Communication method, communication apparatus, communication system, and storage medium

Abstract

Disclosed in the embodiments of the present application are a communication method, a communication apparatus, a communication system, and a storage medium, which are applied to the technical field of communications, and used for implementing multi-vendor model learning. The method in the embodiments of the present application comprises: sending a first message to a first terminal device, wherein the first message is used for acquiring a first model, and the first model is obtained by means of training by a first network server; and receiving the first model, wherein the first model is used for being trained with a second model, so as to obtain a third model, and the second model is obtained by means of training by a second network server. The embodiments of the present application can implement model transfer between servers by means of exchanging models or server address information between devices, thereby implementing multi-vendor model learning.

Description

A communication method, communication device, communication system, and storage medium

[0001] This application claims priority to Chinese Patent Application No. 202411001965.2, filed on July 24, 2024, entitled "A Communication Method, Communication Device, Communication System and Storage Medium", the entire contents of which are incorporated herein by reference. Technical Field

[0002] This application relates to the field of communication technology, and in particular to a communication method, communication device, communication system and storage medium. Background Technology

[0003] The 3rd Generation Partnership Project (3GPP) Release 18 explores the introduction of artificial intelligence (AI) and machine learning (ML) technologies into the radio air interface for applications such as radio channel information compression and reconstruction, beam management, and positioning enhancement. Neural network-based AI / ML utilizes data-driven training to improve the performance of wireless tasks. AI / ML systems encompass data collection, model training, performance monitoring, model management, model inference, and model storage.

[0004] For a two-terminal model consisting of a transmitter and a receiver, the transmitter and receiver models can be trained on the user equipment (UE) or network (NW) side. The trained transmitter or receiver model is then transferred to the NW or UE for deployment or training of the NW or UE model. Model learning is primarily achieved through offline model transfer between vendor servers. The trained model is then transferred from the terminal or network server to its terminal or network, or vice versa, enabling communication between the terminal model and the network model.

[0005] However, since the models transmitted between the terminal and the network are generally pre-trained or compiled models, and the transmission is generally a one-way process, it is not suitable for training models from multiple vendors. Summary of the Invention

[0006] This application provides a communication method, communication device, communication system, and storage medium, which can exchange model or server address information between devices to realize model transfer between servers, thereby enabling multi-vendor model learning.

[0007] The first aspect of this application provides a communication method. Optionally, the executing entity of this method can be a terminal server. The terminal server can be a server, a component or device applied to a server (e.g., a processor, chip, or chip system), or a logic module or software capable of implementing all or part of the server's functions (e.g., a central unit (CU), a distributed unit (DU), or a radio unit (RU)). Taking a first terminal server as an example, in this method, the first terminal server sends a first message to a first terminal device. The first message is used to obtain a first model, which is trained by a first network server. The first terminal server receives the first model, which is used to train a third model with a second model. The second model is trained by a second network server.

[0008] In this embodiment of the application, the first terminal server receives the first model and trains the model on the terminal server, so that the terminal server can train a unified model for different network servers, thereby reducing the complexity of model management.

[0009] A second aspect of this application provides a communication method. Optionally, the executing entity of this method can be a terminal server. The terminal server can be a server, a component or device applied to the server (e.g., a processor, chip, or chip system), or a logic module or software (e.g., CU, DU, or RU) capable of implementing all or part of the server's functions. Taking a server as an example, in this method, the terminal server sends a first message to a first terminal device. The first message is used to obtain a first model, which is trained by a fourth terminal server. The terminal server receives the first model, which is used to train a third model with a second model. The second model is trained by a fifth terminal server.

[0010] In this embodiment, the first terminal server transmits the model through the terminal device, enabling model training between terminal servers. This allows the terminal servers to train a unified model for terminal servers from different manufacturers, achieving model learning between multiple manufacturers and reducing the complexity of model management.

[0011] Based on the first or second aspect of this application, in some possible implementations, a first terminal server receives first information from a first terminal device. The first information includes at least one of the following: the identification (ID) of the cell where the first terminal device is located, the manufacturer ID of the first terminal device, and the model ID of the first terminal device. The terminal server can send a first message to the first terminal device based on the first information.

[0012] Based on the first or second aspect of this application, in some possible implementations, the first message includes the ID of the first model or the ID of the first task, wherein the first model is used for the first task.

[0013] Based on the first or second aspect of this application, in some possible implementations, the first terminal server can train a third model based on the first model and the second model.

[0014] Based on the first or second aspect of this application, in some possible implementations, the first terminal server may receive a first model from a first terminal device, wherein the first model is obtained by the first terminal device from the first network server through a first network device, or the first model is obtained by the first terminal device from the fourth terminal server through a fourth terminal device.

[0015] Based on the first or second aspect of this application, in some possible implementations, the first terminal server may send a third model to the first terminal device.

[0016] Based on the first or second aspect of this application, in some possible implementations, the first message further includes the address of a first terminal server, which is used by a first network server to establish a connection with the first terminal server, or by a fourth terminal server to establish a connection with the first terminal server. The first terminal server can receive a first model from the first network server.

[0017] Optionally, the first terminal server may send the third model to the first network server.

[0018] Optionally, the first model and the second model are decoder models;

[0019] or,

[0020] The first and second models are encoder models.

[0021] A third aspect of this application provides a communication method. Optionally, the execution subject of this method can be a terminal device, a component or device applied to the terminal device (e.g., a processor, chip, or chip system), or a logic module or software capable of implementing all or part of the functions of the terminal device. Taking a first terminal device as an example, in this method, the first terminal device receives a first message from a first terminal server. The first message is used to obtain a first model. The first model can be trained by a first network server or by a fourth terminal server. The first terminal device sends the first model to the first terminal server. The first model is obtained from the first network server through the first network device, or obtained from the fourth terminal server through the fourth terminal device.

[0022] Optionally, the first terminal device may send first information to the first terminal server. The first information includes at least one of the following: the ID of the cell where the first terminal device is located, the ID of the manufacturer of the first terminal device, or the ID of the model of the first terminal device.

[0023] Optionally, the first message includes the ID of the first model or the ID of the first task, whereby the first model is used for the first task.

[0024] A fourth aspect of this application provides a communication method. Optionally, the executing entity of this method can be a network server. The network server can be a server, a component or device applied to the server (e.g., a processor, chip, or chip system), or a logic module or software capable of implementing all or part of the server's functions. Taking a third network server as an example, the third network server receives a fourth model, which is trained by a second terminal server. The fourth model is used to train a sixth model with a fifth model, which is also trained by the third terminal server. The third network server trains the sixth model based on the fourth and fifth models.

[0025] The fifth aspect of this application provides a communication method. Optionally, the executing entity of this method can be a network server. The network server can be a server, a component or device applied to the server (e.g., a processor, chip, or chip system), or a logic module or software capable of implementing all or part of the server's functions. Taking a third network server as an example, the third network server receives a fourth model, which is trained by the fourth network server. The fourth model is used to train a sixth model with a fifth model, which is trained by the fifth network server. The third network server trains the sixth model based on the fourth and fifth models.

[0026] Based on the fourth or fifth aspect of this application, in some possible implementations, the third network server may receive a fourth model from the second network device, the fourth model being obtained by the second network device from the second terminal server through the second terminal device, or the fourth model being obtained by the second network device from the fourth network server through the third network device.

[0027] Based on the fourth or fifth aspect of this application, in some possible implementations, the third network server may receive a fifth model from the second network device, the fifth model being obtained by the second network device from the third terminal server through the third terminal device, or the fifth model being obtained by the second network device from the fifth network server through the third network device.

[0028] Based on the fourth or fifth aspect of this application, in some possible implementations, the third network server may receive the fourth model from the second terminal server, or receive the fourth model from the fourth network server.

[0029] Based on the fourth or fifth aspect of this application, in some possible implementations, the third network server may receive the fifth model from the third terminal server, or receive the fifth model from the fifth network server.

[0030] Based on the fourth or fifth aspect of this application, in some possible implementations, the fourth model and the fifth model are decoder models;

[0031] or,

[0032] The fourth and fifth models are encoder models.

[0033] The sixth aspect of this application provides a communication method. Optionally, the subject executing the method can be a network device, a component or device applied to the network device (e.g., a processor, chip, or chip system), or a logic module or software capable of implementing all or part of the functions of the network device. Taking a second network device as an example, in this method, the second network device generates a second message, which is used to request a fourth model. The fourth model is trained by a second terminal server, or by a fourth network server. The second network device sends the second message to a second terminal device or a third network device.

[0034] Optionally, the second network device may receive second information from the second terminal device, the second information including at least one of the cell ID of the second terminal device, the manufacturer ID of the second terminal device, or the model ID of the second terminal device. The network device may send a second message to the second terminal device based on the second information.

[0035] Optionally, the second message may include the ID of the fourth model or the ID of the second task, wherein the fourth model is used for the second task.

[0036] Optionally, the second network device may receive a fourth model from the second terminal device, which is obtained by the second terminal device from the second terminal server.

[0037] Optionally, the second network device may receive a fourth model from the third network device, which is obtained by the third network device from the fourth network server.

[0038] Optionally, the second message may also include the address of a third network server, which is used by the second terminal server to establish a connection with the third network server, or by the fourth network server to establish a connection with the third network server.

[0039] A seventh aspect of this application provides a communication device, comprising:

[0040] The processing module is used to generate the first message;

[0041] The interface module is used to send a first message to the first terminal device. The first message is used to obtain the first model, which is trained by the first network server.

[0042] The interface module is also used to receive the first model, which is used to train the third model with the second model. The second model is trained by the second network server.

[0043] The eighth aspect of this application provides a communication device, comprising:

[0044] The processing module is used to generate the first message;

[0045] The interface module is used to send a first message to the first terminal device. The first message is used to obtain the first model, which is trained by the fourth terminal server.

[0046] The interface module is also used to receive the first model, which is used to train the third model with the second model. The second model is trained by the fifth terminal server.

[0047] Based on the seventh or eighth aspect of this application, in some possible implementations, the interface module is further configured to receive first information from the first terminal device, the first information including at least one of the ID of the cell where the first terminal device is located, the ID of the manufacturer of the first terminal device, and the ID of the model of the first terminal device.

[0048] The interface module is specifically used to send a first message to the first terminal device based on the first information.

[0049] Based on the seventh or eighth aspect of this application, in some possible implementations, the first message includes the ID of the first model or the ID of the first task, wherein the first model is used for the first task.

[0050] Based on the seventh or eighth aspect of this application, in some possible implementations, the processing module is further configured to train a third model based on the first model and the second model.

[0051] Based on the seventh or eighth aspect of this application, in some possible implementations, the interface module is specifically used to receive a first model from a first terminal device, the first model being obtained by the first terminal device from a first network server via a first network device, or from a fourth terminal server via a fourth terminal device.

[0052] Based on the seventh or eighth aspect of this application, in some possible implementations, the interface module is further configured to send a third model to the first terminal device.

[0053] Based on the seventh or eighth aspect of this application, in some possible implementations, the first message further includes the address of the first terminal server, which is used by the first network server to establish a connection with the first terminal server, or by the fourth terminal server to establish a connection with the first terminal server.

[0054] The interface module is specifically used to receive the first model from the first network server or the fourth terminal server.

[0055] Optionally, the interface module can also be used to send the third model to the first network server.

[0056] The ninth aspect of this application provides a communication device, comprising:

[0057] The interface module receives a first message from the first terminal server. The first message is used to obtain the first model, which is trained by the first network server or the fourth terminal server.

[0058] The interface module is also used to send a first model to the first terminal server, wherein the first model is obtained from the first network server through the first network device, or obtained from the fourth terminal server through the fourth terminal device.

[0059] The processing module is used to maintain the first model.

[0060] Optionally, the interface module is also used to send first information to the first terminal server. The first information includes at least one of the following: the ID of the cell where the first terminal device is located, the ID of the manufacturer of the first terminal device, or the ID of the model of the first terminal device.

[0061] Optionally, the first message includes the ID of the first model or the ID of the first task, whereby the first model is used for the first task.

[0062] The tenth aspect of this application provides a communication device, comprising:

[0063] The interface module is used to receive the fourth model, which is trained by the second terminal server. The fourth model is used to train the sixth model with the fifth model, which is trained by the third terminal server.

[0064] The processing module is used to train the sixth model based on the fourth and fifth models.

[0065] The eleventh aspect of this application provides a communication device, comprising:

[0066] The interface module is used to receive the fourth model, which is trained by the fourth network server. The fourth model is used to train the sixth model with the fifth model, which is trained by the fifth network server.

[0067] The processing module is used to train the sixth model based on the fourth and fifth models.

[0068] Based on the tenth or eleventh aspect of this application, in some possible implementations, the interface module is specifically used to receive a fourth model from a second network device, which is obtained by the second network device from a second terminal server through a second terminal device, or from a fourth network server through a third network device.

[0069] Optionally, the interface module is also used to receive a fifth model from the second network device, which is obtained by the second network device from the third terminal server through the third terminal device, or from the fifth network server through the fourth network device.

[0070] Optionally, the interface module is specifically used to receive a fourth model from a second terminal server or a fourth network server.

[0071] Optionally, the interface module is also used to receive a fifth model from a third terminal server or a fifth network server.

[0072] The twelfth aspect of this application provides a communication device, comprising:

[0073] The processing module is used to generate a second message, which is used to request a fourth model. The fourth model is trained by a second terminal server or a fourth network server.

[0074] The interface module is used to send a second message to the second terminal device.

[0075] Optionally, the interface module is also used to receive second information from the second terminal device, the second information including at least one of the cell ID where the second terminal device is located, the manufacturer ID of the second terminal device, or the model ID of the second terminal device;

[0076] The interface module is specifically used to send a second message to the second terminal device based on the second information.

[0077] Optionally, the second message may include the ID of the fourth model or the ID of the second task, wherein the fourth model is used for the second task.

[0078] Optionally, the interface module is specifically used to receive the fourth model from the second terminal device, which is obtained by the second terminal device from the second terminal server.

[0079] or,

[0080] Receive the fourth model from the third network device, which is obtained by the third network device from the fourth network server.

[0081] Optionally, the second message may also include the address of a third network server, which is used by the second terminal server to establish a connection with the third network server, or by the fourth network server to establish a connection with the third network server.

[0082] The thirteenth aspect of this application provides a communication device, which may be a terminal server, network server, terminal device, or network device; it may also be a component (e.g., processor, chip, or chip system) applied to a terminal server, network server, terminal device, or network device; or it may be a logic module or software (e.g., CU, DU, or RU) capable of implementing all or part of the functions of a terminal server, network server, terminal device, or network device. The communication device includes:

[0083] A processor for executing a program that causes the communication device to perform the method described in any of the first to sixth aspects and any possible implementation thereof.

[0084] Optionally, the communication device further includes a memory, and the processor is coupled to the memory; the memory is used to store programs.

[0085] The fourteenth aspect of this application provides a chip or chip system including at least one processor and a communication interface, the communication interface and at least one processor being interconnected via a line, the at least one processor being used to run computer programs or instructions to perform the communication methods described in any of the first to sixth aspects and any possible implementation thereof.

[0086] The communication interface in the chip can be an input / output interface, pins, or circuits.

[0087] In one possible implementation, the chip or chip system described above in this application further includes at least one memory storing instructions. The memory can be an internal storage unit of the chip, such as a register or cache, or it can be a storage unit of the chip itself, such as a read-only memory or random access memory.

[0088] The fifteenth aspect of this application provides a communication system, including communication means for performing the first aspect and any possible implementation thereof, communication means for performing the third aspect and any possible implementation thereof, communication means for performing the fourth aspect and any possible implementation thereof, and communication means for performing the sixth aspect and any possible implementation thereof.

[0089] The sixteenth aspect of this application provides a computer-readable storage medium including instructions that, when executed on a computer, cause the computer to perform the method described in the first aspect above, or cause the computer to perform the method described in the second aspect above, or cause the computer to perform the method described in the third aspect above, or cause the computer to perform the method described in the fourth aspect above, or cause the computer to perform the method described in the fifth aspect above, or cause the computer to perform the method described in the sixth aspect above.

[0090] The seventeenth aspect of this application provides a computer program product containing instructions that, when run on a computer, cause the computer to perform the method described in the first aspect above, or cause the computer to perform the method described in the second aspect above, or cause the computer to perform the method described in the third aspect above, or cause the computer to perform the method described in the fourth aspect above, or cause the computer to perform the method described in the fifth aspect above, or cause the computer to perform the method described in the sixth aspect above. Attached Figure Description

[0091] Figure 1 is a network structure diagram in an embodiment of this application;

[0092] Figure 2 illustrates a possible application scenario of the communication method in this application embodiment;

[0093] Figure 3 is a schematic diagram of an embodiment of multi-vendor model learning in this application;

[0094] Figure 4 is a schematic diagram of an embodiment of the communication method in this application;

[0095] Figure 5 is a schematic diagram of an embodiment of model training in this application;

[0096] Figure 6 is a schematic diagram of another embodiment of model training in this application;

[0097] Figure 7 is a schematic diagram of another embodiment of model training in this application;

[0098] Figure 8 is a schematic diagram of another embodiment of model training in this application;

[0099] Figure 9 is a schematic diagram of another embodiment of the communication method in this application;

[0100] Figure 10 is a schematic diagram of another embodiment of the communication method in this application;

[0101] Figure 11 is a schematic diagram of another embodiment of the communication method in this application;

[0102] Figure 12 is a schematic diagram of another embodiment of the communication method in this application;

[0103] Figure 13 is a schematic diagram of another embodiment of the communication method in this application;

[0104] Figure 14 is a schematic diagram of another embodiment of the communication method in this application;

[0105] Figure 15 is a schematic diagram of another embodiment of the communication method in this application;

[0106] Figure 16 is a schematic diagram of an embodiment of the communication device in this application;

[0107] Figure 17 is a schematic diagram of another embodiment of the communication device in this application;

[0108] Figure 18 is a schematic diagram of another embodiment of the communication device in this application;

[0109] Figure 19 is a schematic diagram of another embodiment of the communication device in this application;

[0110] Figure 20 is a schematic diagram of another embodiment of the communication device in this application;

[0111] Figure 21 is a schematic diagram of another embodiment of the communication device in this application. Detailed Implementation

[0112] This application provides a communication method, communication device, communication system, and storage medium, which are applied in the field of communication technology. Servers exchange model or server address information with terminal devices through network devices to realize model transfer between servers, thereby enabling multi-vendor model learning.

[0113] The embodiments of this application will now be described with reference to the accompanying drawings. Those skilled in the art will recognize that, with technological advancements and the emergence of new scenarios, the technical solutions provided in the embodiments of this application are equally applicable to similar technical problems.

[0114] The terms "first," "second," etc., used in the specification, claims, and drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such terms are interchangeable where appropriate; this is merely a way of distinguishing objects with the same attributes in the description of embodiments of this application. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion, so that a process, method, system, product, or apparatus that comprises a series of elements is not necessarily limited to those elements, but may include other elements not expressly listed or inherent to those processes, methods, products, or apparatuses.

[0115] First, some technical terms involved in the embodiments of this application will be introduced.

[0116] 1) Joint training

[0117] Joint training is a strategy that involves training multiple models or different parts of a model simultaneously, sharing information or objectives during training to improve overall performance or achieve more complex tasks. Joint training is typically used to address problems that require consideration of multiple functions, perspectives, modalities, or tasks simultaneously, such as joint optimization of encoders and decoders, multimodal learning, multi-task learning, or joint optimization problems in complex scenarios.

[0118] Multi-task learning (MTL) is a scenario that utilizes joint training. In this scenario, the model is designed to learn multiple related tasks simultaneously, leveraging commonalities between tasks by sharing representation layers or certain parameters, thereby improving performance across all tasks. Joint training here means simultaneously optimizing the loss functions of multiple tasks, enabling the model to learn feature representations that are useful for all tasks.

[0119] 2) Model distillation;

[0120] Model distillation is a model compression technique that trains a small model (student model) to mimic the predictive capabilities of a pre-trained large model (teacher model), thereby reducing model complexity and computational cost while maintaining or approaching the performance of the teacher model. Model distillation uses the output probability distribution of the teacher model as a target, allowing the student model to learn how to approximate the teacher model's predictions. This knowledge transfer process helps improve the generalization ability and performance of the student model.

[0121] Please refer to Figure 1. The network architecture on which the communication method in this embodiment is based is briefly described below:

[0122] Figure 1 is a possible, non-limiting system schematic diagram. As shown in Figure 1, the communication system 10 includes a radio access network (RAN) 100 and a core network (CN) 200. RAN 100 includes at least one RAN node (110a and 110b in Figure 1, collectively referred to as 110) and at least one terminal (120a-120j in Figure 1, collectively referred to as 120). RAN 100 may also include other RAN nodes, such as wireless relay equipment and / or wireless backhaul equipment (not shown in Figure 1). Terminal 120 is wirelessly connected to RAN node 110. RAN node 110 is wirelessly or wired connected to core network 200. The core network equipment in core network 200 and RAN node 110 in RAN 100 can be different physical devices, or they can be the same physical device integrating core network logical functions and radio access network logical functions.

[0123] RAN 100 can be a cellular system related to the 3rd Generation Partnership Project (3GPP), such as 4G, 5G mobile communication systems, or future mobile communication systems. RAN 100 can also be an open-radio access network (ORAN), a cloud-radio access network (CRAN), or a wireless fidelity (WiFi) system. RAN 100 can also be a communication system that integrates two or more of the above systems.

[0124] RAN node 110, sometimes also referred to as access network equipment, RAN entity, or access node, constitutes part of the communication system and is used to help terminals achieve wireless access. Multiple RAN nodes 110 in communication system 10 can be of the same type or different types. In some scenarios, the roles of RAN node 110 and terminal 120 are relative. For example, network element 120i in Figure 1 can be a helicopter or drone, which can be configured as a mobile base station. For terminals 120j accessing RAN 100 through network element 120i, network element 120i is a base station; but for base station 110a, network element 120i is a terminal. RAN node 110 and terminal 120 are sometimes both referred to as communication devices. For example, network elements 110a and 110b in Figure 1 can be understood as communication devices with base station functions, and network elements 120a-120j can be understood as communication devices with terminal functions.

[0125] In one possible scenario, the RAN node can be a base station (BS), an evolved NodeB (eNodeB), an access point (AP), a transmission reception point (TRP), a next-generation NodeB (gNB), a next-generation base station in a future mobile communication system, a base station in a future mobile communication system, or an access node in a WiFi system. The RAN node can be a macro base station (as shown in Figure 1, 110a), a micro base station or indoor station (as shown in Figure 1, 110b), a relay node or donor node, or a radio controller in a CRAN scenario. Optionally, the RAN node can also be a server, wearable device, vehicle, or in-vehicle equipment. For example, the access network equipment in vehicle-to-everything (V2X) technology can be a roadside unit (RSU). All or part of the functions of the RAN node in this application can also be implemented through software functions running on hardware, or through virtualization functions instantiated on a platform (e.g., a cloud platform). The RAN node can also be equipped with communication modules, circuits, or chips that perform corresponding communication functions. The RAN node can also be configured with program instructions for performing corresponding communication functions, as well as corresponding program instructions. The RAN node in this application can also be a logical node, logical module, or software capable of implementing all or part of the RAN node's functions.

[0126] In another possible scenario, multiple RAN nodes collaborate to assist the terminal in achieving wireless access, with each RAN node performing a portion of the base station's functions. For example, RAN nodes can be central units (CUs), distributed units (DUs), CU-CPs, CU-UPs, or radio units (RUs). CUs and DUs can be configured separately or included in the same network element, such as a baseband unit (BBU). RUs can be included in radio equipment or radio units, such as remote radio units (RRUs), active antenna units (AAUs), or remote radio heads (RRHs).

[0127] In different systems, CU (or CU-CP and CU-UP), DU, or RU may have different names, but those skilled in the art will understand their meaning. For example, in an ORAN system, CU can also be called O-CU (open CU), DU can also be called O-DU, CU-CP can also be called O-CU-CP, CU-UP can also be called O-CU-UP, and RU can also be called O-RU. For ease of description, this application uses CU, CU-CP, CU-UP, DU, and RU as examples. Any of the units among CU (or CU-CP, CU-UP), DU, and RU in this application can be implemented through software modules, hardware modules, or a combination of software and hardware modules.

[0128] A terminal can be a device or module that accesses the aforementioned communication system and has corresponding communication functions. A terminal can also be called a terminal device, user equipment (UE), mobile station, mobile terminal, etc. Terminals can be widely used in various scenarios, such as device-to-device (D2D), vehicle-to-everything (V2X) communication, machine-type communication (MTC), Internet of Things (IoT), virtual reality, augmented reality, industrial control, autonomous driving, telemedicine, smart grids, smart homes, smart offices, smart wearables, intelligent transportation, smart cities, etc. Terminals can be mobile phones, tablets, computers with wireless transceiver capabilities, wearable devices, vehicles, drones, helicopters, airplanes, ships, robots, robotic arms, smart home devices, transportation vehicles with wireless communication capabilities, communication modules, etc. The embodiments of this application do not limit the device form of the terminal. Terminals typically contain communication modules, circuits, or chips that perform corresponding communication functions. Terminals can also be configured with program instructions for performing corresponding communication functions.

[0129] Furthermore, the embodiments of this application can also be applied to other future communication technologies. The network architecture and service scenarios described in this application are for the purpose of more clearly illustrating the technical solutions of this application, and do not constitute a limitation on the technical solutions provided in this application. As those skilled in the art will understand, with the evolution of network architecture and the emergence of new service scenarios, the technical solutions provided in this application are also applicable to similar technical problems.

[0130] Figure 2 illustrates an application scenario applicable to an embodiment of this application. Currently, for a dual-end model consisting of a transmitter and a receiver, the transmitter and receiver models can be trained on a terminal server or a network server, and the trained transmitter or receiver model can be passed to a network device or a terminal device for deployment or training of the network-side model or terminal-side model. For example, terminal server 202 is responsible for training the transmitter model, and terminal server 202 passes the trained terminal-side model to terminal device 201; network server 204 is responsible for training the receiver model, and network server 204 passes the trained network-side model to network device 203, and network device 203 can pass the network-side model to terminal device 201.

[0131] However, since the models transmitted between the terminal and the network are generally pre-trained or compiled models, and the transmission is generally a one-way process, it is not suitable for training models from multiple vendors.

[0132] Based on this, this application provides a method. Referring to Figure 3, a terminal server or network server can obtain a model from a network device or network server through interaction between the terminal device and the network device, train a terminal-side model for multiple cells, and feed it back to the network server; the network server can obtain a model from a terminal device or terminal server through interaction between the network device and the terminal device, train a network-side model, and feed it back to the terminal server; through multi-round model information interaction between the terminal server and the network server, multi-vendor model learning is achieved. Simultaneously, the terminal server can also obtain the counterpart model through interaction between terminal devices. The network server can obtain the counterpart model through interaction between network devices.

[0133] The following sections will explain the transfer process based on the model.

[0134] 1. The first terminal server obtains the first model from the first network device;

[0135] Please refer to Figure 4. One communication method in this embodiment includes:

[0136] 401. The first terminal server sends a first message to the first terminal device, and correspondingly, the first terminal device receives the first message from the first terminal server;

[0137] The first terminal server sends a first message to the first terminal device. The first message is used to obtain the first model, which is trained by the first network server.

[0138] Specifically, the first terminal server selects the first terminal device as the terminal device participating in the model transmission and sends the first message to the first terminal device.

[0139] It should be noted that the first terminal server has multiple terminal devices connected to it, including the first terminal device itself. The first terminal server selects a subset of these terminal devices as participating devices in model transmission, where each of these terminal devices is located in a different cell. Alternatively, the first terminal server selects one terminal device from each of the multiple terminal devices in each different cell to obtain a subset of terminal devices, which then participate in model transmission. The first terminal server sends a message to each of these terminal devices, requesting the network-side model from the network equipment in the cell where that subset of terminal devices resides.

[0140] Optionally, the first message includes the identification (ID) of the first model or the ID of the first task, and the first model is used for the first task.

[0141] Optionally, the first terminal server can select some terminal devices in turn, or it can select them based on the geographical coordinates of the connected terminal devices. The specifics are not limited here.

[0142] Optionally, the first terminal server can select a terminal device with better channel quality from among multiple terminal devices, or it can randomly select a terminal device; the specific choice is not limited here.

[0143] In practical applications, the first message can also be referred to as the first request or the first model request. This application does not limit the naming of the first message.

[0144] The first terminal device can request the first model from the first network device based on the ID of the first model or the ID of the first task. The first network device can be the RAN node of the cell where the first terminal device is located, and the first network device is connected to the first network server. Then, the first terminal device receives the first model from the first network device.

[0145] 402. The first terminal server receives the first model from the first terminal device, and correspondingly, the first terminal device sends the first model to the first terminal server;

[0146] The first terminal server receives the first model, which is used to train the third model with the second model. The second model is trained by the second network server.

[0147] Specifically, the first terminal server receives a first model from the first terminal device. This first model is then fused with a second model to obtain a third model. The second model originates from and is trained by the second network server.

[0148] Optionally, the first terminal server receives a second model transmitted from the second terminal device, which is used to fuse with the first model to obtain a third model.

[0149] It should be noted that the first model can be carried in a response message sent by the first terminal device to the first terminal server, and this response message is a response message to the first message. This response message can also be called the first response or the first model response; the naming of this application embodiment is not limited.

[0150] Optionally, the first and second models can be encoder models, and the third model can be a decoder model or an encoder model.

[0151] Optionally, the first and second models can be decoder models, and the third model can be either an encoder model or a decoder model.

[0152] The encoder model can also be called the transmitting model, and the decoder model can also be called the receiving model. The transmitting model includes at least one of the following functions: channel state information compression, channel coding, symbol modulation, resource mapping, waveform modulation, and radio frequency (RF) processing. The receiving model includes at least one of the following functions: channel state information reconstruction, channel decoding, symbol demodulation, de-resource mapping, waveform demodulation, or RF processing.

[0153] In this embodiment, the first terminal server transmits the model through the terminal device, enabling the training of the model on the terminal server. This allows the terminal server to train a unified model for different network servers, thereby reducing the complexity of model management.

[0154] It is understandable that the first model received by the first terminal server can be a model trained locally by the first network server, or a model obtained by the first network server training multiple terminal-side models; the specifics are not limited here.

[0155] Optionally, the embodiment shown in Figure 4 further includes step 400a. Step 400a may be performed before step 401.

[0156] 400a. The first network server sends the first model to the first network device, and correspondingly, the first network device receives the first model from the first network server;

[0157] The first network device can obtain the first model from the first network server and maintain it.

[0158] Optionally, the embodiment shown in Figure 4 further includes step 400b. Step 400b may be performed before step 401.

[0159] 400b. The first terminal server sends an information reporting instruction to the first terminal device, and the first terminal device receives the information reporting instruction from the first terminal server accordingly.

[0160] The first terminal server sends an information reporting instruction to at least one of a plurality of terminal devices. The information reporting instruction is used to instruct the at least one terminal device to report first information, wherein the at least one terminal device includes the first terminal device.

[0161] Optionally, the first terminal server sends information reporting instructions to all terminal devices connected to it.

[0162] Optionally, the first terminal server can send information reporting instructions to the terminal device with the best channel quality among multiple terminal devices, or it can randomly select a terminal device to send information reporting instructions to; the specifics are not limited here.

[0163] Optionally, the embodiment shown in Figure 4 further includes step 400c. Step 400c may be performed before step 401.

[0164] 400c. The first terminal device sends first information to the first terminal server, and correspondingly, the first terminal server receives the first information from the first terminal device.

[0165] The first terminal server receives first information from the first terminal device. The first information includes at least one of the following: the ID of the cell where the first terminal device is located, the ID of the vendor of the first terminal device, and the ID of the model of the first terminal device.

[0166] Specifically, the first terminal device responds to the information reporting instruction from the first terminal server by sending the first information to the first terminal server.

[0167] Optionally, the first terminal server can determine a first terminal device from multiple terminal devices as the terminal device participating in model transmission based on the first information. For example, the first terminal server can select one terminal device in each cell for model transmission for each task, application scenario, or configured model.

[0168] Optionally, the embodiment shown in FIG4 further includes step 401a. Step 401a may be performed after step 401.

[0169] 401a. The first terminal device sends a first message to the first network device, and correspondingly, the first network device receives the first message from the first terminal device;

[0170] The first terminal device transmits the first message from the first terminal server to the first network device in order to request the first model from the first network device.

[0171] Optionally, the embodiment shown in FIG4 further includes step 401b. Step 401b may be performed after step 401a.

[0172] 401b. The first network device sends a first model to the first terminal device, and correspondingly, the first terminal device receives the first model from the first network device;

[0173] In response to the first message, the first network device sends the network-side model to the first terminal device, which then transmits the first model to the first terminal server.

[0174] It should be noted that the first model is sent from the first network server to the first network device; that is, the first network server sends the first model to the first network device, and the first network device then sends the first model to the first terminal device. Therefore, it can also be said that the first model is obtained by the first terminal device from the first network server through the first network device.

[0175] Optionally, the embodiment shown in Figure 4 further includes step 403. Step 403 may be performed after step 402.

[0176] 403. The first terminal server performs model fusion;

[0177] The first terminal server fuses network-side models obtained from multiple terminal devices to obtain a third model. The process by which the first terminal server obtains models from each terminal device is similar to the process in steps 401 to 402 where the first terminal server obtains the first model from the first terminal device; please refer to the aforementioned description. For example, the first terminal server receives a first model from a first terminal device and a second model from a second terminal device. Then, the first terminal server fuses the first and second models to obtain the third model.

[0178] In one possible implementation, the terminal devices selected by the first terminal server include both a first terminal device and a second terminal device, wherein the cell where the first terminal device is located is different from the cell where the second terminal device is located. Therefore, the network-side model corresponding to the first terminal device is different from the network-side model corresponding to the second terminal device; that is, the first model and the second model are different.

[0179] Specifically, the first terminal server can train multiple network-side models using joint training or model distillation. Taking receiver training as an example, the i-th transmitter model obtained from the peer end is g. i (·;θi ), θ i Let f(·φ) be the transmitter model parameters, and f(·φ) be the receiver model parameters to be trained. Joint training is shown in Figure 5. The receiver model parameters are updated with the goal of achieving end-to-end performance in communication with multiple transmitter models (e.g., minimizing the receiver's loss function). After several iterations, the receiver model is fed back to the peer for model training.

[0180] The training process based on model distillation is shown in Figure 6. First, a common transmitter model is trained based on multiple transmitter models, making its output close to the outputs of each transmitter model. Then, using the common transmitter model as the transmitter, the receiver parameters are updated with the goal of improving end-to-end communication performance with the receiver (e.g., minimizing the receiver's loss function). The encoder or decoder model is then fed back to the peer for model training.

[0181] For example, in uplink communication, the terminal device needs to encode the message, and the network device needs to decode the message. The uplink communication model is shown in Figure 7. Taking the terminal device as UE and the network device as BS as an example, UE Vendor A represents the terminal device of vendor A; UE Vendor B represents the terminal device of vendor B, and the terminal devices of vendor A and vendor B belong to different terminal servers. A terminal device of vendor A can represent multiple terminal devices under a terminal server of vendor A. Similarly, a terminal device of vendor B can represent multiple terminal devices under a terminal server of vendor B. BS0 represents the first network device, and BS1 represents the second network device, and the first and second network devices belong to different network servers. As shown in Figure 7, the first and second network devices each maintain a decoder model. For UEs from different vendors, the decoder models maintained by the first and second network devices are different. Similarly, the terminal devices of vendor A and vendor B each maintain an encoder model. For different network devices, the encoder models maintained by the terminal devices of vendor A and vendor B are different.

[0182] In one possible implementation, the network server corresponding to the first network device receives encoder models from terminal devices of vendor A and vendor B through the first network device, and trains a decoder model based on the multiple encoder models. The first network server then sends the decoder model to the terminal devices of vendor A and vendor B respectively through the first network device.

[0183] The terminal server corresponding to the terminal device of manufacturer A, that is, the terminal server of manufacturer A, receives decoder models from the first network device and the second network device through the terminal device of manufacturer A, and trains an encoder model based on multiple decoder models.

[0184] In another possible implementation, a first network server receives encoder models from terminal devices of vendor A and vendor B via a first network device, and trains a first decoder model and / or encoder model based on the multiple encoder models. This first decoder model is shared by multiple network devices under the first network server. The first network server then sends the first decoder and / or the encoder model of the first network server to the terminal devices of vendor A and vendor B via the first network device.

[0185] Manufacturer A's terminal server receives encoder models from a first network device and a second network device through Manufacturer A's terminal devices, and trains a first encoder model based on multiple encoder models, or trains a first encoder model and / or a second decoder model based on first decoder models from multiple network devices. The first encoder model is shared by multiple terminal devices under Manufacturer A's terminal server.

[0186] For example, in downlink communication, network devices need to encode messages and terminal devices need to decode messages. The downlink communication model is shown in Figure 8. The first network device and the second network device each maintain an encoder model for UEs from different manufacturers; terminal devices from manufacturer A and terminal devices from manufacturer B each maintain a decoder model for different network devices.

[0187] In one possible implementation, a first network server receives decoder models from terminal devices of vendor A and vendor B via a first network device, and trains an encoder model based on the multiple decoder models. The first network server then sends the encoder model to the terminal devices of vendor A and vendor B respectively via the first network device.

[0188] Manufacturer A's terminal server receives encoder models from the first network device and the second network device through Manufacturer A's terminal equipment, and trains a decoder model based on the multiple encoder models.

[0189] In another possible implementation, a first network server receives decoder models from terminal devices of vendor A and vendor B via a first network device, and trains a second encoder model and / or decoder model based on the multiple decoder models. This second encoder model is shared by multiple network devices under the first network server. The first network server then sends the second encoder and / or the decoder model of the first network server to the terminal devices of vendor A and vendor B via the first network device.

[0190] Manufacturer A's terminal server receives second encoder models from a first network device and a second network device through Manufacturer A's terminal devices, and trains a fourth decoder model based on multiple second encoder models, or trains a fourth decoder model based on decoder models from multiple network devices. The fourth decoder model is a decoder model shared by multiple terminal devices under Manufacturer A's terminal server.

[0191] Optionally, the embodiment shown in Figure 4 further includes step 404. Step 404 may be performed after step 403.

[0192] 404. The first terminal server sends the third model to the first terminal device, and correspondingly, the first terminal device receives the third model from the first terminal server.

[0193] The first terminal server sends the trained third model to the first terminal device. This third model can be an encoder model or a decoder model, as can be found in step 403. The details will not be elaborated here.

[0194] Optionally, the embodiment shown in Figure 4 further includes step 405. Step 405 may be performed after step 404.

[0195] 405. The first terminal device sends a third model to the first network device, and correspondingly, the first network device receives the third model from the first terminal device;

[0196] The first terminal device transmits the third model from the first terminal server to the first network device.

[0197] Optionally, the embodiment shown in Figure 4 further includes step 406. Step 406 may be performed after step 405.

[0198] 406. The first network device sends the third model to the first network server, and correspondingly, the first network server receives the third model from the first network device.

[0199] The first network device can transmit the third model from the first terminal device to the first network server.

[0200] In this embodiment, the first terminal server transmits the model through the terminal device, enabling the training of the model on the terminal server. This allows the terminal server to train a unified model for different network servers, achieving model learning between multiple vendors and reducing the complexity of model management.

[0201] II. The first terminal server obtains the first model from the first network server;

[0202] Please refer to Figure 9. One communication method in this embodiment includes:

[0203] 901. The first terminal server sends a first message to the first terminal device, and correspondingly, the first terminal device receives the first message from the first terminal server;

[0204] The first terminal server sends a first message to the first terminal device. The first message is used to obtain the first model, which is trained by the first network server.

[0205] In one possible implementation, the first message includes the address of a first terminal server, which is used by the first network server to establish a connection with the first terminal server.

[0206] In another possible implementation, the first message includes request information for requesting an address from the first network server; in other words, the first message is used to request the address of the first network server.

[0207] Optionally, the address of the first terminal server or the address of the first network server can be an Internet Protocol (IP) address or a Uniform Resource Locator (URL).

[0208] The other contents of the first message are similar to the first message in step 401 of the embodiment shown in Figure 4. Please refer to the foregoing related descriptions. Specific details will not be repeated here.

[0209] 902. The first terminal server receives the first model from the first network server, and correspondingly, the first network server sends the first model to the first terminal server;

[0210] The first terminal server establishes a connection with the first network server, enabling the first terminal server and the first network server to communicate directly without needing to transfer models through the first terminal device and the first network device.

[0211] In one possible implementation, the first terminal server sends its address to the first network server via a first message. After receiving the address of the first terminal server, the first network server establishes a connection with the first terminal server and sends the first model to the first terminal server through the connection.

[0212] In another possible implementation, the first terminal server requests the address of the first network server from the first network server via a first message. After obtaining the address of the first network server, it establishes a connection with the first network server. The first network server then sends the first model to the first terminal server through this connection.

[0213] It is understandable that the first model received by the first terminal server can be a model trained locally by the first network server, or a model obtained by the first network server training multiple terminal-side models; the specifics are not limited here.

[0214] In this embodiment, the first terminal server interacts with the first network device through the first terminal device to exchange server address information, thereby realizing model transfer between the first terminal server and the first network server, and thus realizing multi-vendor model learning.

[0215] Optionally, the embodiment shown in FIG9 further includes step 900a. Step 900a may be performed before step 901.

[0216] 900a. The first terminal server sends an information reporting instruction to the first terminal device, and the first terminal device receives the information reporting instruction from the first terminal server accordingly.

[0217] Optionally, the embodiment shown in Figure 9 further includes step 900b. Step 900b may be performed before step 901.

[0218] 900b. The first terminal device sends first information to the first terminal server, and correspondingly, the first terminal server receives the first information from the first terminal device.

[0219] In this embodiment, steps 900a to 900b are similar to steps 400b to 400c in the embodiment shown in Figure 4 above, and will not be described in detail here.

[0220] Optionally, the embodiment shown in FIG9 further includes step 901a. Step 901a may be performed after step 901.

[0221] 901a. The first terminal device sends a first message to the first network device, and correspondingly, the first network device receives the first message from the first terminal device;

[0222] Step 901a in this embodiment is similar to step 401a in the embodiment shown in Figure 4 above, and will not be described in detail here.

[0223] Optionally, the embodiment shown in FIG9 further includes step 901b. Step 901b may be performed after step 901.

[0224] 901b. The first network device sends a first message to the first network server, and correspondingly, the first network server receives the first message from the first network device.

[0225] In one possible implementation, the first message includes the address of the first terminal server, and the first network server establishes a connection with the first terminal server based on the address of the first terminal server.

[0226] In another possible implementation, the first message includes request information requesting the address of a first network server. The first network server then sends its address back to the first terminal server based on this request information.

[0227] Optionally, the embodiment shown in FIG9 further includes step 901c. Step 901c may be performed after step 901.

[0228] 901c. The first network server sends a third message to the first network device, and correspondingly, the first network device receives the third message from the first network server.

[0229] The first network device receives a third message from the first network server, which is a response message to the first message, wherein the third message includes the address of the first network server.

[0230] Optionally, the embodiment shown in Figure 9 further includes step 901d. Step 901d may be performed after step 901.

[0231] 901d. The first network device sends a third message to the first terminal device, and correspondingly, the first terminal device receives the third message from the first network device.

[0232] The first network device transmits the third message from the first network server to the first terminal device.

[0233] Optionally, the embodiment shown in FIG9 further includes step 901e. Step 901e may be performed after step 901.

[0234] 901e. The first terminal device sends a third message to the first terminal server, and correspondingly, the first terminal server receives the third message from the first terminal device.

[0235] The first terminal device transmits the third message from the first network device to the first terminal server.

[0236] Optionally, the embodiment shown in FIG9 further includes step 903. Step 903 may be performed after step 902.

[0237] 903. The first terminal server performs model fusion;

[0238] Step 903 in this embodiment is similar to step 403 in the embodiment shown in Figure 4 above, and will not be described in detail here.

[0239] The process by which the first terminal server obtains models from various terminal devices is similar to the process in steps 901 to 902 described above, where the first terminal server obtains the first model from the first network server. Refer to the aforementioned description for further details. For example, the first terminal server receives the first model from the first network server and the second model from the second network server. Then, the first terminal server fuses the first and second models to obtain the third model.

[0240] Optionally, the embodiment shown in FIG9 further includes step 904. Step 904 may be performed after step 903.

[0241] 904. The first terminal server sends the third model to the first network server, and correspondingly, the first network server receives the third model from the first terminal server.

[0242] The first terminal server sends the third model to the first network server through a connection with the first network server.

[0243] In this embodiment of the application, the first terminal server interacts with the first network server to exchange server address information, thereby realizing the model transfer between the first terminal server and the first network server, and thus realizing multi-vendor model learning.

[0244] Third, the third network server obtains the fourth model from the second terminal device;

[0245] Please refer to Figure 10. One communication method in this embodiment includes:

[0246] 1001. The second network device sends a second message to the second terminal device, and correspondingly, the second terminal device receives the second message from the second network device;

[0247] The second network device sends a second message to the second terminal device. The second message is used to request a fourth model, which is trained by the second terminal server.

[0248] Specifically, the second network device selects the second terminal device as the terminal device participating in the model transmission and sends a second message to the second terminal device.

[0249] It should be noted that the cell served by the second network device includes multiple terminal devices, including the second terminal device itself. The second network device selects a subset of these terminal devices as participating terminals in model transmission. Each of these terminal devices belongs to a different terminal server or vendor server. Alternatively, the second network device selects one terminal device from each different vendor among the multiple terminal devices to obtain a subset of terminal devices, which then participate in model transmission. The second network device sends a message to each of these terminal devices, requesting a terminal-side model from the terminal server where that terminal device resides. The number of requested terminal-side models is the same as the number of selected terminal devices.

[0250] Optionally, the second message may include the ID of the fourth model or the ID of the second task, wherein the fourth model is used for the second task.

[0251] Optionally, the second network device can select some terminal devices in turn, or it can select them according to the geographical coordinates of the terminal devices in the cell. The specifics are not limited here.

[0252] Optionally, the second network device can select the terminal device with the best channel quality among multiple terminal devices, or it can randomly select terminal devices; the specific choice is not limited here.

[0253] In practical applications, the second message may also be referred to as the second request or the second model request. This application does not limit the naming of the message.

[0254] 1002. The second network device receives the fourth model from the second terminal device, and correspondingly, the second terminal device sends the fourth model to the second network device.

[0255] The second network server receives the fourth model, which is used to train the fifth model to obtain the sixth model. The fifth model is trained by the third terminal server.

[0256] Specifically, the second network device receives the fourth model transmitted by the second terminal device. This fourth model is then fused with models transmitted by other terminal devices to obtain the sixth model.

[0257] It should be noted that the fourth model is sent from the second terminal server to the second terminal device; that is, the second terminal server sends the fourth model to the second terminal device, and the second terminal device then sends the fourth model to the second network device. Therefore, it can also be said that the fourth model is obtained by the second network device from the second terminal server through the second terminal device.

[0258] In one possible implementation, the selected terminal devices of the second network device include both a second terminal device and a third terminal device, wherein the terminal server hosting the second terminal device is different from the terminal server hosting the third terminal device. Therefore, the terminal-side model corresponding to the second terminal device is different from the terminal-side model corresponding to the third terminal device.

[0259] Optionally, the terminal device receives a fifth model transmitted from a third terminal device, which is used to fuse with the fourth model to obtain a sixth model.

[0260] It should be noted that the fourth model can be carried in a response message sent by the second terminal device to the second network device, and this response message is a response message to the second message. This response message can also be called the second response or the second model response; the naming of this application embodiment is not limited.

[0261] Optionally, the fourth and fifth models can be encoder models, and the sixth model can be a decoder model or an encoder model.

[0262] Optionally, the fourth and fifth models can be decoder models, and the sixth model can be an encoder model or a decoder model.

[0263] In this embodiment, the second network device transmits the model through the terminal device to enable the model training on the network server, so that the network server can train a unified model for different terminal servers, thereby reducing the complexity of model management.

[0264] It is understandable that the fourth model received by the second network device can be a model trained locally by the second terminal server, or a model obtained by the second terminal server training multiple network-side models; the specifics are not limited here.

[0265] Optionally, the embodiment shown in FIG10 further includes step 1000a. Step 1000a may be performed before step 1001.

[0266] 1000a. The second terminal server sends the fourth model to the second terminal device, and correspondingly, the second terminal device receives the fourth model from the second network server.

[0267] The second terminal device can obtain the fourth model from the second terminal server and maintain it.

[0268] Optionally, the embodiment shown in FIG10 further includes step 1000b. Step 1000b may be performed before step 1001.

[0269] 1000b. The second network device sends an information reporting instruction to the second terminal device, and the second terminal device receives the information reporting instruction from the second network device accordingly.

[0270] The second network device sends an information reporting instruction to at least one of a plurality of terminal devices. The information reporting instruction is used to instruct the at least one terminal device to report second information, wherein the at least one terminal device includes the second terminal device.

[0271] Optionally, the second network device sends information reporting instructions to all terminal devices in the serving cell.

[0272] Optionally, the second network device can send information reporting instructions to the terminal device with the best channel quality among multiple terminal devices, or it can randomly select a terminal device to send information reporting instructions to; the specifics are not limited here.

[0273] Optionally, the embodiment shown in FIG10 further includes step 1000c. Step 1000c may be performed before step 1001.

[0274] 1000c. The second terminal device sends second information to the second network device, and correspondingly, the second network device receives the second information from the second terminal device.

[0275] The second network device receives second information from the second terminal device, the second information including at least one of the manufacturer ID of the second terminal device and the model ID of the second terminal device.

[0276] Specifically, the second terminal device responds to the information reporting command of the second network device and sends second information to the second network device.

[0277] Optionally, the second network device can determine a second terminal device from multiple terminal devices as the terminal device to participate in model transmission based on the second information. For example, the second network device selects one terminal device from each vendor's terminal devices for model transmission for each task, application scenario, or configured model.

[0278] Optionally, the embodiment shown in FIG10 further includes step 1003. Step 1003 may be performed after step 1002.

[0279] 1003. The second network device sends the fourth model to the third network server, and the third network server receives the fourth model from the second network device accordingly.

[0280] The second network device reports the fourth model to the third network server, which then processes the fourth model.

[0281] Specifically, the second network device reports the models sent by multiple terminal devices to the third network server, which then performs model fusion.

[0282] Optionally, the embodiment shown in FIG10 further includes step 1004. Step 1004 may be performed after step 1002.

[0283] 1004. The third network server performs model fusion;

[0284] In this embodiment, step 1004 is similar to step 403 in the embodiment shown in Figure 4 above. The action performed by the first terminal server in step 403 is performed by the third network server in step 1004. The specific details will not be repeated here.

[0285] The process by which the second network device obtains the model from each terminal device is similar to the process by which the second network device obtains the fourth model from the second terminal device in steps 1001 to 1003 above, and can be referred to the relevant descriptions above. For example, the second network device receives the fourth model from the second terminal device, and the third network device receives the fifth model from the third terminal device. Then, the third network server fuses the fourth and fifth models to obtain the sixth model.

[0286] Optionally, the embodiment shown in FIG10 further includes step 1005. Step 1005 may be performed after step 1002.

[0287] 1005. The third network server sends the sixth model to the second network device, and correspondingly, the second network device receives the sixth model from the third network server.

[0288] The third network server sends the trained sixth model to the second network device.

[0289] Optionally, the embodiment shown in FIG10 further includes step 1006. Step 1006 may be performed after step 1002.

[0290] 1006. The second network device sends the sixth model to the second terminal device, and the second terminal device receives the sixth model from the second network device accordingly.

[0291] The second network device transmits the sixth model from the third network server to the second terminal device.

[0292] Optionally, the embodiment shown in FIG10 further includes step 1007. Step 1007 may be performed after step 1002.

[0293] 1007. The second terminal device sends the sixth model to the second terminal server, and the second terminal server receives the sixth model from the second terminal device accordingly.

[0294] The second terminal device transmits the sixth model from the second network device to the second terminal server.

[0295] In this embodiment, the second network device transmits the model through the terminal device to enable the model training on the network server, so that the network server can train a unified model for different terminal servers, thereby reducing the complexity of model management.

[0296] Fourth, the third network server obtains the fourth model from the second terminal server;

[0297] Please refer to Figure 11. One communication method in this embodiment includes:

[0298] 1101. The second network device sends a second message to the second terminal device, and correspondingly, the second terminal device receives the second message from the second network device;

[0299] The second network device sends a second message to the second terminal device. The second message is used to obtain the fourth model, which is trained by the second terminal server.

[0300] In one possible implementation, the second message includes the address of a third network server, which is used by the second terminal server to establish a connection with the third network server.

[0301] In another possible implementation, the second message includes request information for requesting the address of a second terminal server, which is used by the third network server to establish a connection with the second terminal server.

[0302] The other contents of the second message are similar to the second message in step 1001 of the embodiment shown in Figure 10, and will not be described in detail here.

[0303] 1102. The third network server receives the fourth model from the second terminal server, and correspondingly, the second terminal server sends the fourth model to the third network server.

[0304] The third network server establishes a connection with the second terminal server, enabling the third network server and the second terminal server to communicate directly without needing to transfer models through the second terminal device and the second network device.

[0305] In one possible implementation, the third network server sends its local address to the second terminal server via a second message. After receiving the address from the third network server, the second terminal server establishes a connection with the third network server and sends the fourth model to the third network server through this connection.

[0306] In another possible implementation, the third network server requests the address of the second terminal server from the second terminal server via the second message. After obtaining the address of the second terminal server, it establishes a connection with the second terminal server. The second terminal server then sends the fourth model to the third network server through this connection.

[0307] It is understandable that the fourth model received by the third network server can be a model trained locally by the second terminal server, or a model obtained by the second terminal server training multiple network-side models; the specifics are not limited here.

[0308] In this embodiment, the third network server interacts with the second network device through the second terminal device to exchange server address information, thereby realizing model transfer between the third network server and the second terminal server, and thus realizing multi-vendor model learning.

[0309] Optionally, the embodiment shown in FIG11 further includes step 1100a. Step 1100a may be performed before step 1101.

[0310] 1100a. The second network device sends an information reporting instruction to the second terminal device, and the second terminal device receives the information reporting instruction from the second network device accordingly.

[0311] Optionally, the embodiment shown in FIG11 further includes step 1100b. Step 1100b may be performed before step 1101.

[0312] 1100b. The second terminal device sends second information to the second network device, and correspondingly, the second network device receives the second information from the second terminal device;

[0313] In this embodiment, steps 1100a to 1100b are similar to steps 1000b to 1000c in the embodiment shown in Figure 10 above, and will not be described in detail here.

[0314] Optionally, the embodiment shown in FIG11 further includes step 1101a. Step 1101a may be performed before step 1102.

[0315] 1101a. The second terminal device sends a second message to the second terminal server, and correspondingly, the second terminal server receives the second message from the second terminal device.

[0316] In one possible implementation, the second message includes the address of the third network server, and the second terminal server establishes a connection with the third network server based on the address of the third network server.

[0317] In another possible implementation, the second message includes a request for the address of a second terminal server. The second terminal server then sends its address back to the third network server based on this request.

[0318] Optionally, the embodiment shown in FIG11 further includes step 1101b. Step 1101b may be performed before step 1102.

[0319] 1101b. The second terminal server sends a fourth message to the second terminal device, and correspondingly, the second terminal device receives the fourth message from the second terminal server.

[0320] The second terminal device receives a fourth message from the second terminal server, which is a response message to the second message, wherein the fourth message includes the address of the second terminal server.

[0321] Optionally, the embodiment shown in FIG11 further includes step 1101c. Step 1101c may be performed before step 1102.

[0322] 1101c. The second terminal device sends a fourth message to the second network device, and correspondingly, the second network device receives the fourth message from the second terminal device.

[0323] The second terminal device transmits the fourth message from the second terminal server to the second network device.

[0324] Optionally, the embodiment shown in FIG11 further includes step 1101d. Step 1101d may be performed before step 1102.

[0325] 1101d. The second network device sends a fourth message to the third network server, and the third network server receives the fourth message from the second network device accordingly.

[0326] The second network device transmits the fourth message from the second terminal device to the second network server.

[0327] Optionally, the embodiment shown in FIG11 further includes step 1103. Step 1103 may be performed after step 1102.

[0328] 1103. The third network server performs model fusion;

[0329] In this embodiment, step 1103 is similar to step 1004 in the embodiment shown in Figure 10 above, and will not be described in detail here.

[0330] The process by which the third network server obtains models from each terminal server is similar to the process in steps 1101 to 1102 above, where the third network server obtains the fourth model from the second terminal server. Refer to the aforementioned description for further details. For example, the third network server receives the fourth model from the second terminal server, and the third network server also receives the fifth model from the third terminal server. Then, the third network server fuses the fourth and fifth models to obtain the sixth model.

[0331] Optionally, the embodiment shown in FIG11 further includes step 1104. Step 1104 may be performed after step 1103.

[0332] 1104. The third network server sends the sixth model to the second terminal server, and correspondingly, the third network server receives the sixth model from the second terminal server.

[0333] The third network server sends the sixth model to the second terminal server through a connection with the second terminal server.

[0334] In this embodiment of the application, the third network server interacts with the second terminal server to exchange server address information, thereby realizing the model transfer between the second terminal server and the third network server, and thus realizing multi-vendor model learning.

[0335] 5. The first terminal server obtains the first model from the fourth terminal device;

[0336] Please refer to Figure 12. One communication method in this embodiment includes:

[0337] 1201. The first terminal server sends a first message to the first terminal device, and correspondingly, the first terminal device receives the first message from the first terminal server;

[0338] 1202. The first terminal server receives the first model from the first terminal device, and correspondingly, the first terminal device sends the first model to the first terminal server;

[0339] In this embodiment, steps 1201 to 1202 are similar to steps 401 to 202 in the embodiment shown in Figure 4 above, and will not be described in detail here.

[0340] Optionally, the embodiment shown in FIG12 further includes step 1200a. Step 1200a may be performed before step 1201.

[0341] 1200a. The fourth terminal server sends the first model to the fourth terminal device, and the fourth terminal device receives the first model from the fourth terminal server accordingly.

[0342] The fourth terminal device can obtain the first model from the fourth terminal server and maintain it.

[0343] Optionally, the embodiment shown in FIG12 further includes step 1200b. Step 1200b may be performed before step 1201.

[0344] 1200b. The first terminal server sends an information reporting instruction to the first terminal device, and the first terminal device receives the information reporting instruction from the first terminal server accordingly.

[0345] Optionally, the embodiment shown in FIG12 further includes step 1200c. Step 1200c may be performed before step 1201.

[0346] 1200c. The first terminal device sends first information to the first terminal server, and correspondingly, the first terminal server receives the first information from the first terminal device.

[0347] In this embodiment, steps 1200b to 1200c are similar to steps 400b to 400c in the embodiment shown in Figure 4 above, and will not be described in detail here.

[0348] Optionally, the embodiment shown in FIG12 further includes step 1201a. Step 1201a may be performed after step 1201.

[0349] 1201a. The first terminal device sends a first message to the fourth terminal device, and correspondingly, the fourth terminal device receives the first message from the first terminal device;

[0350] The first terminal device transmits the first message from the first terminal server to the fourth terminal device in order to request the first model from the fourth terminal device.

[0351] Optionally, the embodiment shown in FIG12 further includes step 1201b. Step 1201b may be performed after step 1201a.

[0352] 1201b. The fourth terminal device sends the first model to the first terminal device, and correspondingly, the first terminal device receives the first model from the fourth terminal device.

[0353] The fourth terminal device responds to the first message by sending the fourth model to the first terminal device, which then transmits the first model to the first terminal server.

[0354] It should be noted that the first model is sent from the fourth terminal server to the fourth terminal device; that is, the fourth terminal server sends the first model to the fourth terminal device, and the fourth terminal device then sends the first model back to the first terminal device. Therefore, it can also be said that the first model is obtained by the first terminal device from the fourth terminal server through the fourth terminal device.

[0355] Optionally, the embodiment shown in FIG12 further includes step 1203. Step 1203 may be performed after step 1202.

[0356] 1203. The first terminal server performs model fusion;

[0357] Optionally, the embodiment shown in FIG12 further includes step 1204. Step 1204 may be performed after step 1203.

[0358] 1204. The first terminal server sends the third model to the first terminal device, and correspondingly, the first terminal device receives the third model from the first terminal server.

[0359] In this embodiment, steps 1203 to 1204 are similar to steps 403 to 404 in the embodiment shown in Figure 4 above, and will not be described in detail here.

[0360] Optionally, the embodiment shown in FIG12 further includes step 1205. Step 1205 may be performed after step 1202.

[0361] 1205. The first terminal device sends the third model to the fourth terminal device, and correspondingly, the fourth terminal device receives the third model from the first terminal device;

[0362] The first terminal device transmits the third model from the first terminal server to the fourth terminal device.

[0363] Optionally, the embodiment shown in FIG12 further includes step 1206. Step 1206 may be performed after step 1205.

[0364] 1206. The fourth terminal device sends the third model to the fourth terminal server, and the fourth terminal server receives the third model from the fourth terminal device accordingly.

[0365] The fourth terminal device can transmit the third model from the first terminal device to the fourth terminal server.

[0366] In this embodiment, the first terminal server transmits the model through the terminal device, enabling model training between terminal servers. This allows the terminal servers to train a unified model for terminal servers from different manufacturers, achieving model learning between multiple manufacturers and reducing the complexity of model management.

[0367] VI. The first terminal server obtains the first model from the fourth terminal server;

[0368] Please refer to Figure 13. One communication method in this embodiment includes:

[0369] 1301. The first terminal server sends a first message to the first terminal device, and correspondingly, the first terminal device receives the first message from the first terminal server;

[0370] 1302. The first terminal server receives the first model from the fourth terminal server, and correspondingly, the fourth terminal server sends the first model to the first terminal server.

[0371] In this embodiment, steps 1301 to 1302 are similar to steps 901 to 902 in the embodiment shown in Figure 9 above, and will not be described in detail here.

[0372] Optionally, the embodiment shown in FIG13 further includes step 1300a. Step 1300a may be performed before step 1301.

[0373] 1300a. The first terminal server sends an information reporting instruction to the first terminal device, and the first terminal device receives the information reporting instruction from the first terminal server accordingly.

[0374] Optionally, the embodiment shown in FIG13 further includes step 1300b. Step 1300b may be performed before step 1301.

[0375] 1300b. The first terminal device sends first information to the first terminal server, and correspondingly, the first terminal server receives the first information from the first terminal device.

[0376] In this embodiment, steps 1300a to 1300b are similar to steps 900a to 900b in the embodiment shown in Figure 9 above, and will not be described in detail here.

[0377] Optionally, the embodiment shown in FIG13 further includes step 1301a. Step 1301a may be performed after step 1301.

[0378] 1301a. The first terminal device sends a first message to the fourth terminal device, and correspondingly, the fourth terminal device receives the first message from the first terminal device.

[0379] Step 1301a in this embodiment is similar to step 1201a in the embodiment shown in Figure 12 above, and will not be described in detail here.

[0380] Optionally, the embodiment shown in FIG13 further includes step 1301b. Step 1301b may be performed after step 1301.

[0381] 1301b. The fourth terminal device sends a first message to the fourth terminal server, and correspondingly, the fourth terminal server receives the first message from the fourth terminal device.

[0382] In one possible implementation, the first message includes the address of the first terminal server, and the fourth terminal server establishes a connection with the first terminal server based on the address of the first terminal server.

[0383] In another possible implementation, the first message includes a request for the address of the fourth terminal server. The fourth terminal server then sends its address back to the first terminal server based on this request.

[0384] 1301c. The fourth terminal server sends a third message to the fourth terminal device, and the fourth terminal device receives the third message from the fourth terminal server accordingly.

[0385] The fourth terminal device receives a third message from the fourth terminal server, which is a response message to the first message, wherein the third message includes the address of the fourth terminal server.

[0386] Optionally, the embodiment shown in FIG13 further includes step 1301d. Step 1301d may be performed after step 1301.

[0387] 1301d. The fourth terminal device sends a third message to the first terminal device, and correspondingly, the first terminal device receives the third message from the fourth terminal device.

[0388] The fourth terminal device transmits the third message from the fourth terminal server to the first terminal device.

[0389] Optionally, the embodiment shown in FIG13 further includes step 1301e. Step 1301e may be performed after step 1301.

[0390] 1301e. The first terminal device sends a third message to the first terminal server, and correspondingly, the first terminal server receives the third message from the first terminal device.

[0391] The first terminal device transmits the third message from the fourth terminal device to the first terminal server.

[0392] Optionally, the embodiment shown in FIG13 further includes step 1303. Step 1303 may be performed after step 1302.

[0393] 1303. The first terminal server performs model fusion;

[0394] Step 1303 in this embodiment is similar to step 403 in the embodiment shown in Figure 4 above, and will not be described in detail here.

[0395] The process by which the first terminal server obtains models from various terminal devices is similar to the process in steps 1301 to 1302 described above, where the first terminal server obtains the first model from the fourth terminal server. Refer to the aforementioned description for further details. For example, the first terminal server receives the first model from the fourth terminal server and the second model from the fifth terminal server. Then, the first terminal server fuses the first and second models to obtain the third model.

[0396] Optionally, the embodiment shown in FIG13 further includes step 1304. Step 1304 may be performed after step 1303.

[0397] 1304. The first terminal server sends the third model to the fourth terminal server, and correspondingly, the fourth terminal server receives the third model from the first terminal server.

[0398] The first terminal server sends the third model to the fourth terminal server through a connection with the fourth terminal server.

[0399] VII. The third network server obtains the fourth model from the third network device;

[0400] Please refer to Figure 14. One communication method in this embodiment includes:

[0401] 1401. The second network device sends a second message to the third network device, and correspondingly, the third network device receives the second message from the second network device;

[0402] 1402. The second network device receives the fourth model from the third network device, and correspondingly, the third network device sends the fourth model to the second network device.

[0403] In this embodiment, steps 1401 to 1402 are similar to steps 1001 to 1002 in the embodiment shown in Figure 10 above, and will not be described in detail here.

[0404] Optionally, the embodiment shown in FIG14 further includes step 1400a. Step 1400a may be performed before step 1401.

[0405] 1400a. The fourth network server sends the fourth model to the third network device, and the third network device receives the fourth model from the fourth network server accordingly.

[0406] The third network device can obtain the fourth model from the fourth network server and maintain it.

[0407] Optionally, the embodiment shown in FIG14 further includes step 1403. Step 1403 may be performed after step 1402.

[0408] 1403. The second network device sends the fourth model to the third network server, and the third network server receives the fourth model from the second network device accordingly.

[0409] Optionally, the embodiment shown in FIG14 further includes step 1404. Step 1404 may be performed after step 1402.

[0410] 1404. The third network server performs model fusion;

[0411] Optionally, the embodiment shown in FIG14 further includes step 1405. Step 1405 may be performed after step 1402.

[0412] 1405. The third network server sends the sixth model to the second network device, and correspondingly, the second network device receives the sixth model from the third network server.

[0413] Optionally, the embodiment shown in FIG14 further includes step 1406. Step 1406 may be performed after step 1402.

[0414] 1406. The second network device sends the sixth model to the third network device, and correspondingly, the third network device receives the sixth model from the second network device.

[0415] Optionally, the embodiment shown in FIG14 further includes step 1407. Step 1407 may be performed after step 1402.

[0416] 1407. The third network device sends the sixth model to the fourth network server, and the fourth network server receives the sixth model from the third network device accordingly.

[0417] In this embodiment, steps 1403 to 1407 are similar to steps 1003 to 1007 in the embodiment shown in Figure 10 above, and will not be described in detail here.

[0418] 8. The third network server obtains the fourth model from the fourth network server;

[0419] Please refer to Figure 15. One communication method in this embodiment includes:

[0420] 1501. The second network device sends a second message to the third network device, and correspondingly, the third network device receives the second message from the second network device;

[0421] 1502. The third network server receives the fourth model from the fourth network server, and correspondingly, the fourth network server sends the fourth model to the third network server.

[0422] In this embodiment, steps 1501 to 1502 are similar to steps 1101 to 1102 in the embodiment shown in Figure 11 above, and will not be described in detail here.

[0423] Optionally, the embodiment shown in FIG15 further includes step 1500a. Step 1500a may be performed before step 1501.

[0424] Optionally, the embodiment shown in FIG15 further includes step 1501a. Step 1501a may be performed before step 1502.

[0425] 1501a. The third network device sends a second message to the fourth network server, and the fourth network server receives the second message from the third network device accordingly.

[0426] Optionally, the embodiment shown in FIG15 further includes step 1501b. Step 1501b may be performed before step 1502.

[0427] 1501b. The fourth network server sends a fourth message to the third network device, and the third network device receives the fourth message from the fourth network server accordingly.

[0428] Optionally, the embodiment shown in FIG15 further includes step 1501c. Step 1501c may be performed before step 1502.

[0429] 1501c. The third network device sends a fourth message to the second network device, and correspondingly, the second network device receives the fourth message from the third network device.

[0430] Optionally, the embodiment shown in FIG15 further includes step 1501d. Step 1501d may be performed before step 1502.

[0431] 1501d. The second network device sends a fourth message to the third network server, and the third network server receives the fourth message from the second network device accordingly.

[0432] In this embodiment, steps 1501a to 1501d are similar to steps 1101a to 1101d in the embodiment shown in Figure 11 above, and will not be described in detail here.

[0433] Optionally, the embodiment shown in FIG11 further includes step 1103. Step 1103 may be performed after step 1102.

[0434] 1103. The third network server performs model fusion;

[0435] Optionally, the embodiment shown in FIG11 further includes step 1104. Step 1104 may be performed after step 1103.

[0436] 1104. The third network server sends the sixth model to the fourth network server, and correspondingly, the fourth network server receives the sixth model from the third network server.

[0437] In this embodiment, steps 1503 to 1504 are similar to steps 1103 to 1104 in the embodiment shown in Figure 11 above, and will not be described in detail here.

[0438] The information transmission method in the embodiments of this application has been described above. The communication device in the embodiments of this application is described below. Referring to Figure 16, the communication device 1600 can be used to execute the process performed by the terminal server in the above embodiments. For details, please refer to the relevant descriptions in the foregoing method embodiments. The communication device 1600 can be a server, a component or device applied to a server (e.g., a processor, chip, or chip system), or a logic module or software capable of implementing all or part of the server's functions.

[0439] The communication device 1600 includes an interface module 1601 and a processing module 1602.

[0440] The processing module 1602 is used for data processing. The interface module 1601 can implement corresponding communication functions. The interface module 1601 can also be called a communication interface or a communication module.

[0441] Optionally, the communication device 1600 may further include a storage module, which can be used to store program code, program instructions and / or data. The processing module 1602 can read the instructions and / or data in the storage module so that the communication device 1600 can implement the aforementioned method embodiments.

[0442] The communication device 1600 can be used to perform the actions performed by the terminal server in the above method embodiments. For example, it can be a terminal server, a communication module within a terminal server, or a circuit or chip within a terminal server responsible for communication functions. The communication device 1600 can be a terminal server or a component configurable on a terminal server. The processing module 1602 is used to perform processing-related operations on the terminal server side in the above method embodiments. The interface module 1601 is used to perform receiving-related operations on the terminal server side in the above method embodiments.

[0443] Optionally, interface module 1601 may include a sending module and a receiving module. The sending module is used to perform the sending operation in the above method embodiments. The receiving module is used to perform the receiving operation in the above method embodiments.

[0444] It should be noted that the communication device 1600 may include a transmitting module but not a receiving module. Alternatively, the communication device 1600 may include a receiving module but not a transmitting module. Specifically, it depends on whether the above-described scheme executed by the communication device 1600 includes both transmitting and receiving actions. For example, the communication device 1600 is used to execute the actions performed by the terminal server in the embodiments shown in Figures 4, 9, or 11. For details, please refer to the relevant descriptions in the embodiments shown in Figures 4, 9, or 11; these will not be elaborated upon here.

[0445] For example, the communication device 1600 is used to execute the following scheme:

[0446] Processing module 1602 is used to generate the first message;

[0447] Interface module 1601 is used to send a first message to the first terminal device. The first message is used to obtain a first model. The first model is trained by the first network server.

[0448] The interface module 1601 is also used to receive a first model, which is used to train a third model with a second model. The second model is trained by a second network server.

[0449] For example, the communication device 1600 is used to execute the following scheme:

[0450] Processing module 1602 is used to generate the first message;

[0451] Interface module 1601 is used to send a first message to the first terminal device. The first message is used to obtain the first model, which is trained by the fourth terminal server.

[0452] The interface module 1601 is also used to receive the first model, which is used to train the third model with the second model. The second model is trained by the fifth terminal server.

[0453] In one possible implementation, the interface module 1601 is further configured to receive first information from the first terminal device, the first information including at least one of the ID of the cell where the first terminal device is located, the ID of the manufacturer of the first terminal device, and the ID of the model of the first terminal device;

[0454] The interface module 1601 is specifically used to send a first message to the first terminal device based on the first information.

[0455] In another possible implementation, the first message includes the ID of the first model or the ID of the first task, with the first model used for the first task.

[0456] In another possible implementation, the processing module 1602 is also used to train a third model based on the first and second models.

[0457] In another possible implementation, the interface module 1601 is specifically used to receive a first model from the first terminal device, which is obtained by the first terminal device from the first network server through the first network device, or from the fourth terminal server through the fourth terminal device.

[0458] In another possible implementation, interface module 1601 is also used to send a third model to the first terminal device.

[0459] In another possible implementation, the first message also includes the address of the first terminal server, which is used by the first network server to establish a connection with the first terminal server, or by the fourth terminal server to establish a connection with the first terminal server.

[0460] The interface module 1601 is specifically used to receive the first model from the first network server or the fourth terminal server.

[0461] In another possible implementation, interface module 1601 is also used to send a third model to the first network server.

[0462] It should be understood that the specific procedures for each module to perform the above-mentioned corresponding processes have been described in detail in the above method embodiments, and will not be repeated here for the sake of brevity.

[0463] The processing module 1602 in the above embodiments can be implemented by at least one processor or processor-related circuitry. The interface module 1601 can be implemented by a transceiver or transceiver-related circuitry. The interface module 1601 can also be referred to as a communication module or communication interface. The storage module can be implemented by at least one memory.

[0464] The following is another structural schematic diagram of the communication device according to an embodiment of this application. Referring to Figure 17, the communication device can be used to execute the process performed by the terminal device in the above embodiments, and for details, please refer to the relevant description in the foregoing method embodiments.

[0465] The communication device 1700 includes an interface module 1701. Optionally, a processing module 1702.

[0466] The processing module 1702 is used for data processing. The interface module 1701 can implement corresponding communication functions. The interface module 1701 can also be called a communication interface or a communication module.

[0467] Optionally, the communication device 1700 may further include a storage module, which can be used to store program code, program instructions and / or data. The processing module 1702 can read the instructions and / or data in the storage module so that the communication device 1700 can implement the aforementioned method embodiments.

[0468] The communication device 1700 can be used to perform the actions performed by the terminal device in the above method embodiments. For example, it can be a terminal device, a communication module within a terminal device, or a circuit or chip within a terminal device responsible for communication functions. The communication device 1700 can be a terminal device or a component configurable on a terminal device. The processing module 1702 is used to perform processing-related operations on the terminal device side in the above method embodiments. The interface module 1701 is used to perform reception-related operations on the terminal device side in the above method embodiments.

[0469] Optionally, interface module 1701 may include a sending module and a receiving module. The sending module is used to perform the sending operation in the above method embodiments. The receiving module is used to perform the receiving operation in the above method embodiments.

[0470] It should be noted that the communication device 1700 may include a transmitting module but not a receiving module. Alternatively, the communication device 1700 may include a receiving module but not a transmitting module. Specifically, it depends on whether the above-described scheme performed by the communication device 1700 includes both transmitting and receiving actions. For example, the communication device 1700 is used to perform the actions performed by the terminal device in the embodiments shown in Figures 4, 9, 10, or 11. For details, please refer to the relevant descriptions in the above embodiments; they will not be elaborated upon here.

[0471] For example, the communication device 1700 is used to execute the following scheme:

[0472] Interface module 1701 receives a first message from the first terminal server. The first message is used to obtain the first model, which is trained by the first network server.

[0473] Interface module 1701 is also used to send a first model to a first terminal server, wherein the first model is obtained from the first network server through a first network device;

[0474] Processing module 1702 is used to maintain the first model.

[0475] In one possible implementation, the interface module 1701 is further configured to send first information to the first terminal server, the first information including at least one of the ID of the cell where the first terminal device is located, the ID of the manufacturer of the first terminal device, or the ID of the model of the first terminal device.

[0476] In another possible implementation, the first message includes the ID of the first model or the ID of the first task, with the first model used for the first task.

[0477] It should be understood that the specific procedures for each module to perform the above-mentioned corresponding processes have been described in detail in the above method embodiments, and will not be repeated here for the sake of brevity.

[0478] Optionally, when the communication device 1700 is a terminal device or a communication module within a terminal device, the processing module 1702 in the above embodiments can be implemented by at least one processor or processor-related circuitry. Specifically, the processor may include a modem chip, or a system-on-chip (SoC) chip containing a modem core, or a system-in-package (SIP) chip. The interface module 1701 can be implemented by a transceiver or transceiver-related circuitry. The interface module 1701 may also be referred to as a communication module or communication interface. The storage module can be implemented by at least one memory.

[0479] Optionally, when the communication device 1700 is a circuit or chip in a terminal device responsible for communication functions, such as a modem chip or a SoC chip or SIP chip containing a modem core, the function of the processing module 1702 can be implemented by a circuit system in the aforementioned chip that includes one or more processors or processing cores. The function of the interface module 1701 can be implemented by the interface circuit or data transceiver circuit on the aforementioned chip.

[0480] The following is another schematic diagram of the communication device according to an embodiment of this application. Referring to FIG18, the communication device can be used to execute the process performed by the network server in the embodiments shown in FIG9, FIG10 or FIG11. For details, please refer to the relevant description in the foregoing method embodiments.

[0481] The communication device 1800 includes an interface module 1801. Optionally, a processing module 1802.

[0482] The processing module 1802 is used for data processing. The interface module 1801 can implement corresponding communication functions. The interface module 1801 can also be called a communication interface or a communication module.

[0483] Optionally, the communication device 1800 may further include a storage module, which can be used to store program code, program instructions and / or data. The processing module 1802 can read the instructions and / or data in the storage module so that the communication device 1800 can implement the aforementioned method embodiments.

[0484] The communication device 1800 can be used to perform the actions performed by the network server in the above method embodiments. For example, it can be a network server, a communication module within a network server, or a circuit or chip within a network server responsible for communication functions. The communication device 1800 can be a network server or a component configurable on a network server. The processing module 1802 is used to perform processing-related operations on the network server side in the above method embodiments. The interface module 1801 is used to perform receiving-related operations on the network server side in the above method embodiments.

[0485] Optionally, the interface module 1801 may include a sending module and a receiving module. The sending module is used to perform the sending operation in the above method embodiments. The receiving module is used to perform the receiving operation in the above method embodiments.

[0486] It should be noted that the communication device 1800 may include a transmitting module but not a receiving module. Alternatively, the communication device 1800 may include a receiving module but not a transmitting module. Specifically, it depends on whether the above-described scheme executed by the communication device 1800 includes both transmitting and receiving actions. For example, the communication device 1800 is used to execute the actions performed by the network server in the above embodiments. For details, please refer to the relevant descriptions in the above embodiments; they will not be elaborated upon here.

[0487] For example, the communication device 1800 is used to execute the following scheme:

[0488] Interface module 1801 is used to receive the fourth model, which is trained by the second terminal server. The fourth model is used to train the sixth model with the fifth model, which is trained by the third terminal server.

[0489] Processing module 1802 is used to train a sixth model based on the fourth and fifth models.

[0490] For example, the communication device 1800 is used to execute the following scheme:

[0491] Interface module 1801 is used to receive the fourth model, which is trained by the fourth network server. The fourth model is used to train the sixth model with the fifth model, which is trained by the fifth network server.

[0492] Processing module 1802 is used to train a sixth model based on the fourth and fifth models.

[0493] In one possible implementation, the interface module 1801 is specifically used to receive a fourth model from the second network device, which is obtained by the second network device from the second terminal server through the second terminal device, or from the fourth network server through the third network device.

[0494] In another possible implementation, interface module 1801 is also used to receive a fifth model from the second network device, which is obtained by the second network device from the third terminal server through the third terminal device, or from the fifth network server through the fourth network device.

[0495] In another possible implementation, interface module 1801 is specifically used to receive a fourth model from a second terminal server or a fourth network server.

[0496] In another possible implementation, interface module 1801 is also used to receive a fifth model from a third terminal server or a fifth network server.

[0497] It should be understood that the specific procedures for each module to perform the above-mentioned corresponding processes have been described in detail in the above method embodiments, and will not be repeated here for the sake of brevity.

[0498] The processing module 1802 in the above embodiments can be implemented by at least one processor or processor-related circuitry. The interface module 1801 can be implemented by a transceiver or transceiver-related circuitry. The interface module 1801 can also be referred to as a communication module or communication interface. The storage module can be implemented by at least one memory.

[0499] The following is another structural schematic diagram of the communication device according to an embodiment of this application. Referring to FIG19, the communication device can be used to execute the process performed by the network device in the embodiments shown in FIG4, FIG9, FIG10 and FIG11. For details, please refer to the relevant description in the foregoing method embodiments.

[0500] The communication device 1900 includes an interface module 1901. Optionally, a processing module 1902.

[0501] The processing module 1902 is used for data processing. The interface module 1901 can implement corresponding communication functions. The interface module 1901 can also be called a communication interface or a communication module.

[0502] Optionally, the communication device 1900 may further include a storage module, which can be used to store program code, program instructions and / or data. The processing module 1902 can read the instructions and / or data in the storage module so that the communication device 1900 can implement the aforementioned method embodiments.

[0503] The communication device 1900 can be used to perform the actions performed by the network device in the above method embodiments. For example, it can be a network device or a communication module within a network device, or a circuit or chip within a network device responsible for communication functions. The communication device 1900 can be a network device or a component configurable within a network device. The processing module 1902 is used to perform processing-related operations on the network device side in the above method embodiments. The interface module 1901 is used to perform reception-related operations on the network device side in the above method embodiments.

[0504] Optionally, interface module 1901 may include a sending module and a receiving module. The sending module is used to perform the sending operation in the above method embodiments. The receiving module is used to perform the receiving operation in the above method embodiments.

[0505] It should be noted that the communication device 1900 may include a transmitting module but not a receiving module. Alternatively, the communication device 1900 may include a receiving module but not a transmitting module. Specifically, it depends on whether the above-described scheme performed by the communication device 1900 includes both transmitting and receiving actions. For example, the communication device 1900 is used to perform the actions performed by the network device in the embodiments shown in Figures 9, 10, or 11. For details, please refer to the relevant descriptions in the embodiments shown in Figures 9, 10, or 11; these will not be elaborated upon here.

[0506] For example, the communication device 1900 is used to execute the following scheme:

[0507] Processing module 1902 is used to generate a second message, which requests a fourth model, which is trained by a second terminal server or a fourth network server.

[0508] Interface module 1901 is used to send a second message to a second terminal device.

[0509] In one possible implementation, the interface module 1901 is further configured to receive second information from the second terminal device, the second information including at least one of the cell identifier ID of the second terminal device, the manufacturer ID of the second terminal device, or the model ID of the second terminal device.

[0510] The interface module 1901 is specifically used to send a second message to the second terminal device based on the second information.

[0511] In another possible implementation, the second message includes the ID of the fourth model or the ID of the second task, with the fourth model used for the second task.

[0512] In another possible implementation, the interface module 1901 is specifically used to receive a fourth model from the second terminal device, which is obtained by the second terminal device from the second terminal server.

[0513] or,

[0514] Receive the fourth model from the third network device, which is obtained by the third network device from the fourth network server.

[0515] In another possible implementation, the second message also includes the address of a third network server, which is used by the second terminal server to establish a connection with the third network server, or by the fourth network server to establish a connection with the third network server.

[0516] It should be understood that the specific procedures for each module to perform the above-mentioned corresponding processes have been described in detail in the above method embodiments, and will not be repeated here for the sake of brevity.

[0517] The processing module 1902 in the above embodiments can be implemented by at least one processor or processor-related circuitry. The interface module 1901 can be implemented by a transceiver or transceiver-related circuitry. The interface module 1901 can also be referred to as a communication module or communication interface. The storage module can be implemented by at least one memory.

[0518] The following describes a communication device provided in an embodiment of this application. Please refer to Figure 20, which is a schematic diagram of the structure of a communication device provided in an embodiment of this application. The communication device may be a terminal server, network server, terminal device, or network device as described in the above method embodiments, or it may be a chip, chip system, or processor that supports the terminal server, network server, terminal device, or network device in implementing the above methods. This communication device can be used to implement the methods described in the above method embodiments, and for details, please refer to the description in the above method embodiments.

[0519] The communication device may include one or more processors 2001, which are connected to a memory 2002, an input / output unit 2003, and a bus 2004. The processor 2001 may be a general-purpose processor or a dedicated processor, such as a baseband processor or a central processing unit (CPU). The baseband processor can be used to process communication protocols and communication data, while the CPU can be used to control the communication device (e.g., base station, baseband chip, terminal, terminal chip, DU or CU, etc.), execute software programs, and process data from the software programs.

[0520] Optionally, the communication device may include one or more memories 2002, which may store instructions that can be executed on the processor 2001 to cause the communication device to perform the methods described in the above method embodiments. Optionally, the memories 2002 may also store data. The processor 2001 and the memories 2002 may be provided separately or integrated together.

[0521] Optionally, the communication device may also include a transceiver and an antenna. A transceiver, also called a transceiver unit, transceiver, or transceiver circuit, is used to implement transmission and reception functions. A transceiver may include a receiver and a transmitter; the receiver, also called a receiver circuit, is used to implement the receiving function; the transmitter, also called a transmitter or transmitting circuit, is used to implement the transmitting function.

[0522] In another possible design, the processor 2001 may include a transceiver for implementing receive and transmit functions. For example, the transceiver may be a transceiver circuit, an interface, or an interface circuit. The transceiver circuit, interface, or interface circuit for implementing receive and transmit functions may be separate or integrated. The aforementioned transceiver circuit, interface, or interface circuit may be used for reading and writing code / data, or for transmitting or relaying signals.

[0523] In another possible design, the processor 2001 may optionally store instructions that, when executed, cause the communication device to perform the methods described in the above method embodiments. The instructions may be stored in the processor 2001; in this case, the processor 2001 may be implemented in hardware.

[0524] In another possible design, the communication device may include a circuit that can perform the sending or receiving or communication functions of the terminal server, network server, terminal device, or network device in the aforementioned method embodiments. The processor and transceiver described in this application embodiment can be implemented on integrated circuits (ICs), analog ICs, radio frequency integrated circuits (RFICs), mixed-signal ICs, application-specific integrated circuits (ASICs), printed circuit boards (PCBs), electronic devices, etc. The processor and transceiver can also be manufactured using various IC process technologies, such as complementary metal oxide semiconductors (CMOS), n-type metal-oxide-semiconductor (NMOS), p-type metal oxide semiconductors (PMOS), bipolar junction transistors (BJTs), bipolar CMOS (BiCMOS), silicon germanium (SiGe), gallium arsenide (GaAs), etc.

[0525] The communication device described in the above embodiments may be a terminal server, a network server, a terminal device, or a network device. However, the scope of the communication device described in the embodiments of this application is not limited to this, and the structure of the communication device may not be limited to FIG. 20. The communication device may be a standalone device or may be part of a larger device. For example, the communication device may be:

[0526] (1) Independent integrated circuit IC, or chip, or chip system or subsystem;

[0527] (2) A collection of one or more ICs, optionally including a storage component for storing data and instructions;

[0528] (3) ASIC, such as modem;

[0529] (4) Modules that can be embedded in other devices;

[0530] (5) Receivers, terminals, smart terminals, cellular phones, wireless devices, handheld devices, mobile units, vehicle-mounted devices, network devices, cloud devices, artificial intelligence devices, etc.

[0531] (6) Others, etc.

[0532] For communication devices that can be chips or chip systems, please refer to the schematic diagram of the chip structure shown in Figure 21. The chip 2100 shown in Figure 21 includes a processor 2101 and an interface 2102. Optionally, it may also include a memory 2103. The number of processors 2101 can be one or more, and the number of interfaces 2102 can be multiple.

[0533] For cases where the chip is used to implement the functions of the network device or the first device in the embodiments of this application:

[0534] The interface 2102 is used to receive or output signals;

[0535] The processor 2101 is used to perform data processing operations of network devices or terminal devices.

[0536] It is understood that some optional features in the embodiments of this application can be implemented independently in certain scenarios without relying on other features, such as the current solution on which they are based, to solve the corresponding technical problems and achieve the corresponding effects. Alternatively, they can be combined with other features as needed in certain scenarios. Correspondingly, the communication device given in the embodiments of this application can also implement these features or functions, which will not be elaborated here.

[0537] It should be understood that the processor in the embodiments of this application can be an integrated circuit chip with signal processing capabilities. In implementation, the steps of the above method embodiments can be completed by integrated logic circuits in the processor's hardware or by instructions in software form. The processor described above can be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0538] It is understood that the memory in the embodiments of this application can be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of RAK are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous linked dynamic random access memory (SLDRAM), and direct rambus RAM (DR RAM). It should be noted that the memory used in the systems and methods described herein is intended to include, but is not limited to, these and any other suitable types of memory.

[0539] This application also provides a computer-readable storage medium including instructions that, when executed on a computer, cause the computer to perform the methods described in the foregoing embodiments.

[0540] This application also provides a computer program product containing instructions that, when run on a computer, cause the computer to perform the methods described in the foregoing embodiments.

[0541] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0542] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection between apparatuses or units through some interfaces, and may be electrical, mechanical, or other forms.

[0543] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0544] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0545] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0546] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions. When the computer instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible to a computer or a data storage device such as a server or data center that integrates one or more available media. The available media may be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., high-density digital video discs (DVDs)), or semiconductor media (e.g., solid-state disks (SSDs)).

Claims

1. A communication method characterized by comprising: The method comprises: sending a first message to a first terminal device, the first message being used to obtain a first model, the first model being trained by a first network server; receiving the first model, the first model being used to train a third model with a second model, the second model being trained by a second network server.

2. The method of claim 1, wherein, The method further comprises: receiving first information from the first terminal device, the first information comprising at least one of an ID of a cell where the first terminal device is located, an ID of a vendor of the first terminal device, and an ID of a model of the first terminal device; The sending of the first message to the first terminal device comprises: sending the first message to the first terminal device according to the first information.

3. The method according to claim 1 or 2, characterized in that, The first message comprises an ID of the first model or an ID of a first task, the first model being used for the first task.

4. The method according to any one of claims 1 to 3, characterized in that, The method further comprises: training the third model according to the first model and the second model.

5. The method according to any one of claims 1 to 4, characterized in that, The receiving of the first model comprises: receiving the first model from the first terminal device, the first model being obtained by the first terminal device from the first network server through a first network device.

6. The method of claim 5, wherein, The method further comprises: sending the third model to the first terminal device.

7. The method according to any one of claims 1 to 4, characterized in that, The first message further comprises an address of a first terminal server, the address of the first terminal server being used for the first network server to establish a connection with the first terminal server. The receiving of the first model comprises: receiving the first model from the first network server.

8. The method of claim 7, wherein, The method further comprises: sending the third model to the first network server.

9. A communication method characterized by comprising: The method comprises: receiving a first message from a first terminal server, the first message being used to obtain a first model, the first model being trained by a first network server; sending the first model to the first terminal server, the first model being obtained from the first network server through a first network device.

10. The method of claim 9, wherein, The method further comprises: sending first information to the first terminal server, the first information comprising at least one of an ID of a cell where the first terminal device is located, an ID of a vendor of the first terminal device, or an ID of a model of the first terminal device.

11. The method according to claim 9 or 10, characterized in that, The first message comprises an ID of the first model or an ID of a first task, the first model being used for the first task.

12. A communication method characterized by comprising: The method comprises: receiving a fourth model, the fourth model being trained by a second terminal server, the fourth model being used to train a sixth model with a fifth model, the fifth model being trained by a third terminal server; training the sixth model according to the fourth model and the fifth model.

13. The method of claim 12, wherein, The receiving of the fourth model comprises: receiving the fourth model from a second network device, the fourth model being obtained by the second network device from the second terminal server through a second terminal device.

14. The method of claim 13, wherein, The method further comprises: receiving the fifth model from the second network device, the fifth model being obtained by the second network device from the third terminal server through a third terminal device.

15. The method of claim 12, wherein, The receiving of the fourth model comprises: receiving the fourth model from the second terminal server.

16. The method of claim 15, wherein, The method further includes: receiving the fifth model from the third terminal server.

17. A method of communication, comprising: comprising: generating a second message, the second message being used for requesting a fourth model, the fourth model being trained by a second terminal server; sending the second message to a second terminal device.

18. The method of claim 17, wherein, The method further includes: receiving second information from the second terminal device, the second information comprising at least one of an identification ID of a cell where the second terminal device is located, an ID of a vendor of the second terminal device, or an ID of a model of the second terminal device; The sending the second message to a second terminal device comprises: sending the second message to the second terminal device according to the second information.

19. The method of claim 17 or 18, wherein, The second message comprises an ID of the fourth model or an ID of a second task, the fourth model being used for the second task.

20. The method of any one of claims 17-19, wherein, The method further includes: receiving the fourth model from the second terminal device, the fourth model being obtained by the second terminal device from the second terminal server.

21. The method of any one of claims 17-19, wherein, The second message further comprises an address of a third network server, the address of the third network server being used for the fourth terminal server to establish a connection with the third network server.

22. A communications device, characterized by comprising: a processing module, configured to generate a first message; an interface module, configured to send the first message to a first terminal device, the first message being used for obtaining a first model, the first model being trained by a first network server; The interface module is further configured to receive the first model, the first model being used for training a third model with a second model, the second model being trained by a second network server.

23. A communications device, characterized by comprising: an interface module, configured to receive a first message from a first terminal server, the first message being used for obtaining a first model, the first model being trained by a first network server; The interface module is further configured to send the first model to the first terminal server, the first model being obtained by a first network device from the first network server; a processing module, configured to maintain the first model.

24. A communications device, characterized by comprising: an interface module, configured to receive a fourth model, the fourth model being trained by a second terminal server, the fourth model being used for training a sixth model with a fifth model, the fifth model being trained by a third terminal server; a processing module, configured to train the sixth model according to the fourth model and the fifth model.

25. A communications device, characterized by comprising: a processing module, configured to generate a second message, the second message being used for requesting a fourth model, the fourth model being trained by a second terminal server; an interface module, configured to send the second message to a second terminal device.

26. A communications device, characterized by The apparatus comprises modules for performing the method of any one of claims 1 to 8, or the apparatus comprises modules for performing the method of any one of claims 9 to 11, or the apparatus comprises modules for performing the method of any one of claims 12 to 16, or the apparatus comprises modules for performing the method of any one of claims 17 to 21.

27. A communications device, characterized by comprising: a processor configured to execute a program, causing the communication device to perform the method of any one of claims 1 to 8, or causing the communication device to perform the method of any one of claims 9 to 11, or causing the communication device to perform the method of any one of claims 12 to 16, or causing the communication device to perform the method of any one of claims 17 to 21.

28. The communication apparatus according to claim 27, wherein, Further comprising a memory storing the computer program or instructions.

29. The communication apparatus according to claim 27 or 28, wherein, The communication device is a chip or chip system.

30. A communication system, characterized by comprising: a communication device configured to perform the method of any one of steps 1 to 8, a communication device configured to perform the method of any one of steps 9 to 11, a communication device configured to perform the method of any one of steps 12 to 16, and a communication device configured to perform the method of any one of claims 17 to 21.

31. A computer readable storage medium comprising instructions which, when executed on a computer, cause the computer to perform the method of any one of claims 1 to 8, or cause the computer to perform the method of any one of claims 9 to 11, or cause the computer to perform the method of any one of claims 12 to 16, or cause the computer to perform the method of any one of claims 17 to 21.

32. A computer program product comprising instructions which, when executed on a computer, cause the computer to perform the method of any one of claims 1 to 8, or cause the computer to perform the method of any one of claims 9 to 11, or cause the computer to perform the method of any one of claims 12 to 16, or cause the computer to perform the method of any one of claims 17 to 21.

Images