Network element interaction method and device, medium and electronic equipment

Abstract

The invention belongs to the technical field of communication and artificial intelligence, and particularly relates to a network element interaction method, a network element interaction device, a computer readable medium, electronic equipment and a computer program product. The method comprises the following steps: selecting a publisher network element which establishes an information subscription relationship with a subscriber network element in a mobile communication network; performing data alignment between the subscriber network element and the publisher network element to obtain identification information of intersection data; selecting publisher sample data on the publisher network element according to the identification information of the intersection data, and training a machine learning model corresponding to the subscription content according to the publisher sample data to obtain model intermediate information of the publisher network element; and sending the model intermediate information of the publisher network element to the subscriber network element, and training a machine learning model corresponding to the subscription content on the subscriber network element according to the model intermediate information of the publisher network element. According to the invention, the utilization rate of system data in the mobile communication network can be improved, and the system performance of the mobile communication network is optimized.

Description

Technical Field

[0001] The present application belongs to the field of communications and artificial intelligence technology, and specifically relates to a network element interaction method, a network element interaction device, a computer-readable medium, an electronic device, and a computer program product. Background Art

[0002] In mobile communication networks, due to data privacy and security concerns, various functional entities typically do not disclose their own system data. Alternatively, they require complex encryption and verification processes before sharing the data with designated functional entities. This difficulty in sharing system data between different functional entities results in a significant amount of data being underutilized, hindering the optimization of mobile communication network performance. Summary of the Invention

[0003] The present application provides a network element interaction method, a network element interaction device, a computer-readable medium, an electronic device, and a computer program product, the purpose of which is to improve the utilization rate of system data in a mobile communication network and thereby optimize the system performance of the mobile communication network.

[0004] According to one aspect of an embodiment of the present application, a network element interaction method is provided, the method including:

[0005] Selecting, in a mobile communication network, a publisher network element that establishes an information subscription relationship with a subscriber network element, wherein the information subscription relationship indicates that the publisher network element provides data analysis information related to original data of the subscription content to the subscriber network element;

[0006] Performing data alignment between the subscriber network element and the publisher network element to obtain identification information of intersection data, where the intersection data is data having the same identification information between original data held by the subscriber and original data held by the publisher network element;

[0007] Selecting publisher sample data on the publisher network element according to the identification information of the intersection data, and training a machine learning model corresponding to the subscription content according to the publisher sample data to obtain model intermediate information of the publisher network element;

[0008] The model intermediate information of the publisher network element is sent to the subscriber network element, and a machine learning model corresponding to the subscription content is trained on the subscriber network element based on the model intermediate information of the publisher network element.

[0009] According to one aspect of an embodiment of the present application, a network element interaction device is provided, the device including:

[0010] a network element selection module configured to select, in a mobile communication network, a publisher network element that establishes an information subscription relationship with a subscriber network element, wherein the information subscription relationship is used to indicate that the publisher network element provides data analysis information related to original data of the subscription content to the subscriber network element;

[0011] a data alignment module configured to perform data alignment between the subscriber network element and the publisher network element to obtain identification information of intersection data, wherein the intersection data is data having the same identification information between the original data held by the subscriber and the original data held by the publisher network element;

[0012] a sample selection module configured to select publisher sample data on the publisher network element according to the identification information of the intersection data, and train a machine learning model corresponding to the subscription content according to the publisher sample data to obtain intermediate model information of the publisher network element;

[0013] The model training module is configured to send the model intermediate information of the publisher network element to the subscriber network element, and train a machine learning model corresponding to the subscription content on the subscriber network element based on the model intermediate information of the publisher network element.

[0014] In some embodiments of the present application, based on the above technical solution, the data alignment module includes:

[0015] a third-party network element selection module configured to select a third-party network element that has established a trusted computing relationship with both the subscriber network element and the publisher network element, wherein the trusted computing relationship indicates that the third-party network element can perform trusted computing on the original data held by each of the subscriber network element and the publisher network element while maintaining data privacy;

[0016] an identification information sending module, configured to send identification information of the original data respectively held by the subscriber network element and the publisher network element to the third-party network element;

[0017] an identification information alignment module configured to align identification information of the original data respectively held by the subscriber network element and the publisher network element on the third-party network element to obtain identification information of intersection data;

[0018] The intersection data sending module is configured to send the identification information of the intersection data to the subscriber network element and the publisher network element respectively by the third-party network element.

[0019] In some embodiments of the present application, based on the above technical solution, the third-party network element selection module includes:

[0020] A network element set acquisition module is configured to acquire a first network element set that establishes a trusted computing relationship with the subscriber network element and a second network element set that establishes a trusted computing relationship with the publisher network element;

[0021] A first selection module is configured to select a common network element as a third-party network element if there is one between the first network element set and the second network element set;

[0022] The second selection module is configured to select a temporary network element as the third-party network element through a network open function network element in the mobile communication network if there is no common network element between the first network element set and the second network element set.

[0023] In some embodiments of the present application, based on the above technical solution, the temporary network element is a network element selected from a third network element set maintained by the network open function network element, which is different from the first network element set and the second network element set. The temporary network element establishes a trusted computing relationship with a specified validity period with the subscriber network element and the publisher network element respectively through the network open function network element.

[0024] In some embodiments of the present application, based on the above technical solution, the network element selection module includes:

[0025] an information opening request module, configured to send an information opening request of the subscribing network element to a network opening function network element in the mobile communication network, wherein the information opening request carries a network element identifier and subscription content of the subscribing network element;

[0026] The information subscription request module is configured to establish a mapping relationship between the network element identifier of the subscriber network element and the subscription content on the network open function network element, and initiate an information subscription request to establish an information subscription relationship to the publisher network element according to the subscription content.

[0027] In some embodiments of the present application, based on the above technical solution, the network element selection module further includes:

[0028] A network element authority verification module is configured to perform authority verification on the subscriber network element according to the network element identifier of the subscriber network element, wherein the authority verification is used to determine whether the subscriber network element has system access authority;

[0029] The execution rejection module is configured to reject the execution of the information opening request of the subscriber network element if the verification result of the authority verification is verification failure.

[0030] In some embodiments of the present application, based on the above technical solution, the model training module includes:

[0031] a sample data selection module configured to select subscriber sample data on the subscriber network element according to the identification information of the intersection data, and train a machine learning model corresponding to the subscription content according to the subscriber sample data to obtain model intermediate information of the subscriber network element;

[0032] The intermediate information training module is configured to train a machine learning model corresponding to the subscription content on the subscriber network element based on the model intermediate information of the subscriber network element and the model intermediate information of the publisher network element.

[0033] In some embodiments of the present application, based on the above technical solution, the intermediate information training module is further configured to: select a third-party network element that establishes a trusted computing relationship with both the subscriber network element and the publisher network element; send the model intermediate information of the subscriber network element and the model intermediate information of the publisher network element to the third-party network element respectively; aggregate the model intermediate information of the subscriber network element and the model intermediate information of the publisher network element on the third-party network element to obtain global information; and iteratively update the model parameters of the machine learning model corresponding to the subscription content on the subscriber network element and the publisher network element according to the global information.

[0034] In some embodiments of the present application, based on the above technical solution, the subscribing network element is a network element outside the core network of the mobile communication network, and the publishing network element is a network element inside the core network of the mobile communication network.

[0035] In some embodiments of the present application, based on the above technical solution, the subscriber network element is an application function network element for implementing application layer services, and the publisher network element is a network data analysis function network element for providing network data analysis services to the mobile communication network.

[0036] In some embodiments of the present application, based on the above technical solution, the network element interaction device further includes:

[0037] The sample collection module is configured to initiate a sample collection request to the operation, maintenance and management network element in the mobile communication network through the network data analysis function network element; and obtain the original data held by the network data analysis function network element according to the sample collection request.

[0038] In some embodiments of the present application, based on the above technical solution, the original data held by the network data analysis function network element includes channel quality indication information for indicating the quality of the wireless channel.

[0039] In some embodiments of the present application, based on the above technical solution, the subscribing network element and the publishing network element train a machine learning model corresponding to the subscribed content through federated learning.

[0040] In some embodiments of the present application, based on the above technical solution, the subscribing network element and the publishing network element train a machine learning model corresponding to the subscribed content through vertical federated learning; the training samples held by the subscribing network element and the training samples held by the publishing network element have the same sample identifier and different characteristics.

[0041] According to one aspect of an embodiment of the present application, a computer-readable medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the network element interaction method in the above technical solution is implemented.

[0042] According to one aspect of an embodiment of the present application, an electronic device is provided, comprising: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to execute the executable instructions to implement the network element interaction method in the above technical solution.

[0043] According to one aspect of an embodiment of the present application, a computer program product is provided, including a computer program, which implements the network element interaction method in the above technical solution when executed by a processor.

[0044] In the technical solution provided in the embodiment of the present application, by establishing an information subscription relationship between different network elements, data analysis information can be shared between different network elements while keeping the original data held by the network element itself from being leaked, so that the shared data analysis information can be used to implement the training of the machine learning model. On the one hand, it can ensure the privacy and security of the data, and on the other hand, it can improve the wide applicability of the machine learning model in mobile communication networks. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] The accompanying drawings are incorporated into and constitute a part of the specification, illustrate embodiments consistent with the present application, and together with the specification, are used to explain the principles of the present application. Obviously, the drawings described below are only some embodiments of the present application, and those skilled in the art can derive other drawings based on these drawings without inventive effort.

[0046] Figure 1 The network architecture diagram of the mobile communication network applying the technical solution of the present application is schematically shown.

[0047] Figure 2 The network architecture applicable to the embodiments of the present application is schematically shown.

[0048] Figure 3 A flow chart of a network element interaction method in one embodiment of the present application is shown.

[0049] Figure 4 A flowchart of data alignment based on a trusted third party in one embodiment of the present application is shown.

[0050] Figure 5 A flowchart of training a machine learning model based on a third-party network element in one embodiment of the present application is shown.

[0051] Figure 6 The following is an architectural diagram of an embodiment of the present application for implementing network element interaction in an application scenario.

[0052] Figure 7 A flowchart of direct interaction between two network elements in an application scenario in an embodiment of the present application is shown.

[0053] Figure 8 A flowchart of interaction through a third-party network element in an application scenario in an embodiment of the present application is shown.

[0054] Figure 9 The structural block diagram of the network element interaction device provided in an embodiment of the present application is schematically shown.

[0055] Figure 10 The following schematically shows a block diagram of a computer system structure of an electronic device suitable for implementing an embodiment of the present application. DETAILED DESCRIPTION

[0056] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this application will be thorough and complete and will fully convey the concepts of the example embodiments to those skilled in the art.

[0057] In addition, described feature, structure or characteristic can be combined in one or more embodiments in any suitable manner.In the following description, many specific details are provided so as to provide a full understanding of the embodiments of the present application. However, it will be appreciated by those skilled in the art that the technical scheme of the present application can be put into practice without one or more of the specific details, or other methods, components, devices, steps etc. can be adopted. In other cases, known methods, devices, implementations or operations are not shown or described in detail to avoid blurring the various aspects of the application.

[0058] In the embodiments of the present application, the term "module" or "unit" refers to a computer program or a part of a computer program that has a predetermined function and works together with other related parts to achieve a predetermined goal, and can be implemented in whole or in part by using software, hardware (such as processing circuits or memories) or a combination thereof. Similarly, a processor (or multiple processors or memories) can be used to implement one or more modules or units. In addition, each module or unit can be part of an overall module or unit that includes the function of the module or unit.

[0059] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically separate entities. That is, these functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0060] The flowcharts shown in the accompanying drawings are for illustrative purposes only and do not necessarily include all contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may be decomposed, while others may be combined or partially combined. Therefore, the actual execution order may vary depending on the actual situation.

[0061] The relevant terms of the mobile communication network involved in the embodiments of the present application are explained as follows.

[0062] AF: Application function, also known as an AS (application server). This entity implements the control plane functions of third-party application servers, interacting through AF-NEF-PCF or AF-PCF. This entity also implements the user plane functions of third-party application servers, namely the AS-IP transport network-UPF interface.

[0063] SMF: System Management Function, responsible for tunnel maintenance, IP address allocation and management, UPF selection, policy implementation and QoS control, billing data collection, roaming, etc.

[0064] AMF: Access and mobility management function, performs registration, connection, reachability, and mobility management. It provides a session management message transmission channel for the UE and SMF, provides authentication and authorization functions for user access, and serves as the access point for the terminal and the wireless core network control plane.

[0065] PCF: Policy Control Function. A unified policy framework that provides policy rules for control plane functions.

[0066] UPF: User plane function. Packet routing and forwarding, policy enforcement, traffic reporting, and QoS processing.

[0067] NEF (Network Exposure Function) sits between the 5G core network and external third-party application functions (possibly also some internal AFs). It manages publicly available network data. All external applications that wish to access internal data within the 5G core network must go through the NEF. The NEF provides security guarantees for external applications accessing the 3GPP network, including the ability to expose QoS customization capabilities for external applications, subscription to mobility status events, and AF request distribution.

[0068] UE: user equipment; terminal equipment can include one or more of a mobile phone app, a vehicle-mounted IP camera, and other terminal software.

[0069] NWDAF: Network Data Analytics Function. Provides specific network data analysis services to the network.

[0070] ADRF: Analytics Data Repository Function. This functional entity is also introduced to implement network analysis functions. Its main function is to store network analysis data and collected data.

[0071] MFAF: Messaging Framework Adaptor Function. This NF's primary function is to adapt to the message framework, a function not defined in 3GPP. The MFAF enables interaction between the DCCF and the message framework, sending data to the message framework, receiving processed data from the message framework, and performing other processing such as formatting. This NF is also introduced to implement network data analysis.

[0072] AnLF: Analytics Logical Function. Responsible for model reasoning, providing general NWDAF service interfaces such as Nnwdaf_AnalyticsSubscription and Nnwdaf_AnalyticsInfo, and generating analysis results (including static statistical data and dynamic reasoning results) based on requests from consumer network elements.

[0073] MTLF (Model Training Logical Function) is responsible for model training and can provide trained models to AnLF. The model training process is not standardized (meaning that vendors have considerable freedom to implement it). AnLF is the sole consumer network element for the services provided by MTLF. A single NWDAF network element can integrate both AnLF and MTLF.

[0074] Figure 1 The network architecture diagram of the mobile communication network applying the technical solution of the present application is schematically shown.

[0075] like Figure 1 As shown in the figure, the mobile communication network mainly includes an access network 101, a bearer network 102, and a core network 103. The access network 101 is used to connect terminal devices 104, such as mobile phones, to the network. Terminal devices 104 are connected to base stations via an air interface. The bearer network 102 is used to carry data, for example, by optical fiber between various network elements. The core network 103 is the management center of the mobile communication network and is a carrier-grade router used to manage data such as location management, updates, authentication, and connections for terminal devices 101.

[0076] The technical solutions of the embodiments of the present application can be applied to various communication systems, such as: global system for mobile communications (GSM) system, code division multiple access (CDMA) system, wideband code division multiple access (WCDMA) system, general packet radio service (GPRS), long term evolution (LTE) system, LTE frequency division duplex (FDD) system, LTE time division duplex (TDD) system, universal mobile telecommunication system (UMTS), world-wide interoperability for microwave access (WiMAX) communication system, fifth generation (5G) communication system or new radio (NR) communication system, etc.

[0077] Figure 2 The network architecture applicable to the embodiments of the present application is schematically shown, and the various parts involved in the network architecture are described below.

[0078] User equipment (UE) 201: may include various handheld terminal devices, vehicle-mounted devices, and wearable devices with wireless communication functions, such as cellular phones, smartphones, wireless data cards, personal digital assistants (PDAs), laptops, tablet computers, wireless modems, handheld devices, laptop computers, cordless phones, wireless local loop (WLL) stations, machine type communication (MTC) terminals, and other devices that can access the network.

[0079] Access network (AN) network element 202: It is mainly responsible for wireless resource management, quality of service (QoS) management, data compression and encryption, etc. on the air interface side. It is used to provide network access functions for authorized user equipment in a specific area, and can use transmission tunnels of different qualities according to the level of the user equipment and service requirements. In the wireless access scenario, the access network network element 202 can also be called a radio access network (RAN) network element. The RAN network element can manage wireless resources, provide access services for user equipment, and then complete the forwarding of control signals and user equipment data between the user equipment and the core network. The RAN network element can include various forms of base stations, such as macro base stations, micro base stations (also known as small stations), relay stations, access points, etc. In systems using different radio access technologies, the names of RAN network elements with base station functions may vary. For example, in the fifth generation (5G) system, it is called gNB; in the LTE system, it is called evolved NodeB (eNB or eNodeB); in the third generation (3G) system, it is called Node B, etc.

[0080] User plane network element 203: responsible for the encapsulation, routing, forwarding, statistics of user messages, and quality of service (QoS) processing of user plane data. In a 5G communication system, the user plane network element may be a user plane function (UPF) network element.

[0081] Data network element 204: a network for providing data transmission. In a 5G communication system, the data network element may be a data network (DN) element.

[0082] Access and mobility management network element 205: Mainly used for mobility management and access management, etc., and can be used to implement other functions of the mobility management entity (MME) in addition to session management, such as legal detection and access authorization / authentication. In the 5G communication system, the access and mobility management network element can be the access and mobility management function (AMF) network element.

[0083] Session management network element 206: It is mainly used for session management, allocation and management of Internet protocol (IP) addresses for user equipment, selection of endpoints for manageable user plane functions, policy control and charging function interfaces, and downlink data notification. In a 5G communication system, this session management network element can be a session management function (SMF) network element.

[0084] Policy control network element 207: A unified policy framework for guiding network behavior, providing policy rule information to control plane function network elements (such as AMF and SMF network elements). In 4G communication systems, this policy control network element can be a policy and charging rules function (PCRF) network element. In 5G communication systems, this policy control network element can be a policy control function (PCF) network element.

[0085] Authentication server 208: used for authentication services, generating keys to achieve two-way authentication of user equipment, and supporting a unified authentication framework. In the 5G communication system, the authentication server can be an authentication server function (AUSF) network element.

[0086] Data management network element 209: is used to process user equipment identification, access authentication, registration, and mobility management. In 5G communication systems, this data management network element can be a unified data management (UDM) network element; in 4G communication systems, this data management network element can be a home subscriber server (HSS) network element.

[0087] Application network element 210: used for data routing affected by applications, accessing network open function network elements, interacting with the policy framework for policy control, etc. In the 5G communication system, the application network element can be an application function (AF) network element.

[0088] Network storage network element: used to maintain real-time information of all network function services in the network. In the 5G communication system, this network storage network element can be the network registration function (NRF) network element.

[0089] exist Figure 2 In the network architecture shown, the user equipment (UE) connects to the AMF via the N1 interface, the (R)AN connects to the AMF via the N2 interface, and the (R)AN connects to the UPF via the N3 interface. The UPFs are connected to each other via the N9 interface, and the UPFs are interconnected with the DN via the N6 interface. The SMF controls the UPFs via the N4 interface. The AMF connects to the SMF via the N11 interface. The AMF obtains user equipment subscription data from the UDM unit via the N8 interface, and the SMF obtains user equipment subscription data from the UDM unit via the N10 interface.

[0090] It is understandable that the above-mentioned network elements or functions can be network elements in hardware devices, software functions running on dedicated hardware, or virtualized functions instantiated on a platform (e.g., a cloud platform). The above-mentioned network architecture applied to the embodiment of the present application is only an example, and the network architecture applicable to the embodiment of the present application is not limited thereto. Any network architecture that can realize the functions of the above-mentioned network elements is applicable to the embodiment of the present application.

[0091] For example, in some network architectures, network function network element entities such as AMF network elements, SMF network elements, PCF network elements, AUSF network elements and UDM network elements are all called network function (NF) network elements; or, in other network architectures, a collection of network elements such as AMF network elements, SMF network elements, PCF network elements, AUSF network elements, UDM network elements, etc. can be called control plane function network elements.

[0092] It should be noted that Figure 1 and Figure 2 The interface names between the various network elements in the embodiment are only examples. The names of the interfaces in the specific implementation may be other names, and the embodiments of the present application do not specifically limit this. Figure 1 and Figure 2 The names of the various network elements (for example, SMF, AF, UPF, etc.) included in the above are only examples and do not limit the functions of the network elements themselves. In 5GS and other future networks, the above network elements may also have other names, which are not specifically limited in the embodiments of the present application. For example, in a 6G network, some or all of the above network elements may continue to use the terminology in 5G, or may adopt other names. In addition, it should be understood that the names of the messages (or signaling) transmitted between the above network elements are only examples and do not limit the functions of the messages themselves.

[0093] The following describes in detail the technical solutions provided by the present application, including the network element interaction method, network element interaction device, computer-readable medium, electronic device, and computer program product, in combination with specific implementation methods.

[0094] Figure 3 A flowchart of a network element interaction method in an embodiment of the present application is shown. The network element interaction method can be performed by Figure 2 Any one or more network elements in the mobile communication network shown in FIG. Figure 3 As shown, the network element interaction method may include the following steps S310 to S340.

[0095] S310: Selecting a publisher network element in a mobile communication network to establish an information subscription relationship with a subscriber network element, where the information subscription relationship is used to indicate that the publisher network element provides data analysis information related to original data of the subscription content to the subscriber network element.

[0096] S320: Data alignment is performed between the subscriber network element and the publisher network element to obtain identification information of intersection data. The intersection data is data having the same identification information between the original data held by the subscriber and the original data held by the publisher network element.

[0097] S330: Select publisher sample data on the publisher network element according to the identification information of the intersection data, and train a machine learning model corresponding to the subscription content based on the publisher sample data to obtain the model intermediate information of the publisher network element.

[0098] S340: Send the intermediate model information of the publisher network element to the subscriber network element, and train a machine learning model corresponding to the subscription content on the subscriber network element based on the intermediate model information of the publisher network element.

[0099] In the network element interaction method provided in the embodiment of the present application, by establishing an information subscription relationship between different network elements, data analysis information can be shared between different network elements while keeping the original data held by the network element itself from being leaked, so that the shared data analysis information can be used to realize the training of the machine learning model. On the one hand, it can ensure the privacy and security of the data, and on the other hand, it can improve the wide applicability of the machine learning model in the mobile communication network.

[0100] The following describes in detail the various steps of the network element interaction method in this application in combination with multiple embodiments.

[0101] In step S310, a publisher network element is selected in the mobile communication network to establish an information subscription relationship with the subscriber network element. The information subscription relationship is used to indicate that the publisher network element provides data analysis information related to the original data of the subscription content to the subscriber network element.

[0102] The subscription content corresponds to the application requirements of the subscribing network element. For example, the subscribing network element needs to adjust the corresponding service content according to the network quality of the mobile communication network. However, the various network elements in the mobile communication network usually do not share the original data related to the network quality with other network elements. Therefore, the subscribing network element can obtain data analysis information related to the network quality from the publishing network element by establishing an information subscription relationship. Subsequently, the network quality of the mobile communication network can be determined based on the original data analysis information related to the network quality provided by one or more publishing network elements.

[0103] Data analysis information refers to derived data generated by analyzing and processing the original data held by the publisher's network element based on the subscription content. In the application scenario of machine learning model training, data analysis information may include, for example, derived data related to the sample data and derived data related to the model. Derivative data related to the sample data may include, for example, identification information of the sample data and data features extracted from the sample data. Derivative data related to the model may include, for example, model parameters obtained by model training of each network element or gradient information used to update the model parameters.

[0104] In one embodiment of the present application, a subscribing network element is a network element outside the core network of a mobile communication network, and a publishing network element is a network element inside the core network of the mobile communication network. By establishing an information subscription relationship between the subscribing network element and the publishing network element, the network element inside the core network can provide data analysis information related to the original data of the subscribed content to the network element outside the core network. While ensuring that the original data of the core network is not leaked externally, derived data that meets application requirements can be provided to application nodes outside the core network.

[0105] In one embodiment of the present application, the subscribing network element is an application function network element AF for implementing application layer services, and the publishing network element is a network data analysis function network element NWDAF for providing network data analysis services to the mobile communication network.

[0106] 3GPP R18 23.700-80 provides 47 AI / ML-based solutions to support AI / ML-based services / applications. 5GS network resources are used for AI / ML model and data distribution, transmission, and / or training for various applications. To monitor 5G network resource usage, it is necessary to obtain network resource utilization related to terminal performance (e.g., data throughput provided to the terminal) to support the application of AI / ML model / data distribution and sharing related operations.

[0107] In some related technologies, the application function network element AF has label data but lacks core network network information. The core network is unwilling to share it with the application function network element AF or service server. Therefore, the application function network element AF or service server cannot accurately perform QoS or QoE analysis.

[0108] To address this problem, the embodiment of the present application uses the application function network element AF outside the core network as the subscriber network element and the network data analysis function network element NWDAF inside the core network as the publisher network element. The application function network element AF establishes an information subscription relationship with the network data analysis function network element NWDAF, so that the application function network element AF can obtain data analysis information of the original data from the network data analysis function network element NWDAF. On the one hand, it can maintain the privacy and security of the core network data, and on the other hand, it can meet the business needs of the application function network element AF to detect the usage of 5G network resources.

[0109] In one embodiment of the present application, a sample collection request is initiated by the network data analysis function network element NWDAF to the operation administration and maintenance management network element OAM (Operation Administration and Maintenance) in the mobile communication network; and the original data held by the network data analysis function network element NWDAF is obtained according to the sample collection request.

[0110] In one embodiment of the present application, the original data held by the network data analysis function network element includes channel quality indication information CQI (Channel Quality Indicator) used to indicate the quality of the wireless channel. CQI is an indicator used in wireless communication systems to indicate the quality of the wireless channel. CQI is commonly used in wireless communication technologies such as LTE (Long-Term Evolution) and 5G to help dynamically adjust modem parameters between base stations and terminal devices to maximize communication quality and data transmission rate. By analyzing the indications of such information and providing them to AF, better services can be provided to the business, allowing the business to better adapt to complex network changes.

[0111] In one embodiment of the present application, selecting a publishing network element in a mobile communication network to establish an information subscription relationship with a subscribing network element may further include: sending an information open request of the subscribing network element to a network open function network element NEF in the mobile communication network, wherein the information open request carries the network element identifier and subscription content of the subscribing network element; establishing a mapping relationship between the network element identifier of the subscribing network element and the subscription content on the network open function network element, and initiating an information subscription request to establish an information subscription relationship to the publishing network element based on the subscription content.

[0112] The network element identifier of the subscribing network element is identification information used to differentiate different network elements. In this embodiment of the present application, the network open function network element (NEF) provides an information exchange channel between the subscribing network element and the publishing network element. Furthermore, the network open function network element (NEF) maintains the mapping relationship between the network element identifier of the subscribing network element and the subscription content, thereby effectively managing the interaction matching relationship between the subscribing network element and the publishing network element. This improves interaction efficiency while ensuring the stability and security of information interaction.

[0113] In one embodiment of the present application, before establishing a mapping relationship between the network element identifier of the subscriber network element and the subscription content on the network open function network element, the subscriber network element can be authorized to verify based on the network element identifier of the subscriber network element. The authorization verification is used to determine whether the subscriber network element has system access rights; if the verification result of the authorization verification is a verification failure, the information opening request of the subscriber network element is rejected.

[0114] In the embodiment of the present application, before establishing an information subscription relationship, the network open function network element NEF verifies the authority of the subscriber network element that initiates the request. Only when the subscriber network element has system access rights will it initiate an information subscription request to the publisher network element corresponding to the subscription content, thereby improving the information security of the core network.

[0115] In step S320, data alignment is performed between the subscriber network element and the publisher network element to obtain identification information of intersection data. The intersection data is data with the same identification information between the original data held by the subscriber and the original data held by the publisher network element.

[0116] In one embodiment of the present application, a method for aligning data between a subscriber network element and a publisher network element may include private set intersection, i.e., obtaining the intersection data between the original data through private computing without transmitting the original data to each other. Examples of private set intersection methods include Blind RSA-based PSI Protocol with linear complexity, Diffie-Hellman-based schemes, oblivious transfer (OT)-based schemes, Freedman secure intersection protocol, and the like.

[0117] In one embodiment of the present application, in order to further improve the reliability of data privacy and security, data alignment between the subscriber network element and the publisher network element may be performed by configuring a trusted third party.

[0118] Figure 4 A flowchart of data alignment based on a trusted third party in one embodiment of the present application is shown. Figure 4 As shown, based on the above embodiment, the method for aligning data between a subscriber network element and a publisher network element may include the following steps S410 to S440.

[0119] S410: Select a third-party network element that has established a trusted computing relationship with both the subscriber network element and the publisher network element. The trusted computing relationship is used to indicate that the third-party network element can perform trusted computing on the original data held by the subscriber network element and the publisher network element while maintaining data privacy.

[0120] S420: Send the identification information of the original data held by the subscriber network element and the publisher network element respectively to the third-party network element.

[0121] S430: Align identification information of the original data held by the subscriber network element and the publisher network element on the third-party network element to obtain identification information of the intersection data.

[0122] S440: The third-party network element sends identification information of the intersection data to the subscriber network element and the publisher network element respectively.

[0123] Trusted computing is a computer system security technology that combines cryptographic operations with protection to ensure full detectability. The principle of trusted computing is to ensure that the entire chain undergoes trusted authentication, so whether it is an application, operating system, or hardware, it must be authorized for use. One of the core goals of trust is to ensure the integrity of the system and applications, thereby ensuring that the system or software operates in the trusted state expected by the design objectives. The root of trust is the basis for the trustworthiness of a trusted computer system and is divided into the root of trust measurement (RTM, the first piece of software executed when the platform starts), the root of trust storage (RTS, a group of memories and storage root keys called platform configuration registers in the trusted platform module chip), and the root of trust reporting (RTR, the platform configuration registers and endorsement keys in the trusted platform module chip). The trust chain is the technical implementation solution for the trust measurement model, which extends the trust relationship from the root of trust to the entire computer system through the trust chain.

[0124] The fundamental concept of trusted computing is to ensure security through remote attestation of digital rights. For example, in a computer system, a root of trust is first established, followed by a chain of trust. This chain of trust extends from the root of trust to the hardware platform, the operating system, and finally to the application. This chain of trust is then measured and authenticated, and each level of trust is then extended to the entire computer system, ensuring its integrity.

[0125] In one embodiment of the present application, selecting a third-party network element that establishes a trusted computing relationship with both the subscriber network element and the publisher network element may further include: obtaining a first network element set that establishes a trusted computing relationship with the subscriber network element and a second network element set that establishes a trusted computing relationship with the publisher network element; if there are shared network elements between the first network element set and the second network element set, selecting the shared network elements as the third-party network elements; if there are no shared network elements between the first network element set and the second network element set, selecting a temporary network element as the third-party network element through the network open function network element in the mobile communication network.

[0126] By maintaining a set of network elements that establish a trusted computing relationship with themselves on different network elements, the embodiment of the present application can select a shared network element as a third-party network element by taking the intersection of the network element set when it is necessary to select a trusted third-party network element. This can avoid the need to repeatedly establish mutual trust between network elements each time network element interaction is performed. On the one hand, it can improve the efficiency of data alignment between different network elements, and on the other hand, it can reduce the network resource consumption and computing resource consumption of data alignment.

[0127] In one embodiment of the present application, a temporary network element is a network element selected from a third network element set maintained by a network open function network element, which is different from the first network element set and the second network element set. The temporary network element establishes a trusted computing relationship with a specified validity period with the subscriber network element and the publisher network element respectively through the network open function network element.

[0128] The network openness function network element can periodically select network elements from the mobile communication network that meet specified security certification standards to form a third network element set. When the subscribing network element and the publishing network element do not have a mutually trusted network element in common, a temporary network element can be directly selected from the third network element set. Furthermore, this temporary network element does not belong to either the first or second network element sets. This ensures basic network element trustworthiness and prevents the temporary network element from colluding with either the subscribing network element or the publishing network element.

[0129] The temporary network element establishes a trusted computing relationship with the subscriber network element and the publisher network element with a specified validity period through the network open function network element. If the specified validity period is exceeded, the computing relationship needs to be re-established, thereby ensuring the security and timeliness of the temporary network element.

[0130] In step S330, publisher sample data is selected on the publisher network element according to the identification information of the intersection data, and a machine learning model corresponding to the subscription content is trained based on the publisher sample data to obtain the model intermediate information of the publisher network element.

[0131] In one embodiment of the present application, the subscribing network element and the publishing network element may each hold different data contents for the same raw data identification information. For example, for the same terminal device, the data held by the subscribing network element is the battery level of the terminal device, while the data held by the publishing network element is the guaranteed bit rate (GBR) allocated to the terminal device by the mobile communication network. Machine learning models for predicting terminal device performance can be trained based on the sample data held by the subscribing network element and the publishing network element, respectively.

[0132] The intermediate model information may be, for example, model parameters obtained during the model training process, or gradient information used to update the model parameters.

[0133] In step S340, the model intermediate information of the publisher network element is sent to the subscriber network element, and a machine learning model corresponding to the subscription content is trained on the subscriber network element based on the model intermediate information of the publisher network element.

[0134] In one embodiment of the present application, after the subscribing network element receives the model intermediate information sent by the publishing network element, it can continue to train the machine learning model corresponding to the subscribed content based on the model intermediate information and the sample data it holds, thereby achieving collaborative training of the machine learning model without disclosing the original data to each other.

[0135] In one embodiment of the present application, a method for training a machine learning model corresponding to the subscription content on a subscriber network element based on the model intermediate information of the publisher network element may include: selecting subscriber sample data on the subscriber network element based on the identification information of the intersection data, and training the machine learning model corresponding to the subscription content based on the subscriber sample data to obtain the model intermediate information of the subscriber network element; training the machine learning model corresponding to the subscription content on the subscriber network element based on the model intermediate information of the subscriber network element and the model intermediate information of the publisher network element.

[0136] In addition to using the model intermediate information provided by the publisher network element, the subscriber network element itself can also perform synchronous model training in the embodiment of the present application to obtain the corresponding model intermediate information, and then fuse the model intermediate information of the two network elements to jointly train the machine learning model corresponding to the subscribed content. Compared with separate training, the model training speed can be increased exponentially.

[0137] In one embodiment of the present application, the fusion of model intermediate information may be performed by a trusted third party. Figure 5 The flowchart of training a machine learning model based on a third-party network element in one embodiment of the present application is shown. Figure 5 As shown, based on the above embodiments, training a machine learning model corresponding to the subscription content on the subscriber network element according to the model intermediate information of the subscriber network element and the model intermediate information of the publisher network element can include the following steps S510 to S540.

[0138] S510: Select a third-party network element that has established a trusted computing relationship with both the subscriber network element and the publisher network element.

[0139] S520: Send the intermediate model information of the subscriber network element and the intermediate model information of the publisher network element to the third-party network element respectively.

[0140] S530: Aggregate the model intermediate information of the subscriber network element and the model intermediate information of the publisher network element on the third-party network element to obtain global information.

[0141] S540: Iteratively update the model parameters of the machine learning model corresponding to the subscription content on the subscriber network element and the publisher network element respectively according to the global information.

[0142] By aggregating intermediate model information on a third-party network element, the iterative update process of model parameters that originally needed to be performed on multiple network elements can be concentrated on one network element, thereby greatly reducing the network resource consumption and computing resource consumption of model training.

[0143] The method for selecting the third-party network element can be referred to the detailed description in other embodiments and will not be repeated here.

[0144] In one embodiment of the present application, the subscribing network element and the publishing network element train a machine learning model corresponding to the subscription content through federated learning.

[0145] Federated learning is a machine learning approach for training models that can learn from data from multiple parties without requiring the data to be centralized in one place. This approach is particularly useful in situations involving privacy-sensitive data because it allows model training to be performed without sharing the original data.

[0146] In one embodiment of the present application, the subscribing network element and the publishing network element train a machine learning model corresponding to the subscribed content through vertical federated learning; the training samples held by the subscribing network element and the training samples held by the publishing network element have the same sample identifier and different characteristics.

[0147] Federated learning includes two methods: horizontal federated learning and vertical federated learning. Horizontal federated learning means that multiple participants have different sample IDs but the same features; vertical federated learning means that multiple participants have the same sample ID but different features.

[0148] For example, the subscriber network element is an AF network element located outside the core network in the 5G network. The training samples held by the AF network element are the performance data of the terminal device, such as the device's power, memory usage and other performance indicators; the publisher network element is an NWDAF network element located inside the core network in the 5G network. The training samples held by the NWDAF network element are the guaranteed bit rate GBR (Guaranteed Bit Rate) allocated to the terminal device.

[0149] The embodiments of the present application implement joint modeling and training between subscriber and publisher network elements through vertical federated learning. For example, in a 5G network, a service server can request analysis results from the NWDAF, which then uses information from the RAN, UE, and AF to perform joint modeling, analyze the deduction results, and return the results to the service server.

[0150] The following describes the network element interaction method in this application in detail in conjunction with specific application scenarios.

[0151] In one application scenario of this application, it is often encountered that the AF wants to analyze user behavior or QoE status based on the network status, so it needs to collect some network information from the core network, but the core network is unwilling to share it with the AF due to privacy issues. For example, the service requires some information such as UE mobility, but this information is on the core network side. Without this information, the service cannot accurately judge the current QoE level. Generally, it is necessary to judge QoE by collecting network performance parameters in the form of QoS indicators of this information (such as packet delay, bit rate, packet loss, etc.).

[0152] Figure 6 The following is an architectural diagram of an embodiment of the present application for implementing network element interaction in an application scenario.

[0153] like Figure 6 As shown, the AF can exchange information with the NWDAF of the 5G core network through the network element interaction method provided in the embodiment of the present application. If the information of other network elements, such as NEF, PCF, SMF, AMF, UPF, etc., is unwilling to share information with the NWDAF, the network element interaction method provided in the embodiment of the present application can also be used between the NWDAF and these NF network elements. In this case, the network element interaction is performed through direct interaction between the two parties under the premise of mutual trust between the two parties.

[0154] Another approach is to use a trusted third party as a trusted party between the AF and NWDAF, responsible for exchanging intermediate information during interactive training. This trusted third party does not necessarily refer to a third party outside the 5G system. Rather, it refers to a third party involved in information exchange between the AF and NWDAF or the core network. By default, both the AF and NWDAF trust the third party and are willing to exchange data with it. This trusted third party can be a network element within the 5G system or any terminal or server external to the 5G system.

[0155] Figure 7 The flowchart of the direct interaction between two network elements in an application scenario of the embodiment of the present application is shown. Figure 7 As shown, the process based on direct interaction between two network elements may include the following steps S701 to S712.

[0156] S701: The AF / OTT Server initiates an information disclosure request to the NEF.

[0157] The information disclosure request (Nnef_AnalyticsExposurerequest) is used to request subscription or unsubscription of data analysis information. The data analysis information may include, for example, feature / sample alignment information, and may also include model intermediate information such as weights and gradients.

[0158] AF will not request the original information of the network, and at the same time, AF will not expose its own original information to the network, ensuring that AF and NWDAF can exchange information while protecting privacy.

[0159] S702: NEF initiates an information subscription request to NWDAF.

[0160] NEF requests to subscribe to data analysis information or request to unsubscribe data analysis information by calling services, such as Nnef_AnalyticsExposure_Subscribe or Nnef_AnalyticsExposure_Unsubscribe (defined by 3GPP TS23.502). NEF is responsible for controlling the analysis exposure mapping between AF identifiers and allowed analysis IDs and associated parameters / parameter values.

[0161] After receiving the request from AF, NEF can authorize AF and verify whether the user corresponding to AF has the right to access the system. Assuming that the user has the right to access the system, NEF subscribes or unsubscribes to the analytical information by calling the Nnwdaf_AnalyticsSubscription_Subscribe or Nnwdaf_AnalyticsSubscription_Unsubscribe service operation according to the AF's request.

[0162] S703: NWDAF returns the notification analysis information or termination request to NEF.

[0163] If NEF subscribes to the analytics information, NWDAF notifies NEF of the analytics information or the termination request by calling the Nnwdaf_AnalyticsSubscription_Notify service operation.

[0164] S704: NEF notifies AF of analysis information or termination request.

[0165] If NEF receives a notification from NWDAF, NEF notifies AF of the analytics information or termination request by calling the Nnwdaf_AnalyticsSubscription_Notify service operation defined in TS23.502.

[0166] S705: NEF initiates a request to NWDAF to obtain analysis information.

[0167] NEF requests analysis information by calling the Nnwdaf_AnalyticsInfo_Request service operation. The parameters that NWDAF service consumers can provide include feature / sample alignment, inter-layer information (weights, gradients), etc.

[0168] S706: NWDAF collects information from OAM.

[0169] The information that OAM can provide to the NWDAF is clearly defined in 3GPP TS 23.288. The information required here mainly includes CQI (Channel Quality Indicator). Network information clearly defined in TS 23.288 can also be provided.

[0170] In the specifications defined in 3GPP R18, NWDAF can collect information from other NWDAFs, and other network elements, such as NEF, PCF, SMF, AMF, etc., can also use the solution provided in the embodiments of this application to interactively collect information. This situation is generally used when other network elements are unwilling to open information to NWDAF. (The premise that NWDAF can collect information from other NWDAFs is that different NWDAFs are in the same region. NWDAFs in different regions cannot share information for legal reasons).

[0171] S707: AF and NWDAF perform sample alignment.

[0172] Sample alignment is used to find the intersection of cross-domain samples. AF and NWDAF can also align other interactive information, such as gradient information and private key encryption information. Gradient information reflects the optimization direction of the machine learning model; private key encryption information refers to highly sensitive information that requires encryption, such as homomorphic encryption.

[0173] S708: AF trains the model locally based on the aligned information.

[0174] S709: NWDAF trains the model locally based on the aligned information.

[0175] Here, you can use the MTLF module of NWDAF to train (calculate model loss) and test the machine learning model.

[0176] TS23.288 clearly defines the collaboration between NWDAF MTLF and NWDAF service consumers to train ML models.

[0177] S710: The AF initiates an intermediate information request to the NWDAF through the control plane.

[0178] S711: NWDAF returns to AF via the control plane.

[0179] The reply information is the intermediate information, including gradient information, cross-layer information, etc.

[0180] S712: AF executes the deduction process of the machine learning model.

[0181] By repeating the above process, the iterative update of the machine learning model can be completed based on the direct interaction between the two network elements. In some optional implementations, the AF can directly interact with the NWDAF to perform vertical federated learning model training.

[0182] Figure 8 The flowchart of the embodiment of the present application in an application scenario through the third-party network element interaction is shown. Figure 8 As shown, the process of interaction based on the third-party network element may include the following steps S801 to S814.

[0183] S801: The AF / OTT Server initiates an information disclosure request to the NEF.

[0184] The information disclosure request (Nnef_AnalyticsExposurerequest) is used to request subscription or unsubscription of data analysis information. The data analysis information may include, for example, feature / sample alignment information, and may also include model intermediate information such as weights and gradients.

[0185] AF will not request the original information of the network, and at the same time, AF will not expose its own original information to the network, ensuring that AF and NWDAF can exchange information while protecting privacy.

[0186] S802: NEF initiates an information subscription request to NWDAF.

[0187] NEF requests to subscribe to data analysis information or request to unsubscribe data analysis information by calling services, such as Nnef_AnalyticsExposure_Subscribe or Nnef_AnalyticsExposure_Unsubscribe (defined by 3GPP TS23.502). NEF is responsible for controlling the analysis exposure mapping between AF identifiers and allowed analysis IDs and associated parameters / parameter values.

[0188] After receiving the request from AF, NEF can authorize AF and verify whether the user corresponding to AF has the right to access the system. Assuming that the user has the right to access the system, NEF subscribes or unsubscribes to the analytical information by calling the Nnwdaf_AnalyticsSubscription_Subscribe or Nnwdaf_AnalyticsSubscription_Unsubscribe service operation according to the AF's request.

[0189] S803: NWDAF returns the notification analysis information or termination request to NEF.

[0190] If NEF subscribes to the analytics information, NWDAF notifies NEF of the analytics information or the termination request by calling the Nnwdaf_AnalyticsSubscription_Notify service operation.

[0191] S804: NEF notifies AF of analysis information or termination request.

[0192] If NEF receives a notification from NWDAF, NEF notifies AF of the analytics information or termination request by calling the Nnwdaf_AnalyticsSubscription_Notify service operation defined in TS23.502.

[0193] S805: NEF initiates a request to NWDAF to obtain analysis information.

[0194] NEF requests analysis information by calling the Nnwdaf_AnalyticsInfo_Request service operation. The parameters that NWDAF service consumers can provide include feature / sample alignment, inter-layer information (weights, gradients), etc.

[0195] S806: NWDAF collects information from OAM.

[0196] The information that OAM can provide to the NWDAF is clearly defined in 3GPP TS 23.288. The information required here mainly includes CQI (Channel Quality Indicator). Network information clearly defined in TS 23.288 can also be provided.

[0197] In the specifications defined in 3GPP R18, NWDAF can collect information from other NWDAFs, and other network elements, such as NEF, PCF, SMF, AMF, etc., can also use the solution provided in the embodiments of this application to interactively collect information. This situation is generally used when other network elements are unwilling to open information to NWDAF. (The premise that NWDAF can collect information from other NWDAFs is that different NWDAFs are in the same region. NWDAFs in different regions cannot share information for legal reasons).

[0198] S807: AF and NWDAF perform sample alignment through a third-party network element.

[0199] After the trusted third party confirms with both the AF and NWDAF, it initiates sample alignment through the third-party network element to identify cross-domain sample intersections. The AF and NWDAF can also align other interactive information, such as gradient information and private key encryption. Gradient information reflects the optimization direction of the machine learning model; private key encryption refers to highly sensitive information that requires encryption, such as homomorphic encryption.

[0200] In an optional implementation, the AF and NWDAF can each send their own sample feature names (IDs) to a trusted third party. The trusted third party performs a privacy intersection, obtains the intersection ID, and returns the intersection ID to the AF and NWDAF via the control plane. Sample alignment can then be performed on the AF and NWDAF, respectively.

[0201] S808: AF trains the model locally based on the aligned information.

[0202] S809: NWDAF trains the model locally based on the aligned information.

[0203] Here, you can use the MTLF module of NWDAF to train (calculate model loss) and test the machine learning model.

[0204] TS23.288 clearly defines the collaboration between NWDAF MTLF and NWDAF service consumers to train ML models.

[0205] S810: The AF initiates an intermediate information request to the NWDAF through the control plane using a trusted third party.

[0206] S811: The NWDAF uses a trusted third party to reply to the AF through the control plane.

[0207] The reply information is the intermediate information, including gradient information, cross-layer information, etc.

[0208] S812: The trusted third party aggregates the collected parameters into a global model to update the parameters of the global model.

[0209] S813: The trusted third party feeds back the intermediate results to the AF and NWDAF respectively.

[0210] The intermediate results include one or more of gradient information update, private key encryption information update, and other system statistical parameter update.

[0211] S814: AF executes the deduction process of the machine learning model.

[0212] By repeating the above process, it is possible to complete the iterative update of the machine learning model based on the interaction with the third-party network element. In some optional implementations, the AF can interact with the NWDAF through the third-party network element to perform model training for vertical federated learning.

[0213] Based on the introduction of the above embodiments and application scenarios, it can be seen that this invention implements a solution for network element interaction between the core network and application services, that is, it realizes the interactive service process between NWDAF and AF, and enables the core network to open AI services to businesses under the premise of privacy protection through a third party.

[0214] It should be noted that although the steps of the method of the present application are described in a specific order in the drawings, this does not require or imply that the steps must be performed in this specific order, or that all steps must be performed to achieve the desired results. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step, and / or one step may be decomposed into multiple steps.

[0215] The following introduces an embodiment of the device of the present application, which can be used to execute the network element interaction method in the above embodiment of the present application. Figure 9 The structure block diagram of the network element interaction device provided in the embodiment of the present application is schematically shown. Figure 9 As shown, the network element interaction device 900 includes:

[0216] The network element selection module 910 is configured to select a publishing network element in the mobile communication network that establishes an information subscription relationship with a subscribing network element, wherein the information subscription relationship is used to indicate that the publishing network element provides data analysis information related to the original data of the subscribed content to the subscribing network element;

[0217] a data alignment module 920 configured to perform data alignment between the subscriber network element and the publisher network element to obtain identification information of intersection data, where the intersection data is data having the same identification information between the original data held by the subscriber and the original data held by the publisher network element;

[0218] The sample selection module 930 is configured to select publisher sample data on the publisher network element according to the identification information of the intersection data, and train a machine learning model corresponding to the subscription content according to the publisher sample data to obtain intermediate model information of the publisher network element;

[0219] The model training module 940 is configured to send the model intermediate information of the publisher network element to the subscriber network element, and train a machine learning model corresponding to the subscription content on the subscriber network element based on the model intermediate information of the publisher network element.

[0220] In some embodiments of the present application, based on the above technical solution, the data alignment module includes:

[0221] a third-party network element selection module configured to select a third-party network element that has established a trusted computing relationship with both the subscriber network element and the publisher network element, wherein the trusted computing relationship indicates that the third-party network element can perform trusted computing on the original data held by each of the subscriber network element and the publisher network element while maintaining data privacy;

[0222] an identification information sending module, configured to send identification information of the original data respectively held by the subscriber network element and the publisher network element to the third-party network element;

[0223] an identification information alignment module configured to align identification information of the original data respectively held by the subscriber network element and the publisher network element on the third-party network element to obtain identification information of intersection data;

[0224] The intersection data sending module is configured to send the identification information of the intersection data to the subscriber network element and the publisher network element respectively by the third-party network element.

[0225] In some embodiments of the present application, based on the above technical solution, the third-party network element selection module includes:

[0226] A network element set acquisition module is configured to acquire a first network element set that establishes a trusted computing relationship with the subscriber network element and a second network element set that establishes a trusted computing relationship with the publisher network element;

[0227] A first selection module is configured to select a common network element as a third-party network element if there is one between the first network element set and the second network element set;

[0228] The second selection module is configured to select a temporary network element as the third-party network element through a network open function network element in the mobile communication network if there is no common network element between the first network element set and the second network element set.

[0229] In some embodiments of the present application, based on the above technical solution, the temporary network element is a network element selected from a third network element set maintained by the network open function network element, which is different from the first network element set and the second network element set. The temporary network element establishes a trusted computing relationship with a specified validity period with the subscriber network element and the publisher network element respectively through the network open function network element.

[0230] In some embodiments of the present application, based on the above technical solution, the network element selection module includes:

[0231] an information opening request module, configured to send an information opening request of the subscribing network element to a network opening function network element in the mobile communication network, wherein the information opening request carries a network element identifier and subscription content of the subscribing network element;

[0232] The information subscription request module is configured to establish a mapping relationship between the network element identifier of the subscriber network element and the subscription content on the network open function network element, and initiate an information subscription request to establish an information subscription relationship to the publisher network element according to the subscription content.

[0233] In some embodiments of the present application, based on the above technical solution, the network element selection module further includes:

[0234] A network element authority verification module is configured to perform authority verification on the subscriber network element according to the network element identifier of the subscriber network element, wherein the authority verification is used to determine whether the subscriber network element has system access authority;

[0235] The execution rejection module is configured to reject the execution of the information opening request of the subscriber network element if the verification result of the authority verification is verification failure.

[0236] In some embodiments of the present application, based on the above technical solution, the model training module includes:

[0237] a sample data selection module configured to select subscriber sample data on the subscriber network element according to the identification information of the intersection data, and train a machine learning model corresponding to the subscription content according to the subscriber sample data to obtain model intermediate information of the subscriber network element;

[0238] The intermediate information training module is configured to train a machine learning model corresponding to the subscription content on the subscriber network element based on the model intermediate information of the subscriber network element and the model intermediate information of the publisher network element.

[0239] In some embodiments of the present application, based on the above technical solution, the intermediate information training module is further configured to: select a third-party network element that establishes a trusted computing relationship with both the subscriber network element and the publisher network element; send the model intermediate information of the subscriber network element and the model intermediate information of the publisher network element to the third-party network element respectively; aggregate the model intermediate information of the subscriber network element and the model intermediate information of the publisher network element on the third-party network element to obtain global information; and iteratively update the model parameters of the machine learning model corresponding to the subscription content on the subscriber network element and the publisher network element according to the global information.

[0240] In some embodiments of the present application, based on the above technical solution, the subscribing network element is a network element outside the core network of the mobile communication network, and the publishing network element is a network element inside the core network of the mobile communication network.

[0241] In some embodiments of the present application, based on the above technical solution, the subscriber network element is an application function network element for implementing application layer services, and the publisher network element is a network data analysis function network element for providing network data analysis services to the mobile communication network.

[0242] In some embodiments of the present application, based on the above technical solution, the network element interaction device further includes:

[0243] The sample collection module is configured to initiate a sample collection request to the operation, maintenance and management network element in the mobile communication network through the network data analysis function network element; and obtain the original data held by the network data analysis function network element according to the sample collection request.

[0244] In some embodiments of the present application, based on the above technical solution, the original data held by the network data analysis function network element includes channel quality indication information for indicating the quality of the wireless channel.

[0245] The specific details of the network element interaction device provided in each embodiment of the present application have been described in detail in the corresponding method embodiments and will not be repeated here.

[0246] Figure 10 The block diagram schematically shows a computer system structure of an electronic device used to implement an embodiment of the present application.

[0247] It should be noted that Figure 10 The computer system 1000 of the electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0248] like Figure 10 As shown, the computer system 1000 includes a central processing unit (CPU) 1001, which can perform various appropriate actions and processes according to the program stored in the read-only memory 1002 (ROM) or the program loaded from the storage part 1008 into the random access memory 1003 (RAM). Various programs and data required for system operation are also stored in the random access memory 1003. The CPU 1001, the read-only memory 1002, and the random access memory 1003 are connected to each other via a bus 1004. An input / output interface 1005 (i.e., an I / O interface) is also connected to the bus 1004.

[0249] The following components are connected to the input / output interface 1005: an input section 1006 including a keyboard, a mouse, and the like; an output section 1007 including devices such as a cathode ray tube (CRT), a liquid crystal display (LCD), and a speaker; a storage section 1008 including a hard disk; and a communication section 1009 including a network interface card such as a local area network card or a modem. The communication section 1009 performs communication processing via a network such as the Internet. A drive 1010 is also connected to the input / output interface 1005 as needed. Removable media 1011, such as a magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor memory, is installed in the drive 1010 as needed, so that computer programs read therefrom can be installed into the storage section 1008 as needed.

[0250] In particular, according to an embodiment of the present application, the processes described in the various method flow charts can be implemented as computer software programs. For example, an embodiment of the present application includes a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for executing the methods shown in the flow charts. In such an embodiment, the computer program can be downloaded and installed from a network via the communication portion 1009 and / or installed from a removable medium 1011. When the computer program is executed by the central processing unit 1001, the various functions defined in the system of the present application are performed.

[0251] It should be noted that the computer-readable medium shown in the embodiments of the present application may be a computer-readable signal medium or a computer-readable storage medium or any combination of the two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or component, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present application, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, which carries a computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer-readable medium may be transmitted using any suitable medium, including but not limited to wireless, wired, or any suitable combination thereof.

[0252] The flowcharts and block diagrams in the accompanying drawings illustrate the possible implementation architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, program segment, or a part of code, and the above-mentioned module, program segment, or a part of code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in an order different from that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flowchart, and the combination of the boxes in the block diagram or flowchart, can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0253] It should be noted that, although several modules or units of the device for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiment of the application, the features and functions of two or more modules or units described above can be concretized in one module or unit. On the contrary, the features and functions of one module or unit described above can be further divided into multiple modules or units to be concretized.

[0254] Through the description of the above embodiments, it is easy for those skilled in the art to understand that the example embodiments described herein can be implemented by software or by combining software with necessary hardware. Therefore, the technical solution according to the embodiments of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which can be a personal computer, a server, a touch terminal, or a network device, etc.) to execute the method according to the embodiments of the present application.

[0255] Those skilled in the art will readily appreciate other embodiments of the present invention after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present invention that follow the general principles of this application and include common knowledge or customary techniques in the art that are not disclosed herein.

[0256] It should be understood that the present application is not limited to the exact structures described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.

Claims

1. A network element interaction method, characterized in that: include: Selecting, in a mobile communication network, a publisher network element that establishes an information subscription relationship with a subscriber network element, wherein the information subscription relationship indicates that the publisher network element provides data analysis information related to original data of the subscription content to the subscriber network element; Performing data alignment between the subscriber network element and the publisher network element to obtain identification information of intersection data, where the intersection data is data having the same identification information between original data held by the subscriber and original data held by the publisher network element; Selecting publisher sample data on the publisher network element according to the identification information of the intersection data, and training a machine learning model corresponding to the subscription content according to the publisher sample data to obtain model intermediate information of the publisher network element; The model intermediate information of the publisher network element is sent to the subscriber network element, and a machine learning model corresponding to the subscription content is trained on the subscriber network element based on the model intermediate information of the publisher network element.

2. The network element interaction method according to claim 1, characterized in that: Performing data alignment between the subscriber network element and the publisher network element includes: Selecting a third-party network element that has established a trusted computing relationship with both the subscriber network element and the publisher network element, wherein the trusted computing relationship indicates that the third-party network element can perform trusted computing on the original data held by the subscriber network element and the publisher network element while maintaining data privacy; Sending identification information of the original data held by the subscriber network element and the publisher network element to the third-party network element respectively; Performing data alignment on the third-party network element on the identification information of the original data held by the subscriber network element and the publisher network element to obtain identification information of the intersection data; The third-party network element sends the identification information of the intersection data to the subscriber network element and the publisher network element respectively.

3. The network element interaction method according to claim 2, characterized in that: Selecting a third-party network element that has established a trusted computing relationship with both the subscriber network element and the publisher network element includes: Acquire a first network element set that establishes a trusted computing relationship with the subscriber network element and a second network element set that establishes a trusted computing relationship with the publisher network element; If there is a common network element between the first network element set and the second network element set, selecting the common network element as the third-party network element; If there is no common network element between the first network element set and the second network element set, a temporary network element is selected as the third-party network element through a network open function network element in the mobile communication network.

4. The network element interaction method according to claim 3, characterized in that: The temporary network element is a network element selected from a third network element set maintained by the network open function network element, which is different from the first network element set and the second network element set. The temporary network element establishes a trusted computing relationship with a specified validity period with the subscriber network element and the publisher network element respectively through the network open function network element.

5. The network element interaction method according to any one of claims 1 to 4, characterized in that: Selecting a publisher network element in a mobile communication network to establish an information subscription relationship with a subscriber network element includes: Sending an information opening request of the subscribing network element to a network opening function network element in the mobile communication network, wherein the information opening request carries a network element identifier and subscription content of the subscribing network element; A mapping relationship between the network element identifier of the subscriber network element and the subscription content is established on the network open function network element, and an information subscription request for establishing an information subscription relationship is initiated to the publisher network element according to the subscription content.

6. The network element interaction method according to claim 5, characterized in that: Before establishing a mapping relationship between the network element identifier of the subscriber network element and the subscription content on the network open function network element, the method further includes: Performing authority verification on the subscriber network element according to the network element identifier of the subscriber network element, wherein the authority verification is used to determine whether the subscriber network element has system access rights; If the result of the authority verification is verification failure, the information opening request of the subscribing network element is rejected.

7. The network element interaction method according to any one of claims 1 to 4, characterized in that: Training a machine learning model corresponding to the subscription content on the subscriber network element according to the intermediate model information of the publisher network element includes: Selecting subscriber sample data on the subscriber network element according to the identification information of the intersection data, and training a machine learning model corresponding to the subscription content according to the subscriber sample data to obtain model intermediate information of the subscriber network element; A machine learning model corresponding to the subscription content is trained on the subscriber network element based on the model intermediate information of the subscriber network element and the model intermediate information of the publisher network element.

8. The network element interaction method according to claim 7, characterized in that: Training a machine learning model corresponding to the subscription content on the subscriber network element according to the intermediate model information of the subscriber network element and the intermediate model information of the publisher network element, including: Selecting a third-party network element that has established a trusted computing relationship with both the subscriber network element and the publisher network element; Sending the model intermediate information of the subscriber network element and the model intermediate information of the publisher network element to the third-party network element respectively; Aggregating the model intermediate information of the subscriber network element and the model intermediate information of the publisher network element on the third-party network element to obtain global information; According to the global information, model parameters of the machine learning model corresponding to the subscription content are iteratively updated on the subscriber network element and the publisher network element respectively.

9. The network element interaction method according to any one of claims 1 to 4, characterized in that: The subscribing network element is a network element outside the core network of the mobile communication network, and the publishing network element is a network element inside the core network of the mobile communication network.

10. The network element interaction method according to claim 9, characterized in that: The subscriber network element is an application function network element for implementing application layer services, and the publisher network element is a network data analysis function network element for providing network data analysis services to the mobile communication network.

11. The network element interaction method according to claim 10, characterized in that: Before selecting publisher sample data on the publisher network element according to the identification information of the intersection data, the method further includes: Initiating a sample collection request to an operation, maintenance and management network element in the mobile communication network through the network data analysis function network element; The original data held by the network data analysis function network element is obtained according to the sample collection request.

12. The network element interaction method according to claim 11, characterized in that: The original data held by the network data analysis function network element includes channel quality indication information used to indicate the quality of the wireless channel.

13. The network element interaction method according to any one of claims 1 to 4, characterized in that: The subscriber network element and the publisher network element train a machine learning model corresponding to the subscription content through federated learning.

14. The network element interaction method according to claim 13, characterized in that: The subscribing network element and the publishing network element train a machine learning model corresponding to the subscription content through vertical federated learning; the training samples held by the subscribing network element and the training samples held by the publishing network element have the same sample identifier and different characteristics.

15. A network element interaction device, characterized in that: include: a network element selection module configured to select, in a mobile communication network, a publisher network element that establishes an information subscription relationship with a subscriber network element, wherein the information subscription relationship is used to indicate that the publisher network element provides data analysis information related to original data of the subscription content to the subscriber network element; a data alignment module configured to perform data alignment between the subscriber network element and the publisher network element to obtain identification information of intersection data, wherein the intersection data is data having the same identification information between the original data held by the subscriber and the original data held by the publisher network element; a sample selection module configured to select publisher sample data on the publisher network element according to the identification information of the intersection data, and train a machine learning model corresponding to the subscription content according to the publisher sample data to obtain intermediate model information of the publisher network element; The model training module is configured to send the model intermediate information of the publisher network element to the subscriber network element, and train a machine learning model corresponding to the subscription content on the subscriber network element based on the model intermediate information of the publisher network element.

16. A computer-readable medium, characterized in that The computer-readable medium stores a computer program, and when the computer program is executed by a processor, the network element interaction method according to any one of claims 1 to 14 is implemented.

17. An electronic device, characterized in that: include: processor; as well as a memory for storing executable instructions of the processor; The processor is configured to execute the executable instructions to implement the network element interaction method as described in any one of claims 1 to 14.

18. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the network element interaction method according to any one of claims 1 to 14 is implemented.

Images