assigned by the first device to the terminal device on the first interface, and the first interface may be an interface between the first device and a core network element. Taking the first device as an access network device as an example, the third association information may include: Timestamp, RAN UE NGAP ID, and Global RAN Node ID. Among them, Timestamp is a timestamp, RAN UE NGAP ID is identification information allocated by the access network device to the terminal device, and Global RAN Node ID is identification information of the access network device.

[0123] In this embodiment of the present application, the second dataset includes the data corresponding to the second association information in the first dataset, and the second association information is the intersection of the first association information and the third association information. Therefore, the second dataset is a proper subset of the first dataset or the second dataset is identical to the first dataset.

[0124] For example, the first dataset includes 10,000 pieces of data, and the corresponding first association information is represented by ID-1 to ID-10000. The third dataset includes 9,000 pieces of data, and the corresponding third association information is represented by ID-5001 to ID-14000. Therefore, the second dataset includes 5,000 pieces of data, and the corresponding second association information is represented by ID-5001 to ID-10000.

[0125] In this embodiment of the present application, the fourth dataset is a subset of the second dataset or is identical to the second dataset. The fourth dataset includes a training set, which corresponds to the fourth association information. Optionally, the fourth dataset also includes a validation set and a test set. The validation set corresponds to the fifth association information, and the test set corresponds to the sixth association information.

[0126] like Figure 7 As shown, an embodiment of the present application provides a method for determining a service experience model, the method comprising the following steps:

[0127] Step 701: NWDAF obtains first association information corresponding to a first data set.

[0128] Step 702: NWDAF obtains a second data set according to the first association information.

[0129] Step 703: The NWDAF determines a fourth data set according to the first information and the second data set, where the first information includes capability information of the first device and / or capability information of the NWDAF.

[0130] The capability information here includes one or more of the computing power of the central processing unit (CPU), computing power resources of the graphics processing unit (GPU), memory resources, hard disk resources, transmission bandwidth resources between the NWDAF and the first device (RAN device or AF), or latency.

[0131] Step 704 : NWDAF determines a service experience model of the service based on the fourth data set.

[0132] For example, NWDAF can perform vertical federated learning based on the fourth dataset to obtain a business experience model. Figure 3 , NWDAF can be Figure 3 For client A or client B, the specific model training method can refer to the above description.

[0133] Based on the above solution, NWDAF obtains the data of the terminal device's service on the core network element, aligns the data with the first device, and then performs model training based on the aligned data to obtain an accurate service experience model. NWDAF does not need to obtain the data of the terminal device's service on the first device.

[0134] As an implementation method, before step 702, the NWDAF may also obtain the first data set from the second device. For example, the NWDAF sends a first request to the second device, the first request carrying identification information of the first device. The first request is used to request the first data set, or it can be understood that the first request is used to request service data of a terminal device related to the first device on a core network element. The second device then sends the first data set to the NWDAF.

[0135] In the case where NWDAF obtains the first data set, in the above step 702, the method for NWDAF to obtain the second data set may be, for example: NWDAF sends the first association information to the first device, and the first device determines the second association information based on the third association information and the first association information, that is, the intersection of the third association information and the first association information is determined as the second association information, and then the first device sends the second association information to NWDAF, and then NWDAF obtains the second data set based on the second association information and the first data set, that is, NWDAF obtains the data corresponding to the second association information in the first data set to form the second data set.

[0136] In the case where NWDAF obtains the first data set, in the above step 702, the method for NWDAF to obtain the second data set may also be, for example: the first device sends the third association information to NWDAF, and NWDAF determines the second association information based on the third association information and the first association information, that is, NWDAF determines the intersection of the third association information and the first association information as the second association information, and then NWDAF obtains the second data set based on the second association information and the first data set, that is, NWDAF obtains the data corresponding to the second association information in the first data set to form the second data set.

[0137] As another implementation method, the NWDAF may not need to obtain the first data set from the second device, but instead directly obtain the second data set from the second device. When the data size of the second data set is less than that of the first data set, this method can reduce the amount of data transmitted, thereby alleviating data transmission pressure. Based on this method, step 701 described above may be: the NWDAF sends a second request to the second device, the second request carrying the identification information of the first device, and the second request is used to request first association information corresponding to the first data set. Alternatively, it can be understood that the second request is used to request association information corresponding to service data of a terminal device related to the first device on a core network element, and then the second device sends the first association information to the NWDAF. After the NWDAF obtains the first association information from the second device, the NWDAF aligns the association information with the first device. Specifically, the NWDAF determines the second association information based on the first association information. The specific method can be referred to in the above description. The NWDAF then sends a third request to the second device, the third request carrying the second association information, and the third request is used to request the second data set corresponding to the second association information. The second device obtains the second data set based on the second association information and sends the second data set to the NWDAF. That is, in the above step 702, the NWDAF determines the second association information according to the first association information, and then obtains the second data set from the second device according to the second association information.

[0138] As an implementation method, based on the above implementation method, before the NWDAF obtains the first association information, the first data set, or the second data set from the second device, the NWDAF further sends a fourth request to a network element storage function network element (such as an NRF), where the fourth request is used to request the address information of the second device, and then the network element storage function network element sends the address information of the second device to the NWDAF. The second device may be an NWDAF that supports a data lake function (also referred to as a data lake), or an NWDAF that supports a data collection coordination function, or an NWDAF that supports a data collection function.

[0139] As an implementation method, the above-mentioned fourth data set includes a training set, which corresponds to the fourth association information, that is, the data of the training set is composed of the fourth association information. The training set is a true subset of the fourth data set or the training set is the same as the fourth data set. The training set is used to train different algorithms to obtain models corresponding to each algorithm. Therefore, the above-mentioned step 704 can be: NWDAF obtains the training set based on the fourth association information and the fourth data set, NWDAF determines at least one candidate business experience model based on the training set, and NWDAF determines the business experience model of the business from at least one candidate business experience model. Exemplarily, NWDAF performs model training based on algorithm 1 and the training set to obtain candidate business experience model 1, performs model training based on algorithm 2 and the training set to obtain candidate business experience model 2, and performs model training based on algorithm 3 and the training set to obtain candidate business experience model 3. Then NWDAF determines the business experience model of the business from candidate business experience model 1, candidate business experience model 2 and candidate business experience model 3.

[0140] As an implementation method, the fourth data set also includes a validation set, which corresponds to the fifth association information, that is, the data in the validation set is composed of the fifth association information. The validation set is a proper subset of the fourth data set, or the validation set is the same as the fourth data set. The validation set is used to verify the results of each model, and the algorithm can be continuously updated to adjust the model during the training process. The best model can be selected based on the validation results. The NWDAF determines the service experience model of the service from at least one candidate service experience model. For example, the NWDAF can obtain the validation set based on the fifth association information and the fourth data set, determine the validation results corresponding to at least one candidate service experience model based on the validation set, and determine the service experience model of the service based on the validation results corresponding to the at least one candidate service experience model. For example, the candidate service experience model corresponding to the best validation result is determined as the service experience model. For another example, the validation results are sorted, the top N (N is an integer greater than 1) validation results are taken, a validation result is randomly selected from these N validation results, and the candidate service experience model corresponding to the selected validation result is determined as the service experience model.

[0141] As an implementation method, the fourth dataset also includes a test set, which corresponds to the sixth association information. That is, the data in the test set is determined by the sixth association information. The test set is a true subset of the fourth dataset, or the test set is the same as the fourth dataset. The test set is used to determine the test results of the service experience model. For example, NWDAF can obtain the test set based on the sixth association information and the fourth dataset, and then determine the test results of the service experience model of the service based on the test set.

[0142] As an implementation method, NWDAF can divide the fourth dataset into a training set, a validation set, and a test set, wherein the training set, the validation set, and the test set have no intersection with each other. Optionally, the union of the training set, the validation set, and the test set is equal to the fourth dataset.

[0143] As another implementation method, NWDAF can also divide the fourth dataset into a training set, a validation set, and a test set, and the training set, validation set, and test set can have an intersection. Optionally, the union of the training set, validation set, and test set is equal to the fourth dataset.

[0144] As an implementation method, after the above step 704, NWDAF can also send the fourth association information corresponding to the training set to the first device. The first device can then determine the training set on the first device (also referred to as the fifth data set) based on the fourth association information and the third data set, and perform model training based on the fifth data set to obtain at least one candidate business experience model corresponding to the above business on the first device.

[0145] As an implementation method, after the above step 704, NWDAF can also send the fifth association information corresponding to the verification set to the first device. The first device can then determine the verification set on the first device (also referred to as the sixth data set) based on the fifth association information and the third data set, and verify the above at least one candidate business experience model based on the sixth data set, and select a business experience model from at least one candidate business experience model based on the verification result as the business experience model corresponding to the above business on the first device.

[0146] As an implementation method, after the above step 704, NWDAF can also send the sixth association information corresponding to the test set to the first device. The first device can then determine the test set on the first device (also referred to as the seventh data set) based on the sixth association information and the third data set, and test the above-mentioned business experience model based on the seventh data set to obtain the test results of the business experience model corresponding to the above-mentioned business on the first device.

[0147] The service experience model determined by the first device may be referred to as a service experience sub-model of the first device, and the service experience model determined by the NWDAF may be referred to as a service experience sub-model of the core network.

[0148] Based on the above solution, the first device obtains the data of the terminal device's service on the first device, aligns the data with the NWDAF, and then performs model training based on the aligned data to obtain an accurate service experience model. The first device does not need to obtain the data of the terminal device's service on the core network element.

[0149] The following combination Figures 8 to 10 The corresponding specific embodiment, for the above Figure 7 The method shown is described.

[0150] like Figure 8 The figure shows a schematic diagram of a method for determining a service experience model provided by an embodiment of the present application. Based on this method, a data lake exists on the core network side. The NWDAF retrieves the data of the UE in the core network corresponding to the RAN device from the data lake based on the address information of the RAN device. This part of the UE data can be vertically federated with the data of the UE in the RAN device for federated learning.

[0151] The method comprises the following steps:

[0152] Step 801a: RAN equipment collects and stores data of UEs in the access network.

[0153] The data collected by the RAN device includes a third data set. In a possible implementation, the third data set may include data of multiple UEs, and the data of each UE corresponds to a piece of association information (also referred to as third association information).

[0154] The third association information is the association information between the RAN and the AMF, and includes the Global RAN Node ID, RAN UE NGAP ID, and Timestamp. The Global RAN Node ID is the identifier of the RAN, the RAN UE NGAP ID is the identifier assigned by the RAN device to the UE on the N2 interface, and the Timestamp is the timestamp corresponding to the data collected by the RAN device for the UE.

[0155] Step 801b: The data lake collects and stores the data of UEs in the core network.

[0156] For example, the data lake collects UE data on core network elements such as UPF, AMF, SMF, and PCF. In one possible implementation, when the core network element is UPF and / or AMF, the UE data collected by the data lake can refer to Table 1.

[0157] The following example uses the core network element AMF as an example. In order to ensure that the data of the UE in the access network can be associated with the data of the UE in the core network, when the data lake subscribes to the UE data on the AMF through the Namf_EventExposure_Subscribe service operation, it can carry indication information in the service operation. The indication information is used to indicate the type of associated information contained in the UE data reported by the AMF. In one possible implementation, the associated information includes Global RANNode ID, RAN UE NGAP ID and Timestamp. In another possible implementation, the associated information includes Global RAN Node ID, AMF UE NGAP ID, and Timestamp.

[0158] The data lake can obtain the UE data on the core network based on the pairwise association method described above. Ultimately, the association information corresponding to the UE data in the core network includes the Global RAN Node ID, RAN UE NGAP ID, and Timestamp.

[0159] In step 802, the NWDAF sends a request message to the data lake, and the data lake receives the request message accordingly.

[0160] The request message carries the Global RAN node ID and a specific time period, and the request message is used to request the data lake to feedback the data of the UE on the core network related to the RAN device in a specific time period. The Global RAN node ID is the identification information of the RAN device. Optionally, the request message also includes first indication information, and the first indication information includes an event identifier (Event ID) and a network function type (NF type). The event identifier = Correlated CN Dataset, and the network function type includes AMF, UPF, PCF, etc. Optionally, the request message also includes second indication information, and the second indication information is used to instruct the data lake to feedback the associated UE data on the core network related to the RAN device in a specific time period, and the method for the data lake to associate the data of the UE on each network element in the core network domain can refer to the pairwise association method introduced above.

[0161] As an implementation method, NWDAF determines, based on local operator policies, that a service experience model needs to be trained for RAN devices. NWDAF then sends the aforementioned request message to the data lake. For example, a service provider (i.e., AF) reports to the operator that the average service experience for a service within a specific area and during a specific time period is poor. NWDAF, acting as the operator's data analysis network element, decides to train a service experience model based on a vertical federated learning approach, combining RAN devices and the core network, or combining the core network and AF, or combining the core network, RAN devices, and AF.

[0162] The NWDAF may obtain the RAN device identifier by, for example, querying the Operation, Administration, and Management (OAM) device for the identifiers of the RAN devices deployed in the specific area. Since the NWDAF obtains the service experience difference information from the service provider, the NWDAF may obtain the AF address information.

[0163] In step 803, the data lake sends a response message to the NWDAF, and the NWDAF receives the response message accordingly.

[0164] The response message carries data of the UE in the core network related to the RAN device and association information corresponding to the data of the UE, where the association information includes the Global RAN Node ID, RAN UE NGAP ID and Timestamp.

[0165] The UE data obtained by the NWDAF from the data lake is called the first data set, and the associated information corresponding to the UE data obtained by the NWDAF from the data lake is called the first associated information. In one possible implementation, the first data set may include data of multiple UEs, and each UE's data corresponds to a piece of associated information (also called the first associated information).

[0166] Step 804: The NWDAF performs data alignment with the RAN device to obtain a second data set.

[0167] As an implementation method, the NWDAF can send the acquired first association information to the RAN device. The RAN device determines the second association information based on the first association information and the third association information, where the second association information is the intersection of the first association information and the third association information. The RAN device then sends the second association information to the NWDAF. Consequently, the RAN device can determine a candidate data set within the access network based on the second association information and the third data set. The NWDAF can determine a candidate data set within the core network (also referred to as a second data set) based on the second association information and the first data set. This second data set corresponds to the second association information.

[0168] As another implementation method, the RAN device can send the obtained third association information to the NWDAF. The NWDAF determines the second association information based on the first association information and the third association information, where the second association information is the intersection of the first association information and the third association information. The second association information is then sent to the RAN device. Thus, the RAN device can determine the candidate data set within the access network based on the second association information, and the NWDAF can determine the candidate data set within the core network (also referred to as the second data set) based on the second association information and the first data set.

[0169] For example, the candidate dataset on the RAN side includes 10 million UE data items as shown below:

[0170] Data a1 of UE1: corresponds to association information 1;

[0171] UE2's data a2: corresponds to association information 2;

[0172] Data a3 of UE3: corresponds to association information 3;

[0173] Data a4 of UE4: corresponds to associated information 4;

[0174] …

[0175] The candidate dataset (i.e., the second dataset) on the NWDAF side includes 10 million UE data items as shown below:

[0176] UE1's data b1: corresponds to association information 1;

[0177] UE2's data b2: corresponds to association information 2;

[0178] Data b3 of UE3: corresponds to association information 3;

[0179] Data b4 of UE4: corresponds to association information 4;

[0180] …

[0181] In step 805a, the NWDAF sends a request message to the RAN device, and the RAN device receives the request message accordingly.

[0182] This request message is used to request information about the RAN device's ability to participate in vertical federated training within a specific time period. For example, the request message carries information about the specific time period. Optionally, the request message carries indication information indicating that information about the RAN device's ability to participate in vertical federated training within the specific time period should be obtained.

[0183] In step 805b, the RAN device sends the capability information of the RAN device to participate in the vertical federation training in the specific time period to the NWDAF. Accordingly, the NWDAF receives the capability information of the RAN device to participate in the vertical federation training in the specific time period.

[0184] Step 806: The NWDAF determines a fourth data set according to the capability information of the RAN, the capability information of the NWDAF, and the size of the candidate data set (ie, the second data set) on the NWDAF side.

[0185] The NWDAF determines whether the entire second data set can participate in training based on the RAN capability information, the NWDAF capability information, and the size of the second data set. If not, part of the UE data is selected from the second data set to form a fourth data set.

[0186] For example, assuming that it is determined in step 804 that the second data set that can participate in vertical federated learning contains data on 10 million UEs, and every 10% of the capacity on the RAN device can accommodate 1 million UE data for training, and assuming that the NWDAF determines that the remaining 80% of the capacity on the RAN side can be used for training, the NWDAF can randomly select 8 million UE data from the 10 million UE data to participate in federated training, that is, the selected fourth data set contains the 8 million UE data.

[0187] In step 807 , NWDAF divides the fourth data set into a training data set, a validation data set, and a test data set.

[0188] For example, in the above example, for the selected 8 million UE data, the NWDAF can randomly split them into 7 million, 500,000, and 500,000 data, respectively, as a training dataset, a validation dataset, and a test dataset. The training dataset includes one or more data, corresponding to the fourth association information. The validation dataset includes one or more data, corresponding to the fifth association information. The test dataset includes one or more data, corresponding to the sixth association information.

[0189] Step 808: NWDAF sends configuration information to RAN, and RAN receives the configuration information accordingly.

[0190] This configuration information indicates the association information corresponding to the training dataset, validation dataset, and test dataset used for vertical federated learning on the RAN side, namely, the fourth, fifth, and sixth association information. Based on this configuration information, the RAN can select data from some UEs from the candidate dataset on the RAN side to form the training dataset on the RAN side, select data from some UEs to form the validation dataset on the RAN side, and select data from some UEs to form the test dataset on the RAN side.

[0191] Optionally, the configuration information also includes the algorithm type used for vertical federated learning and the parameter information set within each algorithm (such as the number of layers of the neural network algorithm, the activation function used in each layer, the loss function type, etc.).

[0192] Since both the NWDFA and the RAN have defined the training, validation, and test datasets for vertical federated learning, they can subsequently train their models based on the vertical federated learning approach to generate service experience models. For example, referring to the method described above, consider the RAN as Client A and the NWDAF as Client B. Each model is trained based on its own training, validation, and test datasets to generate service experience models.

[0193] Based on the above solution, NWDAF obtains the service data of terminal devices on the core network elements and aligns the data with the RAN equipment. Then, NWDAF performs model training based on the aligned data to obtain an accurate service experience model. NWDAF does not need to obtain the service data of terminal devices on the RAN equipment.

[0194] It should be noted that NWDAF can also send the associated information corresponding to the training data set, verification data set and test data set to the RAN device in different stages (such as training stage, verification stage and test stage) to instruct the RAN device to perform model training or model verification or model testing.

[0195] It should be noted that the NWDAF described above can be a training NWDAF. After training the model using the vertical federation method, the training NWDAF obtains a service experience model, which is then passed to the inference NWDAF for generating data analysis results on the core network side. The inference NWDAF primarily determines data analysis results based on the trained service experience model and new data.

[0196] like Figure 9The figure shows a schematic diagram of a method for determining a service experience model provided by an embodiment of the present application. Based on this method, a data lake exists on the core network side. The NWDAF retrieves the data of the UE in the core network corresponding to the RAN device from the data lake based on the address information of the RAN device. This part of the UE data can be vertically federated with the data of the UE in the RAN device for federated learning.

[0197] The method comprises the following steps:

[0198] Steps 901a to 901b are the same as steps 801a to 801b above, and reference may be made to the above description.

[0199] In step 902, the NWDAF sends a request message to the data lake, and the data lake receives the request message accordingly.

[0200] The request message carries the Global RAN node ID and a specific time period, and the request message is used to request the data lake to feedback the association information corresponding to the data of the UE in the core network related to the RAN device in the specific time period. The GlobalRAN node ID is the identification information of the RAN device. Optionally, the request message also includes first indication information, and the first indication information includes an event identifier (Event ID) and a network function type (NF type). The event identifier = Sample ID list, and the network function type includes AMF, UPF, PCF, etc. Optionally, the request message also includes second indication information, and the second indication information is used to instruct the data lake to feedback the association information corresponding to the data of the associated UE in the core network related to the RAN device in the specific time period, wherein the method for the data lake to associate the data of the UE on each network element in the core network domain can refer to the pairwise association method introduced above.

[0201] As an implementation method, NWDAF determines, based on local operator policies, that a service experience model needs to be trained for RAN devices. NWDAF then sends the aforementioned request message to the data lake. For example, a service provider (i.e., AF) reports to the operator that the average service experience for a service within a specific area and during a specific time period is poor. NWDAF, acting as the operator's data analysis network element, decides to train a service experience model based on a vertical federated learning approach, combining RAN devices and the core network, or combining the core network and AF, or combining the core network, RAN devices, and AF.

[0202] The NWDAF may obtain the identifier of the RAN device by, for example, querying the OAM device for the identifiers of the RAN devices deployed in the specific area. Since the NWDAF obtains the service experience difference information from the service provider, the NWDAF may obtain the address information of the AF.

[0203] In step 903, the data lake sends a response message to the NWDAF, and the NWDAF receives the response message accordingly.

[0204] The response message carries association information corresponding to the data of UEs in the core network associated with the RAN device, including the Global RAN Node ID, RAN UE NGAP ID, and Timestamp. Optionally, the response message also carries the total size of the data of UEs in the core network associated with the RAN device, or the size of the data corresponding to each UE in the data of the core network associated with the RAN device.

[0205] The associated information obtained by NWDAF from the data lake is called the first associated information.

[0206] Step 904: The NWDAF performs data alignment with the RAN device to obtain second association information.

[0207] As an implementation method, the NWDAF can send the acquired first association information to the RAN device. The RAN device then determines the second association information based on the first association information and the third association information. The second association information is the intersection of the first and third association information. The RAN then sends the second association information to the NWDAF. The RAN device can then determine the candidate data sets within the access network based on the second association information and the third data set.

[0208] As another implementation method, the RAN can send the obtained third association information to the NWDAF. The NWDAF determines the second association information based on the first association information and the third association information. The second association information is the intersection of the first association information and the third association information. The second association information is then sent to the RAN device. The RAN device can then determine the candidate data set within the access network based on the second association information and the third data set.

[0209] In step 905, the NWDAF sends a request message to the data lake, and the data lake receives the request message accordingly.

[0210] The request message carries the second association information, and the request message is used to request to obtain the data of the UE corresponding to the second association information.

[0211] In step 906, the data lake sends a response message to the NWDAF, and accordingly, the NWDAF receives the response message.

[0212] The response message carries data of UEs in the core network associated with the RAN device corresponding to the second association information, where the second association information includes a Global RAN Node ID, a RAN UE NGAP ID, and a Timestamp. The set of data of UEs in the core network associated with the RAN device corresponding to the second association information is referred to as a second data set.

[0213] Therefore, through the above steps 902 to 903, NWDAF can obtain the associated information corresponding to the UE data in the core network (i.e., the first associated information) from the data lake, and through the above steps 905 to 906, NWDAF can obtain the UE data corresponding to the aligned second associated information (i.e., the second data set).

[0214] Steps 907a to 910 are the same as steps 805a to 808, and reference may be made to the aforementioned description.

[0215] Since both NWDFA and RAN have determined the training dataset, validation dataset, and test dataset for vertical federated learning, they can subsequently perform model training based on the vertical federated learning method to obtain service experience models. For example, you can refer to the method described above, using RAN as Client A and NWDAF as Client B, and perform model training based on their respective training datasets, validation datasets, and test datasets. Then, Client A and Client B can exchange the intermediate results. d i The two clients jointly obtain the service experience model. Among them, the parameter Θ corresponding to the service experience model is A Retained on Client A, parameter Θ B The two are retained on Client B and used for local reasoning.

[0216] Based on this solution, the NWDAF obtains terminal device service data from core network elements and aligns it with RAN equipment. It then uses this aligned data to train a model to generate an accurate service experience model, eliminating the need for the NWDAF to obtain terminal device service data from RAN equipment. Furthermore, in this solution, the NWDAF does not obtain data for all UEs in the core network from the data lake, but rather obtains data for UEs corresponding to the aligned second association information. This reduces transmission pressure between the NWDAF and the data lake.

[0217] It should be noted that the NWDAF described above can be a training NWDAF. After training the model using the vertical federation method, the training NWDAF obtains a service experience model, which is then passed to the inference NWDAF for generating data analysis results on the core network side. The inference NWDAF primarily determines data analysis results based on the trained service experience model and new data.

[0218] like Figure 10 As shown, it is a schematic diagram of a method for determining a service experience model provided in an embodiment of the present application. In this solution, there is no data lake involved. The data of the UE on the core network side is distributed in different NWDAFs (with data collection function). After each NWDAF collects the data of the UE on the core network, it can register the information of the RAN device corresponding to the data of the UE on the core network on the NRF. The NWDAF that assists in supporting federated learning training addresses these NWDAFs through the NRF, and then requests the data of the UE on the core network related to a certain RAN device from these NWDAFs.

[0219] The method comprises the following steps:

[0220] Step 1001 is the same as step 801a above, and reference may be made to the above description.

[0221] Step 1002: Different NWDAFs on the core network side collect and store data of UEs related to the same RAN device in the core network.

[0222] Among them, different NWDAFs can collect data on UEs related to the same RAN device on the same or different core network elements. For example, NWDAF1 to NWDAF3 collect data on UEs related to RAN devices on AMF, NWDAF4 to NWDAF7 collect data on UEs related to RAN devices on SMF, NWDAF8 to NWDAF10 collect data on UEs related to RAN devices on UPF, and so on.

[0223] Step 1003: Different NWDAFs on the core network side send registration requests to the NRF. Accordingly, the NRF receives the registration requests.

[0224] The registration request is used to request that identification information of an NWDAF be registered with an NRF, where the NWDAF stores data of the UE corresponding to the association information.

[0225] As an implementation method, the registration request carries the Global RAN Node ID, Timestamp, and NWDAFID.

[0226] As another implementation method, the registration request carries the Global RAN Node ID, RAN UE NGAP ID, Timestamp, and NWDAF ID.

[0227] Step 1004: The training NWDAF sends a request message to the NRF. Accordingly, the NRF receives the request message.

[0228] The request message carries the Global RAN node ID and a specific time period. The request message is used to request the NWDAF ID corresponding to the Global RAN node ID and the specific time period. That is, the NWDAF indicated by the NWDAF ID stores the UE data related to the RAN device corresponding to the Global RAN node ID within the specific time period.

[0229] Step 1005: The NRF sends a response message to the training NWDAF. Accordingly, the training NWDAF receives the response message.

[0230] The response message carries one or more NWDAF IDs.

[0231] Step 1006: The training NWDAF sends a request message to different NWDAFs. Correspondingly, the different NWDAFs receive the request message.

[0232] The request message carries the Global RAN node ID and a specific time period, and is used to request feedback of UE data related to the RAN device for the specific time period. Optionally, the request message carries indication information, and the indication information is used to instruct feedback of UE data related to the RAN device for the specific time period.

[0233] Step 1007 : The different NWDAFs send a response message to the training NWDAF, and accordingly, the training NWDAF receives the response message.

[0234] The response message carries UE data related to the RAN device and associated information corresponding to each UE's data. The associated information may include, for example, the Global RAN Node ID, RAN UE NGAP ID, and Timestamp, or the Global RAN Node ID, AN Tunnel Info, and Timestamp, or the Global RAN Node ID, SUPI, and Timestamp.

[0235] Step 1008: Train the NWDAF to process data from UEs of different NWDAFs.

[0236] For example, the NWDAF is trained to unify the associated information corresponding to UE data from different NWDAFs into GlobalRAN Node ID, RAN UE NGAP ID and Timestamp, and to perform deduplication processing on the UE data.

[0237] The set of UE data obtained after the training NWDAF processing is called a first data set, and the associated information corresponding to the first data set is called first associated information.

[0238] Steps 1009 to 1013 are the same as steps 804 to 808 above, and reference may be made to the aforementioned description.

[0239] Based on this solution, the NWDAF obtains terminal device service data from core network elements and aligns it with the RAN equipment. It then trains the model based on this aligned data to generate an accurate service experience model, eliminating the need for the NWDAF to obtain the terminal device service data from the RAN equipment. Furthermore, this solution is designed for scenarios where a data lake does not exist. UE data associated with a particular RAN device may exist in different NWDAFs. This allows the federated learning-supported training NWDAF to request UE data from these NWDAFs based on the RAN device's identity.

[0240] refer to Figure 11 , is a schematic diagram of a communication device provided in an embodiment of the present application. The communication device is used to implement each step corresponding to the data analysis network element in each of the above embodiments, such as Figure 11 As shown, the communication device 1100 includes a correlation information acquisition unit 1110, a data set acquisition unit 1120, a data set determination unit 1130, and a service experience model determination unit 1140. Optionally, it further includes a transceiver unit 1150.

[0241] An associated information acquisition unit 1110 is configured to acquire first associated information corresponding to a first data set, where the first data set includes data of the service on a core network element; a data set acquisition unit 1120 is configured to acquire a second data set based on the first associated information, where the second data set includes data corresponding to the second associated information in the first data set, where the second associated information is the intersection of the first associated information and the third associated information, where the third associated information corresponds to a third data set, and where the third data set includes data of the service on the first device; a data set determination unit 1130 is configured to determine a fourth data set based on the first information and the second data set, where the fourth data set is a subset or all of the second data set, where the first information includes capability information of the first device and / or capability information of the data analysis network element; a service experience model determination unit 1140 is configured to determine a service experience model of the service based on the fourth data set.

[0242] In a possible implementation method, the fourth data set includes a training set, the training set corresponds to fourth association information, and the business experience model determination unit 1140 is specifically used to: obtain the training set based on the fourth association information and the fourth data set; determine at least one candidate business experience model based on the training set; and determine the business experience model of the business from the at least one candidate business experience model.

[0243] In a possible implementation method, the fourth data set also includes a verification set, and the verification set corresponds to the fifth association information. The business experience model determination unit 1140 is specifically used to: obtain the verification set based on the fifth association information and the fourth data set; determine the verification results corresponding to the at least one candidate business experience model based on the verification set; and determine the business experience model of the business based on the verification results corresponding to the at least one candidate business experience model.

[0244] In a possible implementation method, the fourth data set also includes a test set, which corresponds to the sixth association information. The business experience model determination unit 1140 is further used to: obtain the test set based on the sixth association information and the fourth data set; and determine the test result of the business experience model of the business based on the test set.

[0245] In a possible implementation method, the transceiver unit 1150 is configured to send the fourth association information corresponding to the training set to the first device.

[0246] In a possible implementation method, the transceiver unit 1150 is configured to: send a first request to a second device, where the first request carries identification information of the first device and is used to request the first data set; and receive the first data set from the second device.

[0247] In a possible implementation method, the data set acquisition unit 1120 is specifically used to: send the first association information to the first device through the transceiver unit 1150; receive the second association information from the first device through the transceiver unit 1150; and acquire the second data set based on the second association information and the first data set.

[0248] In a possible implementation method, the data set acquisition unit 1120 is specifically used to: receive the third association information from the first device through the transceiver unit 1150; determine the second association information based on the third association information and the first association information; and acquire the second data set based on the second association information and the first data set.

[0249] In a possible implementation method, the association information acquisition unit 1110 is specifically used to: send a second request to the second device through the transceiver unit 1150, where the second request carries the identification information of the first device, and the second request is used to request the first association information corresponding to the first data set; and receive the first association information from the second device through the transceiver unit 1150.

[0250] In one possible implementation method, the data set acquisition unit 1120 is specifically used to: determine the second association information based on the first association information; send a third request to the second device through the transceiver unit 1150, where the third request carries the second association information, and the third request is used to request the second data set; and receive the second data set from the second device through the transceiver unit 1150.

[0251] In a possible implementation method, the transceiver unit 1150 is further used to: send a fourth request to the network element storage function network element, where the fourth request is used to request the address information of the second device; and receive the address information of the second device from the network element storage function network element.

[0252] In a possible implementation method, the second device is a data analysis network element supporting a data lake function, or a data analysis network element supporting a data collection and coordination function, or a data analysis network element supporting a data collection function.

[0253] In a possible implementation method, the first device is an access network device or a service device.

[0254] In a possible implementation method, the first association information includes the following information: identification information of the first device, identification information allocated by the first device to the terminal device, and a timestamp.

[0255] In a possible implementation method, the identification information allocated by the first device to the terminal device is the identification information allocated by the first device to the terminal device on a first interface, and the first interface is an interface between the first device and the core network element.

[0256] Optionally, the communication device may further include a storage unit for storing data or instructions (also referred to as code or program). Each of the above units may interact or couple with the storage unit to implement the corresponding method or function. For example, the processing unit 1120 may read the data or instructions in the storage unit so that the communication device implements the method in the above embodiment.

[0257] It should be understood that the division of units in the above communication device is merely a division of logical functions. In actual implementation, they can be fully or partially integrated into a physical entity, or they can be physically separated. Moreover, the units in the communication device can all be implemented in the form of software called through a processing element; or all be implemented in the form of hardware; or some units can be implemented in the form of software called through a processing element, and some units can be implemented in the form of hardware. For example, each unit can be a separately established processing element, or it can be integrated into a chip of the communication device. In addition, it can also be stored in the form of a program in a memory, called by a processing element of the communication device to perform the function of the unit. In addition, these units can be fully or partially integrated together, or they can be implemented independently. The processing element described here can also be a processor, which can be an integrated circuit with signal processing capabilities. In the implementation process, each step of the above method or each unit above can be implemented by an integrated logic circuit of hardware in the processor element or by software called through a processing element.

[0258] In one example, the unit in any of the above communication devices may be one or more integrated circuits configured to implement the above method, such as one or more application specific integrated circuits (ASICs), or one or more digital singnal processors (DSPs), or one or more field programmable gate arrays (FPGAs), or a combination of at least two of these integrated circuit forms. For another example, when the unit in the communication device can be implemented in the form of a processing element scheduler, the processing element may be a general-purpose processor, such as a central processing unit (CPU) or other processor that can call a program. For another example, these units can be integrated together and implemented in the form of a system-on-a-chip (SOC).

[0259] refer to Figure 12 , is a schematic diagram of a communication device provided in an embodiment of the present application, which is used to implement the operation of the data analysis network element in the above embodiment. Figure 12 As shown, the communication device includes: a processor 1210 and an interface 1230. Optionally, the communication device also includes a memory 1220. The interface 1230 is used to implement communication with other devices.

[0260] The method executed by the data analysis network element in the above embodiment can be implemented by the processor 1210 calling a program stored in a memory (which can be the memory 1220 in the data analysis network element or an external memory). That is, the data analysis network element may include a processor 1210, which executes the method executed by the data analysis network element in the above method embodiment by calling the program in the memory. The processor here can be an integrated circuit with signal processing capabilities, such as a CPU. The data analysis network element can be implemented by configuring one or more integrated circuits to implement the above method. For example: one or more ASICs, or one or more microprocessors DSPs, or one or more FPGAs, etc., or a combination of at least two of these integrated circuit forms. Alternatively, the above implementation methods can be combined.

[0261] Specifically, Figure 11 The functions / implementation processes of the associated information acquisition unit 1110, the data set acquisition unit 1120, the data set determination unit 1130, the service experience model determination unit 1140 and the transceiver unit 1150 can be realized by Figure 12 The processor 1210 in the communication device 1200 shown calls the computer executable instructions stored in the memory 1220 to implement. Or, Figure 11 The functions / implementation processes of the associated information acquisition unit 1110, the data set acquisition unit 1120, the data set determination unit 1130, and the service experience model determination unit 1140 can be realized by Figure 12 The processor 1210 in the communication device 1200 shown calls the computer execution instructions stored in the memory 1220 to implement, Figure 11 The function / implementation process of the transceiver unit 1150 can be achieved by Figure 12 12 is implemented by the interface 1230 in the communication device 1200 shown in FIG. 12 . Exemplarily, the function / implementation process of the transceiver unit 1150 can be implemented by the processor calling program instructions in the memory to drive the interface 1230 .

[0262] Those skilled in the art will understand that the various numerical numbers such as the first and second involved in this application are only for the convenience of description and are not used to limit the scope of the embodiments of the present application, and also represent the order of precedence. "And / or" describes the association relationship of associated objects, indicating that three relationships can exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the objects before and after the association are in an "or" relationship. "At least one" refers to one or more. At least two refers to two or more. "At least one", "any one" or similar expressions refer to any combination of these items, including any combination of single items (individuals) or plural items (individuals). For example, at least one item (individual, kind) of a, b, or c can represent: a, b, c, ab, ac, bc, or abc, where a, b, c can be single or multiple. "Multiple" refers to two or more, and other quantifiers are similar.

[0263] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0264] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0265] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When software is used for implementation, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can 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 can be transmitted from a 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 can 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 integrations. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)).

[0266] The various illustrative logic units and circuits described in the embodiments of the present application can be implemented or operated by a general-purpose processor, a digital signal processor, an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic device, a discrete gate or transistor logic, a discrete hardware component, or the design of any combination thereof. The general-purpose processor can be a microprocessor, alternatively, the general-purpose processor can also be any traditional processor, controller, microcontroller or state machine. The processor can also be implemented by a combination of computing devices, such as a digital signal processor and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a digital signal processor core, or any other similar configuration to implement.

[0267] The steps of the method or algorithm described in the embodiments of the present application can be directly embedded in the software unit executed by the hardware, processor, or a combination of the two. The software unit can be stored in a random access memory (RAM), flash memory, read-only memory (ROM), EPROM memory, EEPROM memory, register, hard disk, removable disk, CD-ROM or other storage medium in any form in the art. Exemplarily, the storage medium can be connected to the processor so that the processor can read information from the storage medium and can write information to the storage medium. Alternatively, the storage medium can also be integrated into the processor. The processor and storage medium can be arranged in an ASIC.

[0268] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0269] In one or more exemplary designs, the above-described functions described herein may be implemented in hardware, software, firmware, or any combination thereof. If implemented in software, these functions may be stored on a computer-readable medium or transmitted in the form of one or more instructions or codes on a computer-readable medium. Computer-readable media include computer storage media and communication media that facilitate the transfer of computer programs from one location to another. Storage media can be any available medium that can be accessed by a general-purpose or specialized computer. For example, such computer-readable media may include, but is not limited to, RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to carry or store program code in the form of instructions or data structures and other forms that can be read by a general-purpose or specialized computer, or a general-purpose or specialized processor. In addition, any connection can be appropriately defined as a computer-readable medium. For example, if the software is transmitted from a website, server, or other remote resource via a coaxial cable, fiber optic cable, twisted pair cable, digital subscriber line (DSL), or wirelessly, such as infrared, wireless, and microwave, it is also included in the definition of computer-readable media. Disks and discs include compact disks, laser discs, optical discs, digital versatile discs (DVDs), floppy disks, and Blu-ray discs. Disks typically reproduce data magnetically, while discs typically reproduce data optically using lasers. Combinations of these may also be included in computer-readable media.

[0270] Those skilled in the art will appreciate that, in one or more of the examples above, the functions described herein may be implemented using hardware, software, firmware, or any combination thereof. When implemented using software, these functions may be stored in a computer-readable medium or transmitted as one or more instructions or codes on a computer-readable medium. Computer-readable media include computer storage media and communication media, wherein communication media include any medium that facilitates the transmission of a computer program from one place to another. The storage medium may be any available medium that can be accessed by a general-purpose or special-purpose computer.

[0271] The specific implementation methods described above further explain the purpose, technical solutions and beneficial effects of the present application in detail. It should be understood that the above description is only the specific implementation method of the present application and is not intended to limit the scope of protection of the present application. Any modifications, equivalent replacements, improvements, etc. made on the basis of the technical solutions of the present application should be included in the scope of protection of the present application. The above description of the specification of this application can enable any technical personnel in the field to utilize or implement the contents of this application. Any modification based on the disclosed contents should be considered obvious in the field. The basic principles described in this application can be applied to other variations without departing from the inventive essence and scope of the present application. Therefore, the contents disclosed in this application are not limited to the described embodiments and designs, but can also be extended to the maximum scope consistent with the principles of this application and the disclosed new features.

[0272] Although the present application has been described with reference to specific features and embodiments thereof, it is apparent that various modifications and combinations may be made thereto without departing from the spirit and scope of the present application. Accordingly, this specification and the drawings are merely illustrative of the present application as defined by the appended claims and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, those skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, the present application is intended to include such modifications and variations if they fall within the scope of the claims of the present application and their equivalents.

Claims

1. A method for determining a service experience model, characterized in that: include: The data analysis network element obtains first associated information corresponding to a first data set, where the first data set includes data of a service on a core network network element; The data analysis network element obtains a second data set based on the first association information, where the second data set includes data corresponding to the second association information in the first data set, the second association information is an intersection of the first association information and third association information, the third association information corresponds to a third data set, and the third data set includes data of the service on the first device; The data analysis network element determines a fourth data set based on the first information and the second data set, where the fourth data set is a subset or the entirety of the second data set, and the first information includes capability information of the first device and / or capability information of the data analysis network element; The data analysis network element determines a service experience model of the service based on the fourth data set.

2. The method according to claim 1, characterized in that The fourth data set includes a training set, the training set corresponds to fourth association information, and the data analysis network element determines the service experience model of the service according to the fourth data set, including: The data analysis network element obtains the training set according to the fourth association information and the fourth data set; The data analysis network element determines at least one candidate service experience model based on the training set; The data analysis network element determines the service experience model of the service from the at least one candidate service experience model.

3. The method according to claim 2, characterized in that The fourth data set further includes a verification set, the verification set corresponds to the fifth association information, and the data analysis network element determines the service experience model of the service from the at least one candidate service experience model, including: The data analysis network element obtains the verification set according to the fifth association information and the fourth data set; The data analysis network element determines, based on the verification set, verification results corresponding to the at least one candidate service experience model; The data analysis network element determines the service experience model of the service according to the verification results corresponding to the at least one candidate service experience model.

4. The method according to claim 3, characterized in that The fourth data set further includes a test set, the test set corresponding to the sixth association information, and the method further includes: The data analysis network element obtains the test set according to the sixth association information and the fourth data set; The data analysis network element determines a test result of the service experience model of the service based on the test set.

5. The method according to any one of claims 2 to 4, characterized in that The method further comprises: The data analysis network element sends the fourth association information corresponding to the training set to the first device.

6. The method according to any one of claims 1 to 4, characterized in that The method further comprises: The data analysis network element sends a first request to the second device, where the first request carries identification information of the first device, and the first request is used to request the first data set; The data analysis network element receives the first data set from the second device.

7. The method according to any one of claims 1 to 4, characterized in that The data analysis network element obtains a second data set according to the first correlation information, including: The data analysis network element sends the first association information to the first device; The data analysis network element receives the second association information from the first device; The data analysis network element obtains the second data set according to the second association information and the first data set.

8. The method according to any one of claims 1 to 4, characterized in that The data analysis network element obtains a second data set according to the first correlation information, including: The data analysis network element receives the third association information from the first device; The data analysis network element determines the second association information based on the third association information and the first association information; The data analysis network element obtains the second data set according to the second association information and the first data set.

9. The method according to any one of claims 1 to 4, characterized in that The data analysis network element obtains first associated information corresponding to the first data set, including: The data analysis network element sends a second request to the second device, where the second request carries identification information of the first device, and the second request is used to request first association information corresponding to the first data set; The data analysis network element receives the first association information from the second device.

10. The method according to claim 9, characterized in that The data analysis network element obtains a second data set according to the first correlation information, including: The data analysis network element determines the second association information based on the first association information; The data analysis network element sends a third request to the second device, where the third request carries the second association information and is used to request the second data set; The data analysis network element receives the second data set from the second device.

11. The method according to claim 6, characterized in that The method further comprises: The data analysis network element sends a fourth request to the network element storage function network element, where the fourth request is used to request address information of the second device; The data analysis network element receives the address information of the second device from the network element storage function network element.

12. The method according to claim 9, characterized in that The method further comprises: The data analysis network element sends a fourth request to the network element storage function network element, where the fourth request is used to request address information of the second device; The data analysis network element receives the address information of the second device from the network element storage function network element.

13. The method according to claim 10, characterized in that The method further comprises: The data analysis network element sends a fourth request to the network element storage function network element, where the fourth request is used to request address information of the second device; The data analysis network element receives the address information of the second device from the network element storage function network element.

14. The method according to claim 6, characterized in that The second device is a data analysis network element supporting a data lake function, or a data analysis network element supporting a data collection and coordination function, or a data analysis network element supporting a data collection function.

15. The method according to claim 9, characterized in that The second device is a data analysis network element supporting a data lake function, or a data analysis network element supporting a data collection and coordination function, or a data analysis network element supporting a data collection function.

16. The method according to any one of claims 10 to 13, characterized in that The second device is a data analysis network element supporting a data lake function, or a data analysis network element supporting a data collection and coordination function, or a data analysis network element supporting a data collection function.

17. The method according to any one of claims 1 to 4 and 10 to 15, characterized in that The first device is an access network device or a service device.

18. The method according to any one of claims 1 to 4 and 10 to 15, characterized in that The first association information includes the following information: The identification information of the first device, the identification information allocated by the first device to the terminal device, and the timestamp.

19. The method according to claim 18, characterized in that The identification information allocated by the first device to the terminal device is the identification information allocated by the first device to the terminal device on the first interface, and the first interface is the interface between the first device and the core network element.

20. A communication device, characterized in that: include: A correlation information acquisition unit, configured to acquire first correlation information corresponding to a first data set, where the first data set includes data of a service on a core network element; a data set acquisition unit, configured to acquire a second data set based on the first association information, the second data set including data corresponding to the second association information in the first data set, the second association information being an intersection of the first association information and third association information, the third association information corresponding to a third data set, the third data set including data of the service on the first device; a data set determining unit, configured to determine a fourth data set based on the first information and the second data set, where the fourth data set is a subset or the entirety of the second data set, the first information including capability information of the first device and / or capability information of a data analysis network element; A service experience model determining unit is configured to determine a service experience model of the service based on the fourth data set.

21. The device according to claim 20, characterized in that The fourth data set includes a training set, the training set corresponds to the fourth association information, and the service experience model determination unit is specifically configured to: Acquire the training set according to the fourth association information and the fourth data set; Determining at least one candidate service experience model based on the training set; A service experience model for the service is determined from the at least one candidate service experience model.

22. The device according to claim 21, characterized in that The fourth data set further includes a verification set, the verification set corresponds to the fifth association information, and the service experience model determination unit is specifically configured to: Acquire the verification set according to the fifth association information and the fourth data set; Determining, based on the verification set, verification results corresponding to each of the at least one candidate service experience models; Determine the service experience model of the service according to the verification results corresponding to the at least one candidate service experience model.

23. The device according to claim 22, characterized in that The fourth data set further includes a test set, where the test set corresponds to the sixth association information. The service experience model determination unit is further configured to: Acquire the test set according to the sixth association information and the fourth data set; A test result of the service experience model of the service is determined based on the test set.

24. The device according to any one of claims 21 to 23, characterized in that The apparatus further includes a transceiver unit configured to send the fourth association information corresponding to the training set to the first device.

25. The device according to any one of claims 20 to 23, characterized in that The device further includes a transceiver unit, configured to: Sending a first request to a second device, where the first request carries identification information of the first device and is used to request the first data set; The first data set is received from the second device.

26. The device according to any one of claims 20 to 23, characterized in that The device further comprises a transceiver unit; The data set acquisition unit is specifically used to: sending the first association information to the first device through the transceiver unit; receiving, by the transceiver unit, the second association information from the first device; The second data set is acquired according to the second association information and the first data set.

27. The device according to any one of claims 20 to 23, characterized in that The device further comprises a transceiver unit; The data set acquisition unit is specifically used to: receiving the third association information from the first device through the transceiver unit; determining the second association information according to the third association information and the first association information; The second data set is acquired according to the second association information and the first data set.

28. The device according to any one of claims 20 to 23, characterized in that The device further comprises a transceiver unit; The associated information acquisition unit is specifically configured to: Sending a second request to the second device through the transceiver unit, where the second request carries identification information of the first device, and the second request is used to request first association information corresponding to the first data set; The first association information is received from the second device through the transceiver unit.

29. The device according to claim 28, characterized in that The data set acquisition unit is specifically used to: determining the second association information according to the first association information; sending, by the transceiver unit, a third request to the second device, where the third request carries the second association information and is used to request the second data set; The second data set is received from the second device through the transceiver unit.

30. The device according to claim 25, wherein The transceiver unit is further configured to: Sending a fourth request to the network element storage function network element, where the fourth request is used to request address information of the second device; Receive the address information of the second device from the network element storage function network element.

31. The device according to claim 28, characterized in that The transceiver unit is further configured to: Sending a fourth request to the network element storage function network element, where the fourth request is used to request address information of the second device; Receive the address information of the second device from the network element storage function network element.

32. The device according to claim 29, characterized in that The transceiver unit is further configured to: Sending a fourth request to the network element storage function network element, where the fourth request is used to request address information of the second device; Receive the address information of the second device from the network element storage function network element.

33. The device according to claim 25, characterized in that The second device is a data analysis network element supporting a data lake function, or a data analysis network element supporting a data collection and coordination function, or a data analysis network element supporting a data collection function.

34. The device according to claim 28, wherein The second device is a data analysis network element supporting a data lake function, or a data analysis network element supporting a data collection and coordination function, or a data analysis network element supporting a data collection function.

35. The device according to any one of claims 29 to 32, characterized in that The second device is a data analysis network element supporting a data lake function, or a data analysis network element supporting a data collection and coordination function, or a data analysis network element supporting a data collection function.

36. The device according to any one of claims 20 to 23 and 29 to 34, characterized in that The first device is an access network device or a service device.

37. The device according to any one of claims 20 to 23 and 29 to 34, characterized in that The first association information includes the following information: The identification information of the first device, the identification information allocated by the first device to the terminal device, and the timestamp.

38. The device according to claim 37, characterized in that The identification information allocated by the first device to the terminal device is the identification information allocated by the first device to the terminal device on the first interface, and the first interface is the interface between the first device and the core network element.

39. A communication device, characterized in that: include: A processor, wherein the processor is coupled to a memory, wherein the memory is used to store a program or an instruction, and when the program or the instruction is executed by the processor, the apparatus executes the method according to any one of claims 1 to 19.

40. A chip system, characterized in that: include: The chip system includes at least one processor and an interface circuit, wherein the interface circuit is coupled to the at least one processor, and the processor executes the method according to any one of claims 1 to 19 by running instructions.

41. A computer-readable storage medium, characterized in that The method comprises instructions which, when executed on a computer, cause the computer to perform the method according to any one of claims 1 to 19.

42. A communication system, characterized in that include: A data analysis network element, configured to execute the method according to any one of claims 1 to 19; and A core network element used to communicate with the data analysis network element.

43. A computer program product, characterized in that When the computer program product is run, the method according to any one of claims 1 to 19 is implemented.

Images