Data transmission method and apparatus, and medium, electronic device and program product

Abstract

The present application belongs to the technical field of networks and artificial intelligence, and specifically relates to a data transmission method, a data transmission apparatus, a computer-readable medium, an electronic device and a computer program product. The method comprises: on the basis of a first network quality parameter held by a network-side server, determining first intermediate data, wherein the network-side server is used for providing a network communication service to a terminal device; on the basis of a second network quality parameter held by a terminal-side server, determining second intermediate data, wherein the terminal-side server is used for providing an application function service to the terminal device; and performing a merging operation on the first intermediate data and the second intermediate data on a first edge-side server, so as to obtain a first data discarding policy, wherein the first edge-side server is used for providing an edge computing service to the terminal device, and the first data discarding policy is used for instructing the terminal device to transmit full data or discard part of the full data. The present application can improve the data transmission efficiency and the data transmission quality of network communications.

Description

Data transmission methods, devices, media, electronic equipment and software products

[0001] This application claims priority to Chinese Patent Application No. 202411589901.9, filed on November 7, 2024, entitled "Data Transmission Method, Apparatus, Medium, Electronic Equipment and Program Product". Technical Field

[0002] This application belongs to the field of network and artificial intelligence technology, and specifically relates to a data transmission method, a data transmission device, a computer-readable medium, an electronic device, and a computer program product. Background Technology

[0003] In mobile communication networks, for reasons of data privacy and security, various functional entities usually do not disclose the system data they hold to external parties, or they need to go through a cumbersome information encryption and verification process before they can share the corresponding system data with the designated functional entities.

[0004] Because it is difficult for different functional entities in a mobile communication network to share system data, it is impossible to accurately evaluate the network communication quality of different functional entities, which easily leads to problems such as low data transmission efficiency and poor data transmission quality in the network communication process. Summary of the Invention

[0005] This application provides a data transmission method, a data transmission device, a computer-readable medium, an electronic device, and a computer program product, with the aim of improving the data transmission efficiency and quality of network communication.

[0006] According to one aspect of the embodiments of this application, a data transmission method is provided, the method comprising: determining first intermediate data based on a first network quality parameter held by a network-side server, the network-side server being used to provide network communication services to a terminal device, the first intermediate data being used to represent a first data transmission state of the network communication service; determining second intermediate data based on a second network quality parameter held by a terminal-side server, the terminal-side server being used to provide application function services to the terminal device, the second intermediate data being used to represent a second data transmission state of the application function services; and performing a merging operation on the first intermediate data and the second intermediate data on a first edge-side server to obtain a first data discarding strategy, the first edge-side server being used to provide edge computing services to the terminal device, the first data discarding strategy being used to instruct the terminal device to transmit all data or discard a portion of the all data.

[0007] According to one aspect of the embodiments of this application, a data transmission apparatus is provided, the apparatus comprising:

[0008] The first determining module is configured to determine first intermediate data based on a first network quality parameter held by a network-side server, wherein the network-side server is used to provide network communication services to terminal devices, and the first intermediate data is used to represent the first data transmission status of the network communication service.

[0009] The second determining module is configured to determine second intermediate data based on a second network quality parameter held by a terminal-side server. The terminal-side server is used to provide application function services to the terminal device. The second intermediate data is used to represent the second data transmission status of the application function service.

[0010] The merging module is configured to perform a merging operation on the first intermediate data and the second intermediate data on a first edge-side server to obtain a first data discarding policy. The first edge-side server is used to provide edge computing services to the terminal device. The first data discarding policy is used to instruct the terminal device to transmit all data or discard part of the data in the full data.

[0011] In some embodiments of this application, based on the above technical solutions, the network-side server includes a network data analysis function network element in a mobile communication network; the first determining module includes:

[0012] The first acquisition module is configured to send a data acquisition request to the operation, maintenance and management network element in the mobile communication network.

[0013] The first receiving module is configured to receive response data returned by the operation and maintenance management network element in response to the data collection request. The response data includes the first network quality parameters related to the access network and terminal equipment collected by the operation and maintenance management network element.

[0014] The first inference module is configured to perform first model inference based on the first network quality parameters according to the network-side model configured on the network data analysis function network element, and obtain the first intermediate data.

[0015] In some embodiments of this application, based on the above technical solutions, the first determining module further includes:

[0016] The first training module is configured to train the terminal-side model based on the first data samples held by the terminal-side server to obtain the first intermediate model parameters. The terminal-side model is a first machine learning model configured on the terminal-side server for predicting the second data discarding strategy.

[0017] The first update module is configured to iteratively update the parameters of the first intermediate model based on the second data samples held by the network data analysis function network element to obtain the network-side model.

[0018] In some embodiments of this application, based on the above technical solutions, the terminal-side server includes application function network elements in a mobile communication network; the second determining module includes:

[0019] The second acquisition module is configured to acquire data from the terminal device in the mobile communication network to obtain the second network quality parameters related to the terminal device.

[0020] The second inference module is configured to perform second model inference based on the network quality parameters according to the terminal-side model configured on the application function network element, and obtain the second intermediate data.

[0021] In some embodiments of this application, based on the above technical solutions, the second determining module further includes:

[0022] The second training module is configured to train the network-side model based on the second data samples held by the network-side server to obtain the second intermediate model parameters. The network-side model is a second machine learning model configured on the network-side server for predicting the third data discarding strategy.

[0023] The second update module is configured to iteratively update the parameters of the second intermediate model based on the first data sample held by the application function network element to obtain the terminal-side model.

[0024] In some embodiments of this application, based on the above technical solutions, the merging module includes:

[0025] The data merging module is configured to perform a merging operation on the first intermediate data and the second intermediate data on the first edge-side server to obtain global intermediate data.

[0026] The data inference module is configured to perform third model inference based on the global intermediate data according to the edge-side model configured on the first edge-side server, and obtain the first data discarding strategy.

[0027] In some embodiments of this application, based on the above technical solutions, the merging module further includes:

[0028] The data alignment module is configured to perform data alignment between the network-side server and the terminal-side server to obtain the identification information of the intersection data. The intersection data is data with the same identification information between the first original data held by the network-side server and the second original data held by the terminal-side server.

[0029] The network-side training module is configured to select network-side data samples on the network-side server based on the identification information of the intersection data, and train a network-side model for predicting the fourth data discarding strategy based on the network-side data samples, thereby obtaining the third intermediate model parameters of the network-side model.

[0030] The terminal-side training module is configured to select terminal-side data samples on the terminal-side server based on the identification information of the intersection data, and train a terminal-side model for predicting the fifth data discarding strategy based on the terminal-side data samples, thereby obtaining the fourth intermediate model parameters of the terminal-side model.

[0031] The edge-side training module is configured to iteratively update the third intermediate model parameters of the network-side model and the fourth intermediate model parameters of the terminal-side model based on the data samples held by the first edge-side server, so as to obtain the edge-side model.

[0032] In some embodiments of this application, based on the above technical solutions, the data alignment module includes:

[0033] The network element selection module is configured to select a second edge-side server that has established a trusted computing relationship with both the network-side server and the terminal-side server. The trusted computing relationship indicates that the second edge-side server can perform trusted computing on the first original data held by the network-side server and the second original data held by the terminal-side server while maintaining data privacy.

[0034] The original identifier sending module is configured to send the identifier information of the first original data held by the network-side server and the identifier information of the second original data held by the terminal-side server to the second edge-side server, respectively.

[0035] The identifier alignment module is configured to perform data alignment on the second edge-side server on the identifier information of the first original data held by the network-side server and the identifier information of the second original data held by the terminal-side server to obtain the identifier information of the intersection data.

[0036] The intersection identifier sending module is configured to have the second edge-side server send the identifier information of the intersection data to the network-side server and the terminal-side server respectively.

[0037] In some embodiments of this application, based on the above technical solutions, the network element selection module is further configured to: obtain a first set of network elements that establish a trusted computing relationship with the network-side server and a second set of network elements that establish a trusted computing relationship with the terminal-side server; if there are shared network elements between the first set of network elements and the second set of network elements, then the shared network element is selected as the second edge-side server; if there are no shared network elements between the first set of network elements and the second set of network elements, then a temporary network element is selected as the second edge-side server through the network open function network element in the mobile communication network.

[0038] In some embodiments of this application, based on the above technical solutions, the temporary network element is a network element selected from the third network element set maintained by the network open function network element, which is different from the first network element set and different from the second network element set. The temporary network element establishes a trusted computing relationship with the network-side server and the terminal-side server with a specified effective duration through the network open function network element.

[0039] In some embodiments of this application, based on the above technical solutions, the apparatus further includes:

[0040] The collaboration request module is configured such that the terminal-side server and the first edge-side server respectively send a first collaboration request and a second collaboration request to the network-side server for performing distributed computing.

[0041] The first establishment module is configured to establish a first collaborative relationship between the network-side server and the terminal-side server for performing the distributed computing, based on a first response message returned by the network-side server to the terminal-side server in response to the first collaborative request.

[0042] The second establishment module is configured to establish a second collaborative relationship between the network-side server and the first edge-side server for performing the distributed computing, based on a second response message returned by the network-side server to the first edge-side server in response to the second collaborative request.

[0043] In some embodiments of this application, based on the above technical solutions, the collaboration request module is further configured as follows: the terminal-side server and the first edge-side server respectively send a first information opening request and a second information opening request to the network open function network element in the mobile communication network, wherein the first information opening request carries a first network element identifier for identifying the terminal-side server, and the second information opening request carries a second network element identifier for identifying the first edge-side server; on the network open function network element, according to the first network element identifier and the second network element identifier, the first collaboration request and the second collaboration request for performing the distributed computing are sent to the network-side server in the mobile communication network.

[0044] In some embodiments of this application, based on the above technical solutions, the network-side server and the terminal-side server train the network-side model configured on the network-side server and the terminal-side model configured on the terminal-side server through vertical federated learning; the training samples held by the network-side server and the training samples held by the terminal-side server have the same sample identifier and different sample content.

[0045] According to one aspect of the embodiments of this application, a computer-readable medium is provided, on which a computer program is stored, which, when executed by a processor, implements the data transmission method as described in the above technical solutions.

[0046] According to one aspect of the embodiments of this application, an electronic device is provided, the electronic device 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 data transmission method as described in the above technical solutions.

[0047] According to one aspect of the embodiments of this application, a computer program product is provided, including a computer program that, when executed by a processor, implements the data transmission method as described in the above technical solutions.

[0048] In the technical solution provided in this application embodiment, first intermediate data is determined based on a first network quality parameter held by the network-side server, and second intermediate data is determined based on the network quality parameter held by the terminal-side server. Then, the first and second intermediate data are merged and calculated on the edge-side server to obtain a data discarding policy. Based on the data discarding policy, the terminal device is instructed to transmit all data or discard a portion of the full data. By determining the first and second intermediate data representing the first and second data transmission states on the network-side server and the terminal-side server respectively, an accurate evaluation of network communication quality can be completed without sharing the original data. This allows the edge-side server to specify a data discarding policy for the terminal device, enabling the terminal device to transmit all data when network quality is good and discard a portion of the full data when network quality is poor. Therefore, data transmission efficiency and quality during network communication can be improved.

[0049] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0050] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application. It is obvious that the drawings described below are merely some embodiments of this application, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort.

[0051] Figure 1 schematically illustrates the overall network architecture of a mobile communication network applying the technical solution of this application.

[0052] Figure 2 schematically illustrates the specific network element structure of a mobile communication network applicable to embodiments of this application.

[0053] Figure 3 shows a flowchart of a data transmission method in various embodiments of this application.

[0054] Figure 4 illustrates a flowchart of how the network-side server determines the first intermediate data in various embodiments of this application.

[0055] Figure 5 illustrates a flowchart of how the terminal-side server determines the second intermediate data in various embodiments of this application.

[0056] Figure 6 shows a flowchart of the merging operation of intermediate data by the edge-side server in various embodiments of this application.

[0057] Figure 7 shows a flowchart of joint training of the edge-side model in various embodiments of this application.

[0058] Figure 8 illustrates a distributed collaborative computing method for data transmission in an application scenario according to an embodiment of this application.

[0059] Figure 9 schematically shows a structural block diagram of the data transmission device provided in an embodiment of this application.

[0060] Figure 10 schematically illustrates a computer system architecture block diagram suitable for implementing electronic devices according to embodiments of the present application. Detailed Implementation

[0061] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary 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 to make this application more comprehensive and complete, and to fully convey the concept of the exemplary embodiments to those skilled in the art.

[0062] Furthermore, the described features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. Numerous specific details are provided in the following description to give a thorough understanding of embodiments of this application. However, those skilled in the art will recognize that the technical solutions of this application can be practiced without one or more of the specific details, or other methods, components, apparatuses, steps, etc., can be employed. In other instances, well-known methods, apparatuses, implementations, or operations are not shown or described in detail to avoid obscuring various aspects of this application.

[0063] In this application embodiment, the terms "module" or "unit" refer to a computer program or part of a computer program that has a predetermined function and works with other related parts to achieve a predetermined goal, and can be implemented wholly or partially using software, hardware (such as processing circuitry or memory), or a combination thereof. Similarly, a processor (or multiple processors or memory) can be used to implement one or more modules or units. Furthermore, each module or unit can be part of an overall module or unit that includes the functionality of that module or unit.

[0064] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor devices and / or microcontroller devices.

[0065] The flowcharts shown in the accompanying drawings are merely illustrative and do not necessarily include all content and operations / steps, nor do they necessarily have to be performed in the described order. For example, some operations / steps can be broken down, while others can be combined or partially combined; therefore, the actual execution order may change depending on the specific circumstances.

[0066] The relevant terms related to mobile communication networks involved in the embodiments of this application are explained as follows.

[0067] AF stands for Application Function, and can also be considered as AS (Application Server). This entity can implement the control plane functions of a third-party application server, interacting via AF-NEF-PCF or AF-PCF. This entity can also implement the user plane functions of a third-party application server, for example, interacting via the AS-IP transport network-UPF interface.

[0068] SMF: Session Management function, responsible for tunnel maintenance, IP address allocation and management, UPF selection, policy enforcement and QoS control, billing data collection and / or roaming, etc.

[0069] 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, and provides authentication and authorization functions for user access. It is the core network control plane access point for terminals and the radio.

[0070] PCF: Policy Control function, which has a unified policy framework and provides policy rules for control plane functions.

[0071] UPF: User plane function, which performs packet routing and forwarding, policy enforcement, traffic reporting and / or QoS processing.

[0072] NEF (Network Exposure Function) is located between the 5G core network and external third-party application functions (and possibly some internal AFs). It is responsible for managing the network data exposed to the outside world. All external applications that want to access data inside the 5G core network must go through NEF. NEF provides corresponding security guarantees to ensure the security of external applications to the 3GPP network, and provides functions such as opening up QoS customization capabilities for external applications, subscription to mobility state events, and distribution of AF requests.

[0073] UE: user equipment; terminal equipment may include one or more of the following: mobile terminal mobile phone APP, vehicle-mounted IP camera, and other terminal software.

[0074] NWDAF: Network Data Analytics Function. Provides specific network data analysis services to the network.

[0075] ADRF: Analytics Data Repository Function. This functional entity was introduced to implement network analysis functions, such as saving network analysis data and saving collected data.

[0076] MFAF: Messaging Framework Adaptor Function. This NF enables adaptation to the messaging framework, a function not defined in 3GPP. MFAF facilitates interaction between the DCCF and the messaging framework, sending data to be processed and receiving processed data, performing formatting and other related processing. This NF was also introduced to enable network data analysis functions.

[0077] AnLF: Analytics Logical Function. It is responsible for model inference and provides general NWDAF service interfaces such as Nnwdaf_AnalyticsSubscription and Nnwdaf_AnalyticsInfo. It can generate analysis results (including static statistical data and dynamic inference results) based on requests from consumer network elements.

[0078] MTLF: Model Training Logical Function. It is responsible for model training and can provide a pre-trained model to AnLF. The model training process here is non-standardized (meaning manufacturers have considerable flexibility). AnLF is the only consumer network element provided with services by MTLF. A single NWDAF network element can integrate both AnLF and MTLF.

[0079] Figure 1 schematically illustrates the overall network architecture of a mobile communication network applying the technical solution of this application.

[0080] As shown in Figure 1, the mobile communication network may include an access network 101, a bearer network 102, and a core network 103. The access network 101 enables users' mobile phones and other terminal devices 104 to access the network. The terminal devices 104 connect to the base station via an air interface. The bearer network 102 carries data transmission, for example, transmitting data between network elements via optical fiber. The core network 103 is the management center of the mobile communication network, and is a carrier-grade router used for data management such as location management, updates, authentication, and connection management of the terminal devices 104.

[0081] The technical solutions of this application embodiment 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), Worldwide Interoperability for Microwave Access (WiMAX) system, 5th Generation (5G) system, 6th Generation (6G) system, New Radio (NR) system, etc.

[0082] Figure 2 schematically illustrates the specific network element structure of a mobile communication network applicable to embodiments of this application. The various components involved in this mobile communication network will be described below.

[0083] User equipment (UE) 201: may include various handheld terminal devices, vehicle-mounted devices, and smart wearable devices with wireless communication capabilities, such as cellular phones, smartphones, wireless data cards, personal digital assistants (PDAs), laptops, tablets, 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.

[0084] Access network (AN) element 202: This element is responsible for air interface-side radio resource management, Quality of Service (QoS) management, data compression and encryption, etc. It provides network access functionality to authorized user equipment in specific areas and can use transmission tunnels of different qualities based on the user equipment's level and service requirements. In wireless access scenarios, access network element 202 can also be called a radio access network (RAN) element. RAN elements manage radio resources, provide access services to user equipment, and forward control signals and user equipment data between the user equipment and the core network. RAN elements can include various types of base stations, such as macro base stations, micro base stations (also called small stations), relay stations, access points, etc. The name of the RAN element with base station functionality may differ in systems employing different wireless access technologies. For example, in fifth-generation (5G) systems... th In 5G systems, it is called gNB; in LTE systems, it is called evolved NodeB (eNB or eNodeB); in 3G systems, it is called Node B, etc.

[0085] User plane network element 203: This element is responsible for user packet encapsulation, routing, forwarding, statistics, and quality of service (QoS) processing of user plane data. In 5G communication systems, this user plane network element can be a user plane function (UPF) element.

[0086] Data network element 204: A network used to provide data transmission. In a 5G communication system, this data network element can be a data network (DN) element.

[0087] Access and mobility management network element 205: This element can be used for mobility management and access management, and can implement functions other than session management within the mobility management entity (MME) functionality, such as legitimacy detection and access authorization / authentication. In 5G communication systems, this access and mobility management network element can be an access and mobility management function (AMF) network element.

[0088] Session management network element 206: This can be used for session management, allocation and management of user equipment network protocol (IP) addresses, selection of manageable user plane functions, policy control and charging function interface endpoints, and downlink data notification. In 5G communication systems, this session management network element can be a session management function (SMF) network element.

[0089] Policy control network element 207: A unified policy framework used to guide network behavior, providing policy rule information to control plane function network elements (such as AMF, SMF, etc.). 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.

[0090] Authentication Server 208: Used for authentication services, generating keys, and implementing two-way authentication of user equipment, supporting a unified authentication framework. In 5G communication systems, this authentication server can be an authentication server function (AUSF) network element.

[0091] Data management network element 209: Used to handle 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.

[0092] Application network element 210: Used for data routing affected by applications, accessing network open function network elements, and interacting with the policy framework for policy control, etc. In 5G communication systems, this application network element can be an application function (AF) network element.

[0093] Network storage element (not shown in the diagram): Used to maintain real-time information for all network function services in the network. In 5G communication systems, this network storage element can be a network repository function (NRF) element.

[0094] In the network architecture shown in Figure 2, User Equipment (UE) 201 is connected to AMF 205 via interface N1, (R)AN 202 is connected to AMF 205 via interface N2, and (R)AN 202 is connected to UPF 203 via interface N3. UPFs 203 are connected to each other via interface N9, and UPF 203 is interconnected with DN 204 via interface N6. SMF 206 controls UPF 203 via interface N4. AMF 205 is connected to SMF 206 via interface N11. AMF 205 obtains UE subscription data from UDM 209 via interface N8, and SMF 206 obtains UE subscription data from UDM 209 via interface N10.

[0095] It is understood that the aforementioned network elements or functions can be network components in hardware devices, software functions running on dedicated hardware, or virtualization functions instantiated on a platform (e.g., a cloud platform). The network architecture applied to the embodiments of this application is merely an example, and the network architecture applicable to the embodiments of this application is not limited thereto. Any network architecture capable of implementing the functions of the aforementioned network elements is applicable to the embodiments of this application.

[0096] For example, in some network architectures, network function elements such as AMF, SMF, PCF, AUSF, and UDM are all called network function (NF) elements; or, in other network architectures, a collection of network elements such as AMF, SMF, PCF, AUSF, and UDM can be called control plane function elements.

[0097] The interface names between the network elements in Figures 1 and 2 are merely examples; in actual implementations, the interface names may differ, and this application does not impose any specific limitations on them. The names of the network elements (e.g., SMF, AF, UPF, etc.) included in Figures 1 and 2 are also merely examples and do not limit the functionality of the network elements themselves. In 5GS and future networks, the aforementioned network elements may also have other names, and this application does not impose any specific limitations on them. For example, in 6G networks, some or all of the aforementioned network elements may use the terminology from 5G networks, or they may use other names. Furthermore, it should be understood that the names of the messages (or signaling) transmitted between the aforementioned network elements are also merely examples and do not limit the functionality of the messages themselves.

[0098] The following detailed description, in conjunction with specific embodiments, provides a detailed explanation of the technical solutions provided in this application, including data transmission methods, data transmission devices, computer-readable media, electronic devices, and computer program products.

[0099] Figure 3 shows a flowchart of a data transmission method in various embodiments of this application. This data transmission method can be executed by one network element in the mobile communication network shown in Figure 2, or it can be executed jointly by two or more network elements. As shown in Figure 3, the data transmission method includes the following steps S310 to S330.

[0100] S310: Determine first intermediate data based on the first network quality parameters held by the network-side server. The network-side server is used to provide network communication services to the terminal device. The first intermediate data is used to represent the first data transmission status of the network communication service.

[0101] The network-side server may include, for example, the Network Data Analysis Function (NWDAF) element shown in Figure 2. This element is located within the core network of the mobile communication network and can directly obtain raw data from within the core network. It calculates a first network quality parameter based on the raw data and then determines first intermediate data based on the first network quality parameter. In various embodiments of this application, "parameter" can refer to a measurable attribute (the parameter itself), a parameter value, or both.

[0102] Network quality parameters include one or more of the following: signal strength, latency, jitter, and packet loss rate. Some of these parameters are related to the terminal device, while others are end-to-end parameters related to network transmission. Network quality parameters related to the terminal device are a subset of these, such as signal strength. "Network quality parameters" can also be called other names, such as "network performance parameters" or "network and terminal device performance parameters."

[0103] Network quality parameters may include at least one of the following: Reference Signal Receiving Power (RSRP), Received Signal Strength Indication (RSSI), Reference Signal Receiving Quality (RSRQ), and Signal to Interference plus Noise Ratio (SINR). In various embodiments of this application, the first network quality parameter may be similar to the second network quality parameter, including at least one of the above parameters, or the same as it.

[0104] RSRP refers to the linear average power contribution of resource particles carrying the cell-specific reference signal in a specified measurement frequency band, which can be understood as the power of each subcarrier. RSSI refers to the total power within the receive bandwidth, including the useful signal, interference, and noise floor. RSRQ is defined as the ratio of RSRP to RSSI, for example, RSRQ = N * RSRP / (E-UTRA carrier RSSI), where N is the number of RBs in the E-UTRA carrier RSSI measurement bandwidth. RSRQ reflects the signal-to-noise ratio and interference level of the signal quality.

[0105] SINR represents the ratio of the strength of the received useful signal to the strength of the received interference signal (noise and interference). It reflects the link quality of the current channel and is an important indicator for measuring UE performance parameters.

[0106] The first intermediate data is used to represent the first data transmission status generated by the network-side server when providing network communication services to the terminal device. The first data transmission status may be intermediate data obtained by performing first model inference based on the first network quality parameters and a pre-trained first machine learning model.

[0107] The first intermediate data obtained through the first model inference process reflects the first data transmission status of the network communication service without exposing the original first network quality parameters. While ensuring the privacy of data within the core network is not compromised, the first intermediate data can be transmitted as derived data to the outside of the core network to meet the application requirements of external network elements regarding the first data transmission status of the network communication service.

[0108] S320: Determine the second intermediate data based on the second network quality parameters held by the terminal-side server. The terminal-side server is used to provide application function services to the terminal device. The second intermediate data is used to represent the second data transmission status of the application function services.

[0109] The terminal-side server may include, for example, the application function network element AF shown in Figure 2, specifically used to provide application function services to the terminal device. This network element is an external network element of the core network in the mobile communication network. During the process of providing application function services to the terminal device, it can collect relevant application data, calculate a second network quality parameter based on the application data, and then determine second intermediate data based on the second network quality parameter. The second intermediate data represents the second data transmission state generated by the terminal-side server when providing application function services to the terminal device. This second data transmission state can be intermediate data obtained through second model inference using a pre-trained second machine learning model based on the second network quality parameter. In the absence of original data from within the core network, the second intermediate data is used to meet the application requirements for the second data transmission state of the application function service.

[0110] For example, various types of application software can be installed on terminal devices. These applications can communicate with corresponding terminal-side servers to execute various application functions. The application software installed on the terminal device can include client programs, browser programs, or hosting programs. Client programs correspond to server programs; they can send requests to the application function server and receive response data from it. Browser programs are used to retrieve, present, and browse information resources transmitted over the internet. They can access navigation links provided by the application function server and interact with it via web pages. Hosted programs utilize a host program to provide the runtime environment and can perform corresponding program functions on the terminal device without requiring a client installation.

[0111] The terminal-side server communicates with the application software installed on the terminal device to provide corresponding application functions and services, such as high-definition video playback in the audio and video field, model training and inference in the large model field, and extended reality scenes in the metaverse field. The aforementioned second intermediate data is used to represent the data transmission status of at least one of these application functions and services.

[0112] S330: The first edge-side server performs a merging operation on the first intermediate data and the second intermediate data to obtain a first data discarding policy. The first edge-side server is used to provide edge computing services to the terminal device. The first data discarding policy is used to instruct the terminal device to transmit all data or discard part of the data in the full data.

[0113] The first edge-side server may include, for example, the application function network element (AF) shown in Figure 2, specifically used to provide edge computing services to terminal devices. In one application scenario, the mobile communication network includes at least two application function network elements, one of which acts as a terminal-side server to provide application function services to terminal devices, and the other application function network element acts as an edge-side server to provide edge computing services to terminal devices.

[0114] Edge computing is a distributed computing paradigm that brings computing, storage, and network services closer to the data source or user (the first edge-side server) to reduce latency, improve response speed, and optimize bandwidth usage. Edge computing provides the nearest-end service to end devices by using an open platform that integrates network, computing, storage, and application capabilities on the side closest to the terminal device. Edge computing allows applications to initiate at the edge, resulting in faster network service responses and meeting the industry's basic needs in areas such as real-time business, application intelligence, security, and privacy protection.

[0115] By merging the first intermediate data and the second intermediate data on the first edge server, a first data discarding strategy is achieved that integrates external core network data and internal core network data while ensuring data privacy and security, and can be adaptively adjusted according to the overall data transmission status of the mobile communication network.

[0116] In the data transmission method provided in this application embodiment, by determining first and second intermediate data representing the first and second data transmission states on the network-side server and the terminal-side server respectively, a more accurate evaluation of network communication quality can be achieved without sharing the original data. This allows for specifying a first data discarding policy for the terminal device on the first edge-side server, enabling the terminal device to transmit all data when network quality is good and discard a portion of the full data when network quality is poor. Therefore, this improves data transmission efficiency and quality during network communication, ensuring QoS service requirements. "Full data" can refer to all data packets that the terminal device needs to transmit in a single transmission.

[0117] Figure 4 illustrates a flowchart of the process by which a network-side server determines the first intermediate data in various embodiments of this application. As shown in Figure 4, the network-side server in this embodiment includes a network data analysis function network element in a mobile communication network. Step S310, which determines the first intermediate data based on the first network quality parameters held by the network-side server, may include the following steps S311 to S313.

[0118] S311: Send a data collection request to the operation, maintenance and management network element in the mobile communication network.

[0119] S312: Receive response data from the operation and maintenance management network element in response to the data collection request. The response data includes first network quality parameters related to the access network and terminal equipment collected by the operation and maintenance management network element.

[0120] S313: Based on the network-side model configured on the network data analysis function network element, perform first model inference based on the first network quality parameters to obtain the first intermediate data.

[0121] In this embodiment of the application, the Network Data Analysis Function (NWDAF) can send a data collection request to the Operation and Maintenance Management (OAM) network element. The OAM collects a first network quality parameter related to the Access Network (RAN) and the Terminal Equipment (UE) based on the received data collection request, and then returns response data containing the first network quality parameter to the NWDAF.

[0122] The Network Data Analysis Function (NWDAF) element is configured with a pre-trained network-side model. Using this model, model inference is performed based on the first network quality parameters carried in the response data to obtain the first intermediate data. The network-side model is a first machine learning model configured on the network-side server to predict the second data discarding strategy.

[0123] The first intermediate data is used to represent the data transmission status of the network communication service, and may include predicted network quality parameter values ​​for a future period after the current time. The data transmission status may also include other quantitative indicators, such as light-load data transmission status, heavy-load data transmission status, etc.

[0124] In various embodiments of this application, before performing model inference based on the first network quality parameters, the terminal-side model can be trained according to the first data samples held by the terminal-side server to obtain the first intermediate model parameters. The terminal-side model is a second machine learning model configured on the terminal-side server for predicting the third data discarding strategy. Then, the first intermediate model parameters are iteratively updated according to the second data samples held by the network data analysis function network element to obtain the network-side model.

[0125] The first intermediate model parameters may include model parameters obtained during the training of the terminal-side model, and may also include gradient information used to update the model parameters.

[0126] This application embodiment can utilize the first intermediate model parameters trained on the terminal-side server as initial network-side model parameters or a reference for the network-side model, and continue model training on the network-side server to obtain the network-side model. In this way, the terminal-side server does not need to share raw data with the network-side server. While ensuring data privacy and security, the model training results of the terminal-side server are synchronized to the network-side server, thereby improving the training efficiency of the network-side model. Furthermore, the network-side model incorporates some of the model training results from the terminal-side server, which can improve the reliability of the network-side model.

[0127] Figure 5 illustrates a flowchart of the process by which a terminal-side server determines the second intermediate data in an embodiment of this application. As shown in Figure 5, the terminal-side server in this embodiment includes application function network elements in a mobile communication network. Step S320, which determines the second intermediate data based on the second network quality parameters held by the terminal-side server, may include the following steps S321 to S322.

[0128] S321: Collect data from terminal devices in a mobile communication network to obtain second network quality parameters related to the terminal devices.

[0129] S322: Based on the terminal-side model configured on the application function network element, perform second model inference based on the second network quality parameters to obtain the second intermediate data.

[0130] In this embodiment, the application function network element (AF), acting as a terminal-side server, can collect data from terminal devices in the mobile communication network to obtain second network quality parameters related to the terminal devices. The collected second network quality parameters may be of the same type or partially the same as the first network quality parameters, and may include radio interface-related parameters such as signal strength.

[0131] In various embodiments of this application, a pre-trained terminal-side model is configured on the application function network element (AF). Using this terminal-side model, second model inference is performed based on second network quality parameters related to the terminal device to obtain second intermediate data. The terminal-side model is a second machine learning model configured on the terminal-side server for predicting a third data discarding strategy.

[0132] In various embodiments of this application, before performing the second model inference based on the second network quality parameters, the network-side model is trained according to the second data samples held by the network-side server to obtain the second intermediate model parameters. The network-side model is a second machine learning model configured on the network-side server for predicting the fourth data discarding strategy. Then, the second intermediate model parameters are iteratively updated according to the first data samples held by the application function network element to obtain the terminal-side model.

[0133] This application embodiment can utilize the second intermediate model parameters trained on the network-side server as initial terminal-side model parameters or a reference for the terminal-side model, and continue model training on the terminal-side server to obtain the terminal-side model. In this way, the network-side server does not need to share raw data with the terminal-side server. While ensuring data privacy and security, the model training results from the network-side server are synchronized to the terminal-side server, thereby improving the training efficiency of the terminal-side model. Furthermore, since the terminal-side model incorporates some of the model training results from the network-side server, its reliability can be improved.

[0134] Figure 6 illustrates a flowchart of the merging operation of intermediate data by an edge-side server in various embodiments of this application. This edge-side server may be referred to as the first edge-side server. As shown in Figure 6, step S330, where the first intermediate data and the second intermediate data are merged on the first edge-side server to obtain a first data discarding strategy, may include the following steps S331 to S332.

[0135] S331: On the first edge server, merge the first intermediate data and the second intermediate data to obtain global intermediate data;

[0136] S332: Based on the edge-side model configured on the first edge-side server, perform third-model inference based on global intermediate data to obtain the first data discarding strategy.

[0137] In the embodiments of this application, the network-side server and the terminal-side server can send the first intermediate data and the second intermediate data obtained by their respective model inference to the first edge-side server, and then the first edge-side server performs a merging operation on the first intermediate data and the second intermediate data to obtain global intermediate data.

[0138] In embodiments of this application, the first intermediate data can be denoted as a first vector, and the second intermediate data can be denoted as a second vector. The method by which the first edge-side server performs merging operations on the intermediate data may include concatenating the first vector and the second vector to obtain a merged vector, or it may also include calculating the average of the first vector and the second vector to obtain a mean vector. Alternatively, the first vector and the second vector may be merged in other ways, such as multiplying, adding, or weighted adding the two vectors.

[0139] The first edge server uses a pre-trained edge model to perform model inference based on global intermediate data obtained from merging operations, and obtains the first data discarding strategy of the model output.

[0140] In the embodiments of this application, the edge-side model can be jointly trained with the network-side model and the terminal-side model.

[0141] Figure 7 illustrates a flowchart of joint training of the edge-side model in various embodiments of this application. As shown in Figure 7, the method for joint training of the edge-side model, the network-side model, and the terminal-side model may include one or more of the following steps S710 to S740.

[0142] S710: Perform data alignment between the network-side server and the terminal-side server to obtain the identification information of the intersection data. The intersection data is data with the same identification information between the first original data held by the network-side server and the second original data held by the terminal-side server.

[0143] Intersecting data can be used on the network-side server and the terminal-side server to train the network-side model and the terminal-side model, respectively, to obtain the trained model parameters (which can be referred to as the third and fourth intermediate model parameters, respectively), and then provided to the first edge server. Based on these trained model parameters, the first edge server iteratively updates them using data samples to obtain the edge-side model. This process includes, for example, the following steps.

[0144] S720: Select network-side data samples on the network-side server based on the identification information of the intersection data, and train the network-side model for predicting the fourth data discarding strategy based on the network-side data samples to obtain the third intermediate model parameters of the network-side model.

[0145] S730: Select terminal-side data samples on the terminal-side server based on the identification information of the intersection data, and train the terminal-side model for predicting the fifth data discarding strategy based on the terminal-side data samples to obtain the fourth intermediate model parameters of the terminal-side model.

[0146] S740: Based on the data samples held by the first edge-side server, the third intermediate model parameters of the network-side model and the fourth intermediate model parameters of the terminal-side model are iteratively updated to obtain the edge-side model.

[0147] In embodiments of this application, the method for data alignment between subscriber network elements and publisher network elements (network-side server and terminal-side server) may include privacy set intersection, that is, determining the intersection data between their original data through privacy computation without transmitting their original data to each other. Privacy set intersection methods may include, for example, Blind RSA-based PSI Protocol with linear complexity, Diffie-Hellman-based schemes, oblivious transfer (OT)-based schemes, Freedman secure intersection protocol, etc.

[0148] In embodiments of this application, to further improve the reliability of data privacy and security, data alignment can be performed between the network-side server and the terminal-side server by establishing a second edge-side server with a trusted computing relationship. For example, data alignment can be performed based on the identification information of the first original data held by the network-side server and the identification information of the second original data held by the terminal-side server to obtain the identification information of the intersection data. This second edge-side server may be the same as or different from the first edge-side server.

[0149] The method for data alignment between the network-side server and the terminal-side server may include: selecting a second edge-side server that has established a trusted computing relationship with both the network-side server and the terminal-side server, wherein the trusted computing relationship indicates that the second edge-side server can perform trusted computing on the first original data held by the network-side server and the second original data held by the terminal-side server while maintaining data privacy; sending the identification information of the first original data held by the network-side server and the identification information of the second original data held by the terminal-side server to the second edge-side server respectively; aligning the identification information of the first original data held by the network-side server and the identification information of the second original data held by the terminal-side server on the second edge-side server to obtain the identification information of the intersection data; and having the second edge-side server send the identification information of the intersection data to the network-side server and the terminal-side server respectively.

[0150] Trusted computing is a computer security technology that combines cryptographic operations with security measures, ensuring end-to-end detectability. The principle of trusted computing is that the entire chain undergoes trusted authentication, so applications, operating systems, and hardware must be authorized before 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 goals. The root of trust is the foundation of trust in a trusted computer system, comprising the Trusted Measurement Root (RTM, a piece of software executed first upon platform startup), the Trusted Storage Root (RTS, a set of memories and storage root keys called platform configuration registers in the trusted platform module chip), and the Trusted Report Root (RTR, the platform configuration registers and signature keys in the trusted platform module chip). The chain of trust is the implementation technology of the trust measurement model, extending trust relationships from the root of trust to the entire computer system.

[0151] The basic idea of ​​trusted computing is to use remote proof of digital rights of ownership (DRM) to ensure security. For example, in a computer system, a root of trust is first established, and then a chain of trust is built, starting from the root of trust and extending to the hardware platform, the operating system, and the applications. Each level verifies and trusts the next, extending this trust to the entire computer system, thereby ensuring the trustworthiness of the entire computer system.

[0152] In the embodiments of this application, the second edge-side server that has established a trusted computing relationship with both the network-side server and the terminal-side server can be selected from the shared network elements that have established a trusted computing relationship with both, or other network elements can be selected outside the shared network elements to temporarily serve as the second edge-side server with a specified effective duration.

[0153] For example, selecting a second edge-side server that has established a trusted computing relationship with both the network-side server and the terminal-side server may further include: obtaining a first set of network elements that have established a trusted computing relationship with the network-side server and a second set of network elements that have established a trusted computing relationship with the terminal-side server; if there are shared network elements between the first set of network elements and the second set of network elements, then the shared network element is selected as the second edge-side server; if there are no shared network elements between the first set of network elements and the second set of network elements, then a temporary network element is selected as the second edge-side server through the network open function network element in the mobile communication network.

[0154] This application embodiment maintains a set of network elements that have established trusted computing relationships with themselves on different network elements. When a trusted second edge server needs to be selected, the common network element can be selected as the second edge server by taking the intersection of the network element sets. This avoids the need to repeatedly establish mutual trust between network elements every time network elements interact. 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.

[0155] In the embodiments of this application, the temporary network element is a network element selected from the third network element set maintained by the network open function network element, which is different from the first network element set and different from the second network element set. The temporary network element establishes a trusted computing relationship with the network-side server and the terminal-side server with a specified effective duration through the network open function network element.

[0156] Network open function network elements can periodically select network elements from the mobile communication network that meet specified security authentication standards to form a third network element set. When there are no shared, mutually trusted network elements between the network-side server and the terminal-side server, temporary network elements can be selected from the third network element set. Moreover, this temporary network element belongs neither to the first nor the second network element set. This ensures the basic trustworthiness of the network elements while preventing collusion between the temporary network element and either the network-side server or the terminal-side server, thus improving the reliability and security of data processing.

[0157] Temporary network elements establish trusted computation relationships with both the network-side server and the terminal-side server for a specified validity period through network open function network elements. If the use of a temporary network element exceeds the specified validity period, the computation relationship needs to be re-established, thereby ensuring the security and timeliness of the temporary network element.

[0158] In the embodiments of this application, in order to enable distributed computing among the network-side server, terminal-side server, and first edge-side server, a collaborative relationship needs to be established among the three. The method for establishing the collaborative relationship may include: the terminal-side server and the first edge-side server respectively sending a first collaborative request and a second collaborative request to the network-side server for performing distributed computing; establishing a first collaborative relationship for performing distributed computing between the network-side server and the terminal-side server based on a first response message returned by the network-side server to the terminal-side server in response to the first collaborative request; and establishing a second collaborative relationship for performing distributed computing between the network-side server and the first edge-side server based on a second response message returned by the network-side server to the first edge-side server in response to the second collaborative request.

[0159] This application embodiment sends a first collaboration request and a second collaboration request to the network-side server through the terminal-side server and the first edge-side server, and establishes a collaboration relationship based on the response of the network-side server, thereby realizing efficient collaboration in distributed computing. This application embodiment not only optimizes resource allocation and computing load, but also improves the speed and efficiency of data processing, while reducing latency, enhancing the scalability and flexibility of the system, and providing faster response time and better service quality.

[0160] In embodiments of this application, the terminal-side server and the first edge-side server respectively send a first collaboration request and a second collaboration request for performing distributed computing to the network-side server. This may further include: the terminal-side server and the first edge-side server respectively sending a first information opening request and a second information opening request to a network open function element in the mobile communication network. The first information opening request carries a first network element identifier for identifying the terminal-side server, and the second information opening request carries a second network element identifier for identifying the first edge-side server. On the network open function element, based on the first network element identifier and the second network element identifier, the first collaboration request and the second collaboration request for performing distributed computing are sent to the network-side server in the mobile communication network.

[0161] By sending a first information opening request and a second information opening request to the network open function network elements in the mobile communication network through the terminal-side server and the first edge-side server, and carrying a first network element identifier and a second network element identifier used to identify the terminal-side server and the first edge-side server, more accurate and efficient resource management and task allocation can be achieved. This embodiment not only improves the controllability of the network-side server over requests, but also enhances the availability and robustness of the system. Simultaneously, it reduces data transmission latency, improves response speed, reduces dependence on network bandwidth, and supports real-time data processing and analysis, which is particularly important for scenarios requiring low latency and real-time processing.

[0162] In the embodiments of this application, the network-side server and the terminal-side server train the network-side model configured on the network-side server and the terminal-side model configured on the terminal-side server through vertical federated learning. During model training, the training samples held by the network-side server and the training samples held by the terminal-side server can have the same sample identifier and different sample content.

[0163] Federated learning includes two methods: horizontal federated learning and vertical federated learning. Horizontal federated learning refers to multiple participants with different sample IDs but identical features. Vertical federated learning refers to multiple participants with the same sample IDs but different features.

[0164] This application embodiment utilizes vertical federated learning to achieve joint modeling and training by network-side servers and terminal-side servers. For example, in a 5G network, the service server can request analysis results from the NWDAF, enabling it to perform joint modeling using information from the RAN, UE, and AF, analyze and extrapolate the results, and return the results to the service server.

[0165] The data transmission method in this application will be explained in detail below with reference to specific application scenarios.

[0166] With the increasing number of mobile applications, many business sectors require more stable and timely mobile communication guarantees. For example, in the audio and video field, 4K video is increasingly becoming a pressing need in many scenarios, such as extended reality (XR). The transmission of 4K video relies heavily on high-quality and stable mobile communication. Furthermore, in areas involving large-scale models, there are also many urgent needs for large-scale model technology in 5G and even 6G mobile communication. Since large-scale model data requires the support of cloud servers, while cloud computing power can support such massive data volumes, it cannot guarantee transmission latency within a predetermined range in some areas. Therefore, on the one hand, the high transmission rate of 6G mobile communication is needed; on the other hand, edge computing power is required to ensure latency and reliability requirements are met.

[0167] Based on the above requirements, the data transmission method provided in the embodiments of this application can realize an enhanced method for mobile application and terminal service experience based on cloud-edge collaborative distributed computing for 6G mobile communication. By improving the interaction and distributed architecture process of the core network in the cloud and edge, normal QoS service guarantee can be achieved.

[0168] Figure 8 illustrates a distributed collaborative computing method for data transmission in an application scenario according to various embodiments of this application. In this figure, UE represents a terminal device, AF (Local) network element represents a terminal-side server, AF (MEC) network element represents an edge-side server, NWDAF network element represents a network-side server, and the definitions of other network elements can be found in the mobile communication network architecture shown in Figure 2, which will not be repeated here.

[0169] The distributed collaborative computing method for data transmission based on the above network element architecture may include the following steps S801 to S813.

[0170] S801: The UE initiates service registration information to the core network, starts a PDU session, and starts the service.

[0171] The UE establishes a session connection with the AF (MEC), base station, and core network. The establishment process consists of the following six steps:

[0172] (1) The UE sends an RRC connection request to the core network to establish a connection with the core network. This request includes some necessary information, including the UE's identity information and the protocol versions supported by the UE.

[0173] (2) The UE sends a first PDU session request to the core network. This request includes communication-related information between the UE and the target terminal, such as the target terminal's identity information and the data type being transmitted.

[0174] (3) The core network sends a second PDU session request to the target terminal.

[0175] (4) The target terminal sends the first PDU session request confirmation to the core network.

[0176] (5) The UE sends a second PDU session confirmation to the core network.

[0177] (6) The core network sends a PDU session establishment success message to the UE and the target terminal.

[0178] S802: The AF (Local) initiates a cloud-edge-device distributed computing registration request to the NWDAF network element in the core network, requesting NWDAF-related operations such as analysis and joint modeling. For example, the registration method may include sending an Nnwdaf_AnalyticsInfo_Request or Nnwdaf_AnalyticsSubscribe registration information to the NWDAF via NEF.

[0179] If the AF(Local) is not trusted by the NWDAF, the AF(Local) first sends an open request to the NEF. Then, the NEF registers with the NWDAF by sending the Nnef_Exposure information from the AF(Local).

[0180] During the process of NEF sending registration information to NWDAF, potential NWDAFs need to be discovered to achieve the task of vertical federated learning. The NWDAF discovery process, for example, involves confirming policy information through PCF to discover potential NWDAF network elements in the network that can be used for federated learning.

[0181] S803: The AF (MEC) initiates a registration request for cloud-edge-device distributed computing to the NWDAF network element in the core network, requesting NWDAF-related operations such as analysis and joint modeling. Specific registration methods may include sending an Nnwdaf_AnalyticsInfo_Request or Nnwdaf_AnalyticsSubscribe registration information to the NWDAF via NEF.

[0182] If the AF(MEC) is not trusted by the NWDAF, the AF(MEC) first sends an open request to the NEF by sending the Nnef_Exposure information from the AF(MEC) to the NEF, and then the NEF registers with the NWDAF.

[0183] S804: NWDAF sends confirmation information to NEF through the control plane of 5G core network. After receiving the feedback from NWDAF, NEF replies to AF (Local) and AF (MEC) to confirm the open information.

[0184] The NWDAF sends an Nwdaf_AnalyticsInfo_Request reply or an Nnwdaf_AnalyticsSubscription notification to the NEF, meaning the NWDAF provides confirmation information to the NEF through the 5G core network control. After receiving the feedback from the NWDAF, the NEF replies to the AF(Local) and AF(MEC) with confirmation of the open information. This confirmation information is denoted as Nnef_Exposure. If the AF(Local) and AF(MEC) are trusted by the NWDAF, this step is not required by the NEF.

[0185] S805: The process by which NWDAF initiates data collection from RAN and OAM. AF (MEC) and AF (Local) collect data from UE.

[0186] The information collected during data acquisition may include one or more of RSRP, RSSI, RSRQ, SINR, and cell-id. The collected RSRP, RSSI, RSRQ, and SINR can be referred to as network quality parameters. The network quality parameters obtained by NWDAF can be called the first network quality parameters, and the network quality parameters obtained by AF (Local) can be called the second network quality parameters.

[0187] For example, an NWDAF can collect information from other NWDAFs. Other network elements, such as NEF, PCF, SMF, and AMF, can also collect information using the vertical federated learning method provided in this patent. This is generally used when other network elements are unwilling to share information with the NWDAF.

[0188] NWDAF can collect information from the RAN and UE. Information for the RAN can generally be obtained from the OAM network management system. For the UE, information is obtained through the OTT-Server, but this requires the UE to be related to the relevant service.

[0189] The data collection process may involve joint training using cloud servers to obtain the processed dataset.

[0190] S806: AF (Local) trains the model locally. This model can be called the terminal-side model or the first intermediate model. During the iteration process, the update information of the first intermediate model, the number of service data packets transmitted, and the size of the service data packets transmitted are integrated. The update information of the first intermediate model can be the fourth intermediate model parameters in various embodiments of this application, and the number of service data packets transmitted, the size of the service data packets transmitted, etc., can be the first intermediate data in various embodiments of this application. The fourth intermediate model parameters can also be used as the first intermediate model parameters in various embodiments of this application to be transmitted to NWDAF for joint training of the model. AF (Local) can also iteratively update the second intermediate model parameters according to the second intermediate model parameters transmitted by NWDAF and the first data samples held by AF (Local) to obtain the updated terminal-side model and the fourth intermediate model parameters.

[0191] S807: The NWDAF trains the model using its MTLF and AnLF network elements. The model trained in this step can be called the network-side model or the second intermediate model. Similar to step S806, in this step, the third intermediate model parameters and the second intermediate data can be identified as integrated information. For multiple UEs, there can be multiple third intermediate model parameters and multiple second intermediate data. The third intermediate model parameters can also be used as the second intermediate model parameters in various embodiments of this application to be transmitted to the AF (Local) for joint model training. The NWDAF can also iteratively update the first intermediate model parameters transmitted by the AF (Local) based on the second data samples it holds to obtain the updated network-side model and the third intermediate model parameters.

[0192] The steps S806 and S807 above can be interchanged.

[0193] S808: NWDAF and AF(Local) determine the amount of integrated information that needs to be transmitted to AF(MEC).

[0194] The number of AFs (MECs) is determined by the traffic volume. Generally, one UE service requires one AF (MEC), but a joint service involving multiple UEs requires a separate AF (MEC).

[0195] S809: AF (Local) transmits the integrated information to AF (MEC).

[0196] S810: NWDAF segments the intermediate results and transmits the intermediate results based on each terminal to each AF (MEC) with or without NEF.

[0197] S811: The AF (MEC) receives transmission information from the AF (Local) and NWDAF to perform joint training of the model. The transmission information from the AF (Local) and NWDAF includes, for example, the "third intermediate model parameters" and "fourth intermediate model parameters" in various embodiments of this application. The jointly trained model is, for example, the "edge-side server" or "first edge-side server" in various embodiments of this application.

[0198] S812: The AF(MEC) infers a data discarding policy based on the jointly trained model. The data discarding policy includes at least one of the following: whether to discard service packets, the number of service packets to discard, and the transmission range of each service packet. The data discarding policy inferred by the AF(MEC) can also be referred to as the first data discarding policy.

[0199] For example, in audio and video services, the AF configures XRM services. XRM data packets consist of multiple PDU sets, and each PDU set consists of multiple service frames (I-frames, B-frames, and P-frames). Based on the number of received redundant data packets, the size of the redundant data packets, and redundancy information, the AF configures the forward error correction mechanism for each PDU set and prepares to configure it for the UE. At this time, the cloud-edge-device distributed computing inference through AF (MEC), AF, and NWDAF determines whether to discard redundant data packets, such as whether to discard I-frames, B-frames, and / or P-frames, and the number and magnitude of discarded I-frames, B-frames, and / or P-frames.

[0200] The description of the data discarding strategy in this step applies to data discarding strategies in other embodiments of this application, such as the first, second, third, fourth, and fifth data discarding strategies. The different names used for these data discarding strategies are only for distinguishing their use in different processing stages. Sometimes, even if the names are different, they may represent the same content in different embodiments. For example, the third and fourth data discarding strategies predicted by the network-side model may refer to the same content, and the second and fifth data discarding strategies predicted by the terminal-side model may refer to the same content.

[0201] S813: AF(MEC) makes intelligent AI decisions based on data discarding policies to change the current service behavior of the UE and ensure QoS service requirements.

[0202] Based on the above application scenarios, this application provides a method for enhancing the mobile application and terminal service experience based on cloud-edge collaborative distributed computing for 6G mobile communication. By proposing a method for the interaction and distributed architecture process of the core network in the cloud and edge, it satisfies normal QoS service guarantees and improves the user's service experience.

[0203] It should be noted that although the steps of the method in this application are described in a specific order in the accompanying drawings, this does not require or imply that the steps must be performed in that specific order, or that all the steps shown must be performed to achieve the desired result. Additional or alternative steps may be omitted, multiple steps may be combined into one step, and / or one step may be broken down into multiple steps.

[0204] The following describes an apparatus embodiment of this application, which can be used to execute the data transmission method in the above embodiments of this application.

[0205] Figure 9 schematically illustrates a structural block diagram of the data transmission apparatus provided in an embodiment of this application. As shown in Figure 9, the data transmission apparatus 900 includes:

[0206] The first determining module 910 is configured to determine first intermediate data based on a first network quality parameter held by a network-side server, wherein the network-side server is used to provide network communication services to terminal devices, and the first intermediate data is used to represent the first data transmission status of the network communication service.

[0207] The second determining module 920 is configured to determine second intermediate data based on a second network quality parameter held by a terminal-side server, wherein the terminal-side server is used to provide application function services to the terminal device, and the second intermediate data is used to represent the second data transmission status of the application function service.

[0208] The merging module 930 is configured to perform a merging operation on the first intermediate data and the second intermediate data on a first edge-side server to obtain a data discarding policy. The first edge-side server is used to provide edge computing services to the terminal device, and the first data discarding policy is used to instruct the terminal device to transmit all data or discard part of the data in the full data.

[0209] In some embodiments of this application, based on the above technical solutions, the network-side server includes a network data analysis function network element in a mobile communication network; the first determining module 910 includes:

[0210] The first acquisition module is configured to send a data acquisition request to the operation, maintenance and management network element in the mobile communication network.

[0211] The first receiving module is configured to receive response data returned by the operation and maintenance management network element in response to the data collection request. The response data includes first network quality parameters related to the access network and terminal equipment collected by the operation and maintenance management network element.

[0212] The first inference module is configured to perform first model inference based on the first network quality parameters according to the network-side model configured on the network data analysis function network element, and obtain the first intermediate data.

[0213] In some embodiments of this application, based on the above technical solutions, the first determining module 910 further includes:

[0214] The first training module is configured to train the terminal-side model based on the first data samples held by the terminal-side server to obtain the first intermediate model parameters. The terminal-side model is a first machine learning model configured on the terminal-side server for predicting the second data discarding strategy.

[0215] The first update module is configured to iteratively update the parameters of the first intermediate model based on the second data samples held by the network data analysis function network element to obtain the network-side model.

[0216] In some embodiments of this application, based on the above technical solutions, the terminal-side server includes application function network elements in a mobile communication network; the second determining module 920 includes:

[0217] The second acquisition module is configured to acquire data from terminal devices in the mobile communication network and obtain second network quality parameters related to the terminal devices.

[0218] The second inference module is configured to perform second model inference based on the second network quality parameters according to the terminal-side model configured on the application function network element, and obtain second intermediate data.

[0219] In some embodiments of this application, based on the above technical solutions, the second determining module 920 further includes:

[0220] The second training module is configured to train the network-side model based on the second data samples held by the network-side server to obtain the second intermediate model parameters. The network-side model is a second machine learning model configured on the network-side server for predicting the third data discarding strategy.

[0221] The second update module is configured to iteratively update the parameters of the second intermediate model based on the first data sample held by the application function network element to obtain the terminal-side model.

[0222] In some embodiments of this application, based on the above technical solutions, the merging module 930 includes:

[0223] The data merging module is configured to perform a merging operation on the first intermediate data and the second intermediate data on the first edge-side server to obtain global intermediate data.

[0224] The data inference module is configured to perform third model inference based on the global intermediate data according to the edge-side model configured on the first edge-side server, and obtain the first data discarding strategy.

[0225] In some embodiments of this application, based on the above technical solutions, the merging module 930 further includes:

[0226] The data alignment module is configured to perform data alignment between the network-side server and the terminal-side server to obtain the identification information of the intersection data. The intersection data is data with the same identification information between the first original data held by the network-side server and the second original data held by the terminal-side server.

[0227] The network-side training module is configured to select network-side data samples on the network-side server based on the identification information of the intersection data, and train a network-side model for predicting the fourth data discarding strategy based on the network-side data samples, thereby obtaining the third intermediate model parameters of the network-side model.

[0228] The terminal-side training module is configured to select terminal-side data samples on the terminal-side server based on the identification information of the intersection data, and train a terminal-side model for predicting the fifth data discarding strategy based on the terminal-side data samples, thereby obtaining the fourth intermediate model parameters of the terminal-side model.

[0229] The edge-side training module is configured to iteratively update the third intermediate model parameters of the network-side model and the fourth intermediate model parameters of the terminal-side model based on the data samples held by the first edge-side model, so as to obtain the edge-side model.

[0230] In some embodiments of this application, based on the above technical solutions, the data alignment module includes:

[0231] The network element selection module is configured to select a second edge-side server that has established a trusted computing relationship with both the network-side server and the terminal-side server. The trusted computing relationship indicates that the second edge-side server can perform trusted computing on the first original data held by the network-side server and the second original data held by the terminal-side server while maintaining data privacy.

[0232] The original identifier sending module is configured to send the identifier information of the first original data held by the network-side server and the identifier information of the second original data held by the terminal-side server to the second edge-side server, respectively.

[0233] The identifier alignment module is configured to perform data alignment on the second edge-side server on the identifier information of the first original data held by the network-side server and the identifier information of the second original data held by the terminal-side server to obtain the identifier information of the intersection data.

[0234] The intersection identifier sending module is configured to have the second edge-side server send the identifier information of the intersection data to the network-side server and the terminal-side server respectively.

[0235] In some embodiments of this application, based on the above technical solutions, the network element selection module is further configured to: obtain a first set of network elements that establish a trusted computing relationship with the network-side server and a second set of network elements that establish a trusted computing relationship with the terminal-side server; if there are shared network elements between the first set of network elements and the second set of network elements, then the shared network element is selected as the second edge-side server; if there are no shared network elements between the first set of network elements and the second set of network elements, then a temporary network element is selected as the second edge-side server through the network open function network element in the mobile communication network.

[0236] In some embodiments of this application, based on the above technical solutions, the temporary network element is a network element selected from the third network element set maintained by the network open function network element, which is different from the first network element set and different from the second network element set. The temporary network element establishes a trusted computing relationship with the network-side server and the terminal-side server with a specified effective duration through the network open function network element.

[0237] In some embodiments of this application, based on the above technical solutions, the data transmission device 900 further includes:

[0238] The collaboration request module is configured such that the terminal-side server and the first edge-side server respectively send a first collaboration request and a second collaboration request to the network-side server for performing distributed computing.

[0239] The first establishment module is configured to establish a first collaborative relationship between the network-side server and the terminal-side server for performing distributed computing based on a first response message returned by the network-side server to the terminal-side server in response to the first collaborative request.

[0240] The second establishment module is configured to establish a second collaborative relationship between the network-side server and the first edge-side server for performing distributed computing, based on a second response message returned by the network-side server to the first edge-side server in response to the second collaborative request.

[0241] In some embodiments of this application, based on the above technical solutions, the collaboration request module is further configured to: the terminal-side server and the first edge-side server respectively send a first information opening request and a second information opening request to the network open function network element in the mobile communication network, wherein the first information opening request carries a first network element identifier for identifying the terminal-side server, and the second information opening request carries a second network element identifier for identifying the first edge-side server; on the network open function network element, according to the first network element identifier and the second network element identifier, a first collaboration request and a second collaboration request for performing distributed computing are sent to the network-side server in the mobile communication network.

[0242] In some embodiments of this application, based on the above technical solutions, the network-side server and the terminal-side server train the network-side model configured on the network-side server and the terminal-side model configured on the terminal-side server through vertical federated learning; the training samples held by the network-side server and the training samples held by the terminal-side server have the same sample identifier and different sample content.

[0243] The specific details of the data transmission apparatus provided in the various embodiments of this application have been described in detail in the corresponding method embodiments, and will not be repeated here.

[0244] Figure 10 schematically illustrates a computer system architecture block diagram for implementing an electronic device according to an embodiment of this application.

[0245] It should be noted that the computer system 1000 of the electronic device shown in Figure 10 is only an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0246] As shown in Figure 10, the computer system 1000 includes a central processing unit (CPU) 1001, which can perform various appropriate actions and processes based on programs stored in read-only memory (ROM) 1002 or programs loaded from storage section 1008 into random access memory (RAM) 1003. The RAM 1003 also stores various programs and data required for system operation. The CPU 1001, ROM 1002, and RAM 1003 are interconnected via a bus 1004. An input / output interface 1005 (I / O interface) is also connected to the bus 1004.

[0247] The following components are connected to the input / output interface 1005: an input section 1006 including a keyboard, mouse, etc.; an output section 1007 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and speakers, etc.; a storage section 1008 including a hard disk, etc.; and a communication section 1009 including a network interface card such as a local area network card, modem, etc. 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. A removable medium 1011, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on the drive 1010 as needed so that computer programs read from it can be installed into the storage section 1008 as needed.

[0248] Specifically, according to embodiments of this application, the processes described in the various method flowcharts can be implemented as computer software programs. For example, embodiments of this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 1009, and / or installed from removable medium 1011. When the computer program is executed by central processing unit 1001, it performs various functions defined in the system of this application.

[0249] It should be noted that the computer-readable medium shown in the embodiments of this application can be a computer-readable signal medium, a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fiber, portable compact disc read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this application, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this application, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such transmitted data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. The computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to wireless, wired, etc., or any suitable combination thereof.

[0250] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0251] It should be noted that although several modules or units for the device used to perform actions have been mentioned in the detailed description above, this division is not mandatory. In fact, according to the embodiments of this application, the features and functions of two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided and embodied by multiple modules or units.

[0252] Through the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein can be implemented by software or by combining software with necessary hardware. Therefore, the technical solutions according to the embodiments of this application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) or on a network, including several instructions to cause a computing device (such as a personal computer, server, touch terminal, or network device, etc.) to execute the method according to the embodiments of this application.

[0253] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein.

[0254] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.

Claims

1. A data transmission method, characterized in that, The method includes: First intermediate data is determined based on a first network quality parameter held by a network-side server. The network-side server is used to provide network communication services to terminal devices. The first intermediate data is used to represent the first data transmission status of the network communication service. The second intermediate data is determined based on the second network quality parameters held by the terminal-side server. The terminal-side server is used to provide application function services to the terminal device. The second intermediate data is used to represent the second data transmission status of the application function service. The first intermediate data and the second intermediate data are merged and processed on the first edge-side server to obtain a first data discarding policy. The first edge-side server is used to provide edge computing services to the terminal device. The first data discarding policy is used to instruct the terminal device to transmit all data or discard part of the data in the full data.

2. The data transmission method according to claim 1, characterized in that, The network-side server includes a network data analysis function network element in the mobile communication network; determining the first intermediate data based on the first network quality parameters held by the network-side server includes: Send a data collection request to the operation, maintenance and management network element in the mobile communication network; The system receives response data from the operation and maintenance management network element in response to the data collection request. The response data includes the first network quality parameters related to the access network and terminal equipment collected by the operation and maintenance management network element. Based on the network-side model configured on the network data analysis function network element, the first model inference is performed based on the first network quality parameters to obtain the first intermediate data.

3. The data transmission method according to claim 2, characterized in that, The method further includes: The terminal-side model is trained based on the first data sample held by the terminal-side server to obtain the first intermediate model parameters. The terminal-side model is a first machine learning model configured on the terminal-side server for predicting the second data discarding strategy. The parameters of the first intermediate model are iteratively updated based on the second data sample held by the network data analysis function network element to obtain the network-side model.

4. The data transmission method according to any one of claims 1 to 5, characterized in that, The terminal-side server includes application function network elements in the mobile communication network; determining the second intermediate data based on the second network quality parameters held by the terminal-side server includes: Data is collected from the terminal device in the mobile communication network to obtain the second network quality parameter related to the terminal device; Based on the terminal-side model configured on the application function network element, the second model inference is performed based on the second network quality parameters to obtain the second intermediate data.

5. The data transmission method according to claim 4, characterized in that, The method further includes: The network-side model is trained based on the second data samples held by the network-side server to obtain the second intermediate model parameters. The network-side model is a second machine learning model configured on the network-side server to predict the third data discarding strategy. The terminal-side model is obtained by iteratively updating the parameters of the second intermediate model based on the first data sample held by the application function network element.

6. The data transmission method according to any one of claims 1 to 5, characterized in that, On the first edge-side server, the first intermediate data and the second intermediate data are merged to obtain the first data discarding strategy, including: On the first edge server, the first intermediate data and the second intermediate data are merged to obtain global intermediate data; Based on the edge-side model configured on the first edge-side server, the third model inference is performed based on the global intermediate data to obtain the first data discarding strategy.

7. The data transmission method according to claim 6, characterized in that, The method further includes: Data alignment is performed between the network-side server and the terminal-side server to obtain the identification information of the intersection data. The intersection data is data with the same identification information between the first original data held by the network-side server and the second original data held by the terminal-side server.

8. The data transmission method according to claim 7, characterized in that, The method further includes: Based on the identification information of the intersection data, network-side data samples are selected on the network-side server, and a network-side model for predicting the fourth data discarding strategy is trained based on the network-side data samples to obtain the third intermediate model parameters of the network-side model. Based on the identification information of the intersection data, terminal-side data samples are selected on the terminal-side server, and a terminal-side model for predicting the fifth data discarding strategy is trained based on the terminal-side data samples to obtain the fourth intermediate model parameters of the terminal-side model. The edge-side model is obtained by iteratively updating the third intermediate model parameters of the network-side model and the fourth intermediate model parameters of the terminal-side model based on the data samples held by the first edge-side server.

9. The data transmission method according to claim 7 or 8, characterized in that, Data alignment between the network-side server and the terminal-side server includes: A second edge server is selected that has established a trusted computing relationship with both the network-side server and the terminal-side server. The trusted computing relationship indicates that the second edge server can perform trusted computing on the first original data held by the network-side server and the second original data held by the terminal-side server while maintaining data privacy. The identification information of the first raw data held by the network-side server and the identification information of the second raw data held by the terminal-side server are respectively sent to the second edge-side server; On the second edge-side server, the identification information of the first original data held by the network-side server and the identification information of the second original data held by the terminal-side server are aligned to obtain the identification information of the intersection data; The second edge-side server sends the identification information of the intersection data to the network-side server and the terminal-side server respectively.

10. The data transmission method according to claim 9, characterized in that, A second edge-side server is selected that has established a trusted computing relationship with both the network-side server and the terminal-side server, including: Obtain a first set of network elements that have established a trusted computing relationship with the network-side server and a second set of network elements that have established a trusted computing relationship with the terminal-side server; If there is a shared network element between the first network element set and the second network element set, then the shared network element is selected as the second edge-side server; If there are no shared network elements between the first set of network elements and the second set of network elements, then a temporary network element is selected as the second edge-side server through the network open function network element in the mobile communication network.

11. The data transmission method according to claim 10, characterized in that, The temporary network element is a network element selected from the third network element set maintained by the network open function network element, which is different from the first network element set and different from the second network element set. The temporary network element establishes a trusted computing relationship with the network-side server and the terminal-side server with a specified effective duration through the network open function network element.

12. The data transmission method according to any one of claims 1 to 11, characterized in that, Before performing a merging operation on the first intermediate data and the second intermediate data on the first edge-side server to obtain the first data discarding policy, the method further includes: The terminal-side server and the first edge-side server respectively send a first collaboration request and a second collaboration request to the network-side server for performing distributed computing. Based on the first response message returned by the network-side server to the terminal-side server in response to the first collaboration request, a first collaboration relationship for performing the distributed computing is established between the network-side server and the terminal-side server. Based on the second response message returned by the network-side server to the first edge-side server in response to the second collaboration request, a second collaboration relationship for performing the distributed computing is established between the network-side server and the first edge-side server.

13. The data transmission method according to claim 12, characterized in that, The terminal-side server and the first edge-side server respectively send the first collaboration request and the second collaboration request to the network-side server for executing the distributed computing, including: The terminal-side server and the first edge-side server respectively send a first information opening request and a second information opening request to the network open function network element in the mobile communication network. The first information opening request carries a first network element identifier for identifying the terminal-side server, and the second information opening request carries a second network element identifier for identifying the first edge-side server. On the network open function network element, based on the first network element identifier and the second network element identifier, the first collaboration request and the second collaboration request for performing the distributed computing are sent to the network-side server in the mobile communication network.

14. The data transmission method according to any one of claims 1 to 13, characterized in that, The network-side server and the terminal-side server train the network-side model configured on the network-side server and the terminal-side model configured on the terminal-side server through vertical federated learning; the training samples held by the network-side server and the training samples held by the terminal-side server have the same sample identifiers but different sample content.

15. A data transmission device, characterized in that, The device includes: The first determining module is configured to determine first intermediate data based on a first network quality parameter held by a network-side server, wherein the network-side server is used to provide network communication services to terminal devices, and the first intermediate data is used to represent the first data transmission status of the network communication service. The second determining module is configured to determine second intermediate data based on a second network quality parameter held by a terminal-side server. The terminal-side server is used to provide application function services to the terminal device. The second intermediate data is used to represent the second data transmission status of the application function service. The merging module is configured to perform a merging operation on the first intermediate data and the second intermediate data on a first edge-side server to obtain a first data discarding policy. The first edge-side server is used to provide edge computing services to the terminal device. The first data discarding policy is used to instruct the terminal device to transmit all data or discard part of the data in the full data.

16. A computer-readable medium, characterized in that, The computer-readable medium stores a computer program that, when executed by a processor, implements the data transmission method according to any one of claims 1 to 14.

17. An electronic device, characterized in that, include: processor; as well as Memory for storing the executable instructions of the processor; The processor is configured to execute the executable instructions to implement the data transfer 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, it implements the data transmission method according to any one of claims 1 to 14.

Images