Communication method and related device

Abstract

This application provides a communication method and related apparatus. The communication method may be used for a first processing entity of a first communication device. The method includes: the first processing entity sends a first registration request to a second entity, the first registration request includes a first identifier, the first identifier indicates one or more of the following information of the first processing entity: model information or data information, and the model parameters of the second entity are adjusted based on real-time environmental parameters; the first processing entity receives a first registration response sent by the second entity, the first registration response indicating input parameters of the second entity; the first processing entity sends a first response acknowledgment, the first response acknowledgment indicating whether the output parameters of the first processing entity match the input parameters of the second entity; the method can adapt to environmental changes and support effective data reception and transmission.

Description

[Technical Field]

[0001] [CROSS-REFERENCE TO RELATED APPLICATIONS] This application claims priority to Chinese Patent Application No. 202210283863.9, entitled "COMMUNICATION METHOD AND RELATED APPARATUS," filed with the State Intellectual Property Office of the People's Republic of China on March 21, 2022, the entire contents of which are incorporated by reference.

[0002] [Technical field] This application relates to the field of communications technology, and more particularly to communications methods and related devices. [Background technology]

[0003] Artificial intelligence (AI) technology plays an important role in promoting the development of mobile communication network technology. For example, there are related researches on applying AI technology to the network layer (network optimization, mobility management and resource allocation, etc.), the physical layer (channel coding and decoding, channel prediction and receiver, etc.), etc.

[0004] The current combination of AI and communication is mainly to design an AI model or training method for given system parameters in a specific scenario. When the environment, such as device capabilities, scenarios (e.g., channel conditions), or system parameters, changes, the corresponding model needs to be redesigned and retrained, making it difficult to deploy and use the combination of AI and communication in a universal communication system. Summary of the Invention

[0005] SUMMARY OF THE INVENTION Embodiments of this application provide a communication method and related apparatus for adapting to environmental changes and supporting efficient data reception and transmission.

[0006] According to a first aspect, an embodiment of the present application provides a communication method, which is used in a first processing entity of a first communication device, and includes: the first processing entity sends a first registration request to a second entity, the first registration request including a first identifier, the first identifier indicating one or more of the following information of the first processing entity: model information or data information, and model parameters of the second entity are adjusted based on real-time environmental parameters; the first processing entity receives a first registration response sent by the second entity, the first registration response indicating input parameters of the second entity; the first processing entity sends a first response confirmation response, the first response confirmation response indicating whether output parameters of the first processing entity match the input parameters of the second entity;

[0007] Optionally, the second entity may also be referred to as a self-adaptive entity.

[0008] In the method, a first processing entity performs registration with a second entity to determine that output parameters of the first processing entity match input parameters of the second entity, and model parameters of the second entity may be adjusted based on real-time environmental parameters, which can adapt to environmental changes and support effective data reception and transmission.

[0009] In a possible implementation, the first response acknowledgment indicates a successful match, and the method further includes: a first processing entity performs a first operation on the first input data to obtain first output data, the first output data being an output of the first processing entity; and the first processing entity transmits the first output data to the second entity through the first interface.

[0010] In this implementation, when the output parameters of the first processing entity are successfully matched with the input parameters of the second entity, the first processing entity transmits the first output data to the second entity, thereby realizing effective data transmission.

[0011] In a possible implementation, the first processing entity includes an input adaptation module, a data space transformation module, and an output adaptation module.

[0012] In this implementation, the input adaptation module and output adaptation module of the first processing entity perform processing such as dimensional transformation on the input data and output data of the first processing entity, respectively, to adapt the output dimensions of the pre-processing module of the first processing entity to the input dimensions of the data space transformation module, and to adapt the output dimensions of the first processing entity to the input dimensions of the subsequent processing module (e.g., the second entity) of the first processing entity.

[0013] In a possible implementation, the first processing comprises one or more of the following: quantization, source coding, channel coding, modulation and MIMO precoding, or processing equivalent to one or more of the following: quantization, source coding, channel coding, modulation and MIMO precoding.

[0014] In this implementation, one or more operations performed in conventional data reception and transmission may be implemented via the first processing entity, which makes it easier to achieve global optimization when implementing multiple operations compared to implementing each operation module separately in conventional links.

[0015] In a possible implementation, the first processing includes encoding and modulation, the first input data being a bit stream, and the first output data being modulation symbols.

[0016] In a possible implementation, the first processing includes quantization, source coding, channel coding and modulation, the first input data is a symbol sequence, and the first output data is a modulated symbol sequence.

[0017] From the above possible implementation manners, it can be seen that all first processing entities realizing different functions can be registered with a second entity compatible with the first processing entity to realize effective data reception and transmission.

[0018] In a possible implementation, the first interface defines a procedure and / or signaling for interaction between the first processing entity and the second entity, for example, the first interface defines a transmission process for one or more of a registration request, a registration response, or a response acknowledgment.

[0019] In a possible implementation manner, the first response acknowledgment indicates that the output parameters of the first processing entity are incompatible with the input parameters of the second entity, and the method further includes: the first processing entity sends a second registration request to the fourth entity, the second registration request including a first identifier, the first identifier indicating one or more of the following information of the first processing entity: model information or data information, and the model parameters of the fourth entity may be adjusted based on the real-time environmental parameters; the first processing entity receives a second registration response sent by the fourth entity, the second registration response indicating the input parameters of the fourth entity; the first processing entity sends a second response acknowledgment, the second response acknowledgment indicating whether the output parameters of the first processing entity are compatible with the input parameters of the fourth entity.

[0020] In this implementation, when the output parameters of the first processing entity are incompatible with the input parameters of the second entity, the first processing entity may initiate registration with a fourth entity to ensure that the first processing entity is registered with a self-adaptive entity that is compatible with the first processing entity.

[0021] In a possible implementation, the model information includes one or more of the following: model purpose, model type, model scale, model accuracy, or model performance.

[0022] According to this implementation, the second entity may determine whether the first processing entity matches the model information corresponding to the second entity based on the model information corresponding to the first identifier, and then determine a first registration response. Furthermore, the first processing entity may be registered with the appropriate second entity.

[0023] In a possible implementation, the data information comprises one or more of the following: the type of data to be processed, the dimension of the data to be processed or the precision of the data to be processed.

[0024] According to this implementation, the second entity may determine whether the first processing entity matches the data information corresponding to the second entity based on the data information corresponding to the first identifier, and then determine a first registration response, and the first processing entity may be registered with the appropriate second entity.

[0025] In a possible implementation, the output parameters of the first processing entity include one or more of the following features of the output data of the first processing entity: type, dimension or precision, and the input parameters of the second entity include one or more of the following features of the input data of the second entity: type, dimension or precision.

[0026] In a possible implementation, the first processing entity and the second entity belong to the same communication device.

[0027] According to a second aspect, an embodiment of the present application provides a communication method, which is used by a second entity of a second communication device, and includes: the second entity receives a first registration request sent by a first processing entity, the first registration request including a first identifier, the first identifier indicating one or more of the following information of the first processing entity: model information or data information, and model parameters of the second entity are adjusted based on real-time environmental parameters; the second entity sends a first registration response to the first processing entity, the first registration response indicating input parameters of the second entity, and the input parameters of the second entity are used to determine whether output parameters of the first processing entity match the input parameters of the second entity.

[0028] In the method, the second entity receives the registration request of the first processing entity and feeds back the input parameters of the second entity to the first processing entity, so that the first processing entity determines that the output parameters of the first processing entity match the input parameters of the second entity, and the model parameters of the second entity may be adjusted based on real-time environmental parameters, which can adapt to environmental changes and support effective data reception and transmission.

[0029] In a possible implementation, the second entity receives a first response acknowledgment, where the first response acknowledgment indicates that the output parameters of the first processing entity have successfully matched with the input parameters of the second entity, and the method further includes: the second entity receives first output data through the first interface, where the first output data is output data of the first processing entity; the second entity performs second processing on the first output data to obtain second output data; and the second entity outputs the second output data.

[0030] In a possible implementation manner, the second entity outputting the second output data includes: the second entity transmitting the second output data to a third communication device;

[0031] In a possible implementation manner, the second entity performing the second processing on the first output data to obtain the second output data includes: the second entity mapping the first output data to the second output data, and the mapping manner may be adjusted based on real-time environmental parameters.

[0032] In this implementation, the mapping of the first output data to the second output data is adjusted based on real-time environmental parameters, which can adapt to the changing transmission environment and support effective data reception and transmission.

[0033] In a possible implementation, the first interface defines a procedure and / or signaling for interaction between the first processing entity and the second entity, for example, the first interface defines a transmission process for one or more of a registration request, a registration response, or a response acknowledgment.

[0034] In a possible implementation, when sending the first registration response, the second entity starts a first response acknowledgment time measurement, and when the response acknowledgment time measurement expires, the second entity sends a first response acknowledgment timeout indication to the first control entity.

[0035] In a possible implementation, the model information includes one or more of the following: model purpose, model type, model scale, model accuracy, or model performance.

[0036] In a possible implementation, the data information comprises one or more of the following: the type of data to be processed, the dimension of the data to be processed or the precision of the data to be processed.

[0037] In a possible implementation, the output parameters of the first processing entity include one or more of the following features of the output data of the first processing entity: type, dimension or precision, and the input parameters of the second entity include one or more of the following features of the input data of the second entity: type, dimension or precision.

[0038] In a possible implementation, the first processing entity and the second entity belong to the same communication device.

[0039] The beneficial effects of the possible implementation manner of the second aspect refer to the beneficial effects of the possible implementation manner of the first aspect, and the details will not be described again here.

[0040] According to a third aspect, an embodiment of the present application provides a communication method, which is used by a third processing entity of a third communication device, and includes: the third processing entity sends a second registration request to the second entity, the second registration request including a second identifier, the second identifier indicating one or more of model information or data information of the third processing entity, and model parameters of the second entity are adjusted based on real-time environmental parameters; the third processing entity receives a second registration response sent by the second entity, the second registration response indicating output parameters of the second entity; the third processing entity sends a second response confirmation response, the second response confirmation response indicating whether input parameters of the third processing entity match output parameters of the second entity.

[0041] Optionally, the second entity may also be referred to as a self-adaptive entity.

[0042] In the method, the third processing entity performs registration with the second entity to determine that the input parameters of the third processing entity match the output parameters of the second entity, and the model parameters of the second entity may be adjusted based on real-time environmental parameters, which can adapt to environmental changes and support effective data reception and transmission.

[0043] In a possible implementation manner, the second response acknowledgment indicates that the input parameters of the third processing entity match the output parameters of the second entity, and the method includes: the third processing entity obtains second output data transmitted by the second entity, and performs third processing on the second output data to obtain third data.

[0044] In this implementation method, after the input parameters of the third processing entity are successfully matched with the output parameters of the second entity, the third processing entity receives the second output data transmitted by the second entity, and performs third processing on the second output data to obtain third data, thereby realizing effective data transmission.

[0045] In a possible implementation, the third processing entity includes an input adaptation module, a data space transformation module, and an output adaptation module. The input adaptation module is configured to adapt dimensions of the second output data to input dimensions of the data space transformation module. The data space transformation module is configured to perform a third process on the second output data to obtain second intermediate output data. The output adaptation module is configured to perform a dimensional transformation on the second intermediate output data to obtain third data.

[0046] In this implementation, the input adaptation module and output adaptation module of the third processing entity perform processing such as dimension transformation on the input data and output data of the third processing entity, respectively, to adapt the output dimension of a pre-processing module (e.g., the second entity) of the third processing entity to the input dimension of the data space transformation module, and to adapt the output dimension of the third processing entity to the input dimension of a subsequent processing module of the third processing entity.

[0047] In a possible implementation, the third process is the inverse operation of the first process in the first aspect, and includes one or more of MIMO decoding, demodulation, channel decoding, and source decoding, or processes equivalent to one or more of MIMO decoding, demodulation, channel decoding, and source decoding.

[0048] In this implementation, one or more operations performed in conventional data reception and transmission may be implemented via a third processing entity, which makes it easier to achieve global optimization when implementing multiple operations compared to implementing each operation module separately in conventional links.

[0049] In a possible implementation, the third processing includes demodulation and channel decoding, the second output data is modulation symbols, and the third data is a bit stream.

[0050] In a possible implementation, the third processing includes demodulation, channel decoding and source decoding, the second output data is a modulation symbol sequence, and the third data is a symbol sequence.

[0051] From the above possible implementation manners, it can be seen that all third processing entities realizing different functions can be registered with the second entity compatible with the third processing entities to realize effective data reception and transmission.

[0052] In a possible implementation, the second interface defines a procedure and / or signaling for interaction between the third processing entity and the second entity, for example, the second interface defines a transmission process for one or more of a registration request, a registration response, or a response acknowledgment.

[0053] In a possible implementation, the second response confirmation response indicates that the input parameters of the third processing entity are incompatible with the output parameters of the second entity, and the method further includes: the third processing entity sends a second registration request to the fourth entity, the second registration request including a second identifier, the second identifier indicating one or more of model information or data information of the third processing entity, and the model parameters of the fourth entity may be adjusted based on the real-time environmental parameters; the third processing entity receives a second registration response sent by the fourth entity, the second registration response indicating the input parameters of the fourth entity; the third processing entity sends a second response confirmation response, the second response confirmation response indicating whether the input parameters of the third processing entity are compatible with the input parameters of the fourth entity.

[0054] In this implementation, when the input parameters of the third processing entity are incompatible with the output parameters of the second entity, the third processing entity may initiate registration with a fourth entity to ensure that the third processing entity is registered with a self-adaptive entity that is compatible with the third processing entity.

[0055] In a possible implementation, the model information includes one or more of the following: model purpose, model type, model scale, model accuracy, or model performance.

[0056] According to this implementation, the second entity may be able to determine whether the third processing entity matches the model information corresponding to the second entity based on the model information corresponding to the second identifier, and then determine a second registration response, and the third processing entity may be able to register with the appropriate second entity.

[0057] In a possible implementation, the data information comprises one or more of the following: the type of data to be processed, the dimension of the data to be processed or the precision of the data to be processed.

[0058] According to this implementation, the second entity may determine whether the third processing entity matches the data information corresponding to the second entity based on the data information corresponding to the second identifier, and then determine a second registration response, and the third processing entity may be registered with the appropriate second entity.

[0059] In a possible implementation, the input parameters of the third processing entity include one or more of the type, dimension or precision of the input data of the third processing entity, and the output parameters of the second entity include one or more of the type, dimension or precision of the output data of the second entity.

[0060] According to a fourth aspect, an embodiment of the present application provides a communication method, the method being used by a second entity of a fourth communication device, the method including: the second entity receiving a second registration request sent by a third processing entity, the second registration request including a second identifier, the second identifier indicating one or more of model information or data information of the third processing entity, model parameters of the second entity being adjusted based on real-time environmental parameters; the second entity sending a second registration response to the third processing entity, the second registration response indicating output parameters of the second entity, the output parameters of the second entity being used to determine whether input parameters of the third processing entity are compatible with output parameters of the second entity.

[0061] In the method, the second entity receives a registration request of a third processing entity and feeds back the output parameters of the second entity to the third processing entity, so that the third processing entity determines that the input parameters of the third processing entity match the output parameters of the second entity, and the model parameters of the second entity may be adjusted based on real-time environmental parameters, which can adapt to environmental changes and support effective data reception and transmission.

[0062] In a possible implementation, the second entity receives a second response acknowledgment, where the second response acknowledgment indicates that the input parameters of the third processing entity are successfully matched with the output parameters of the second entity, and the method further includes: the second entity performs a second processing on the first output data to obtain second output data, where the first output data is the output data of the first processing entity; and the second entity transmits the second output data to the third processing entity through the second interface.

[0063] In a possible implementation manner, the second entity performing the second processing on the first output data to obtain the second output data includes: the second entity mapping the first output data to the second output data, and the mapping manner may be adjusted based on real-time environmental parameters.

[0064] In this implementation, the mapping of the first output data to the second output data is adjusted based on real-time environmental parameters, which can adapt to the changing transmission environment and support effective data reception and transmission.

[0065] In a possible implementation, the second interface defines a procedure and / or signaling for interaction between the third processing entity and the second entity, for example, the second interface defines a transmission process for one or more of a registration request, a registration response, or a response acknowledgment.

[0066] In a possible implementation, when sending the second registration response, the second entity starts a second response acknowledgment time measurement, and when the response acknowledgment time measurement expires, the second entity sends a second response acknowledgment timeout indication to the third control entity.

[0067] In a possible implementation, the model information includes one or more of the following: model purpose, model type, model scale, model accuracy, or model performance.

[0068] In a possible implementation, the data information comprises one or more of the following: the type of data to be processed, the dimension of the data to be processed or the precision of the data to be processed.

[0069] In a possible implementation, the input parameters of the third processing entity include one or more of the type, dimension or precision of the input data of the third processing entity, and the output parameters of the second entity include one or more of the type, dimension or precision of the output data of the second entity.

[0070] The beneficial effects of the possible implementation manners of the third and fourth aspects refer to the beneficial effects of the possible implementation manners of the first aspect, and the details will not be described again here.

[0071] According to a fifth aspect, an embodiment of the present application further provides a communication device. The communication device may be used in the first processing entity in the first aspect. The communication device may be a terminal or a network device, a device (e.g., a chip, a chip system, or a circuit) within the terminal or the network device, or a device that can be used with the terminal or the network device.

[0072] In a possible implementation, the communication device may include modules or units that correspond one-to-one to the methods / operations / steps / actions described in the first aspect. The modules or units may be hardware circuits, software, or may be realized by hardware circuits in combination with software.

[0073] In a possible implementation, the communication device may include a processing unit and a transceiver unit. The processing unit is configured to invoke the transceiver unit to perform a receiving function and / or a transmitting function. Specifically, the transceiver unit is configured to send a first registration request to a second entity, the first registration request including a first identifier, the first identifier indicating one or more of model information or data information of the communication device, and model parameters of the second entity are adjusted based on real-time environmental parameters. The transceiver unit is further configured to receive a first registration response sent by the second entity, the first registration response indicating input parameters of the second entity. The transceiver unit is further configured to send a first response confirmation response, the first response confirmation response indicating whether output parameters of the first processing entity match with input parameters of the second entity.

[0074] In a possible implementation, the processing unit is further configured to perform a first operation on the first input data to obtain first output data, and the transceiver unit is further configured to transmit the first output data to the second entity through the first interface.

[0075] In a possible implementation, the transceiver unit includes an input adaptation module and an output adaptation module, and the processing unit includes a data space transformation module. The input configuration module is configured to adapt dimensions of first input data to input dimensions of the data space transformation module. The data space transformation module is configured to perform a first operation on the first input data to obtain first intermediate output data. The output adaptation module is configured to perform a dimensional transformation on the first intermediate output data to obtain the first output data.

[0076] In a possible implementation, the first processing comprises one or more of the following: quantization, source coding, channel coding, modulation and MIMO precoding, or processing equivalent to one or more of the following: quantization, source coding, channel coding, modulation and MIMO precoding.

[0077] In a possible implementation, the first response acknowledgment indicates that the output parameters of the communication device are incompatible with the input parameters of the second entity. The transceiver unit sends a second registration request to the fourth entity, the second registration request including a first identifier, the first identifier indicating one or more of the following information of the first processing entity: model information or data information, and the model parameters of the fourth entity may be adjusted based on the real-time environmental parameters. The transceiver unit is further configured to receive a second registration response sent by the fourth entity, the second registration response indicating the input parameters of the fourth entity. The transceiver unit is further configured to send a second response acknowledgment, the second response acknowledgment indicating whether the output parameters of the first processing entity are compatible with the input parameters of the fourth entity.

[0078] In a possible implementation, the model information includes one or more of the following: model purpose, model type, model scale, model accuracy, or model performance.

[0079] In a possible implementation, the data information comprises one or more of the following: the type of data to be processed, the dimension of the data to be processed or the precision of the data to be processed.

[0080] In a possible implementation, the output parameters of the first processing entity include one or more of the following features of the output data of the first processing entity: type, dimension or precision, and the input parameters of the second entity include one or more of the following features of the input data of the second entity: type, dimension or precision.

[0081] In a possible implementation, the communication unit and the second entity belong to the same communication device.

[0082] In a possible implementation, the processing unit is a processor and the transceiver unit is a transceiver.

[0083] According to a sixth aspect, an embodiment of the present application further provides a communication device. The communication device may be used in the second entity in the second aspect. The communication device may be a terminal or a network device, a device within the network device (e.g., a chip, a chip system, or a circuit), or a device that can be used with the network device.

[0084] In a possible implementation, the communication device may include modules or units that correspond one-to-one to the methods / operations / steps / actions described in the second aspect. The modules or units may be hardware circuits, software, or may be realized by hardware circuits in combination with software.

[0085] In a possible implementation, the communication device may include a processing unit and a transceiver unit. The processing unit is configured to invoke the transceiver unit to perform a receiving function and / or a transmitting function. Specifically, the transceiver unit is configured to receive a first registration request sent by a first processing entity, the first registration request including a first identifier, the first identifier indicating one or more of the following information of the first processing entity: model information or data information, and model parameters of the communication device are adjusted based on real-time environmental parameters. The transceiver unit is configured to send a first registration response to the first processing entity, the first registration response indicating input parameters of the communication device, and the input parameters of the communication device are used to determine whether output parameters of the first processing entity match the input parameters of the communication device.

[0086] In a possible implementation, the transceiver unit is configured to receive a first response acknowledgment, the first response acknowledgment indicating that the output parameters of the first processing entity have successfully matched with the input parameters of the second entity. The transceiver unit is further configured to receive first output data through the first interface, the first output data being output data of the first processing entity. The processing unit is configured to perform second processing on the first output data to obtain second output data. The transceiver unit is further configured to output the second output data.

[0087] In a possible implementation, the transceiver unit is specifically configured to transmit the second output data to a third communication device.

[0088] In a possible implementation, the processing unit is specifically configured to map the first output data to the second output data, and the mapping manner may be adjusted based on real-time environmental parameters.

[0089] In a possible implementation, the processing unit is further configured to initiate a first response acknowledgment time measurement when the transceiver unit transmits the first registration response, and the transceiver unit is further configured to send a first response acknowledgment timeout indication to the first control entity when the response acknowledgment time measurement expires.

[0090] In a possible implementation, the model information includes one or more of the following: model purpose, model type, model scale, model accuracy, or model performance.

[0091] In a possible implementation, the data information comprises one or more of the following: the type of data to be processed, the dimension of the data to be processed or the precision of the data to be processed.

[0092] In a possible implementation, the output parameters of the first processing entity include one or more of the following features of the output data of the first processing entity: type, dimension or precision, and the input parameters of the second entity include one or more of the following features of the input data of the second entity: type, dimension or precision.

[0093] In a possible realisation, the communication unit and the first processing entity belong to the same communication device.

[0094] According to a seventh aspect, an embodiment of the present application further provides a communication device. The communication device may be used in the third processing entity in the third aspect. The communication device may be a terminal or a network device, a device (e.g., a chip, a chip system, or a circuit) within the terminal or the network device, or a device that can be used with the terminal or the network device.

[0095] In a possible implementation, the communication device may include modules or units that correspond one-to-one to the methods / operations / steps / actions described in the third aspect. The modules or units may be hardware circuits, software, or may be realized by hardware circuits in combination with software.

[0096] In a possible implementation, the communication device may include a processing unit and a transceiver unit. The processing unit is configured to call the transceiver unit to perform a receiving function and / or a transmitting function. Specifically, the transceiver unit is configured to send a second registration request to a second entity, the second registration request including a second identifier, the second identifier indicating one or more of model information or data information of the communication device, and the model parameters of the second entity are adjusted based on real-time environmental parameters. The transceiver unit is further configured to receive a second registration response sent by the second entity, the second registration response indicating output parameters of the second entity. The transceiver unit is further configured to send a second response confirmation response, the second response confirmation response indicating whether the input parameters of the communication device match the output parameters of the second entity.

[0097] In a possible implementation, the second response acknowledgment indicates that the input parameters of the third processing entity match the output parameters of the second entity, the transceiver unit is further configured to obtain second output data transmitted by the second entity, and the processing unit is further configured to perform a third processing on the second output data to obtain the third data.

[0098] In a possible implementation, the transceiver unit includes an input adaptation module and an output adaptation module, and the processing unit includes a data space transformation module. The input configuration module is configured to adapt dimensions of the second output data to input dimensions of the data space transformation module. The data space transformation module is configured to perform a third operation on the second output data to obtain second intermediate output data. The output adaptation module is configured to perform a dimensional transformation on the second intermediate output data to obtain third data.

[0099] In a possible implementation, the third process is the inverse operation of the first process in the first aspect, and includes one or more of MIMO decoding, demodulation, channel decoding, and source decoding, or processes equivalent to one or more of MIMO decoding, demodulation, channel decoding, and source decoding.

[0100] In a possible implementation, the second response acknowledgment indicates that the input parameters of the third processing entity are incompatible with the output parameters of the second entity. The transceiver unit is further configured to send a second registration request to the fourth entity, the second registration request including a second identifier, the second identifier indicating one or more of model information or data information of the third processing entity, and the model parameters of the fourth entity may be adjusted based on the real-time environmental parameters. The transceiver unit is further configured to receive a second registration response sent by the fourth entity, the second registration response indicating the input parameters of the fourth entity. The transceiver unit is further configured to send a second response acknowledgment, the second response acknowledgment indicating whether the input parameters of the third processing entity are compatible with the input parameters of the fourth entity.

[0101] According to an eighth aspect, an embodiment of the present application further provides a communication device. The communication device may be used in the second entity in the fourth aspect. The communication device may be a terminal or a network device, a device (e.g., a chip, a chip system, or a circuit) within the terminal or the network device, or a device that can be used with the terminal or the network device.

[0102] In a possible implementation, the communication device may include modules or units that correspond one-to-one to the methods / operations / steps / actions described in the fourth aspect. The modules or units may be hardware circuits, software, or may be realized by hardware circuits in combination with software.

[0103] In a possible implementation, the communication device may include a processing unit and a transceiver unit. The processing unit is configured to invoke the transceiver unit to perform a receiving function and / or a transmitting function. Specifically, the transceiver unit is configured to receive a second registration request sent by a third processing entity, the second registration request including a second identifier, the second identifier indicating one or more of model information or data information of the third processing entity, and model parameters of the communication device are adjusted based on real-time environmental parameters. The transceiver unit is configured to send a second registration response to the third processing entity, the second registration response indicating output parameters of the communication device, and the output parameters of the communication device are used to determine whether input parameters of the third processing entity match output parameters of the second entity.

[0104] In a possible implementation, the transceiver unit is configured to receive a second response acknowledgment, where the second response acknowledgment indicates that the input parameters of the third processing entity have successfully matched with the output parameters of the communication device. The processing unit is configured to perform a second processing on the first output data to obtain second output data, where the first output data is output data of the first processing entity. The transceiver unit is further configured to output the second output data.

[0105] A second processing is performed on the first output data to obtain second output data, and the transceiver unit is further configured to output the second output data.

[0106] In a possible implementation, the processing unit is specifically configured to map the first output data to the second output data, and the mapping manner may be adjusted based on real-time environmental parameters.

[0107] In a possible implementation, the processing unit is further configured to initiate a second response acknowledgment time measurement when the transceiver unit transmits the second registration response, and the transceiver unit is further configured to send a second response acknowledgment timeout indication to the third control entity when the response acknowledgment time measurement expires.

[0108] In a possible implementation, the model information includes one or more of the following: model purpose, model type, model scale, model accuracy, or model performance.

[0109] In a possible implementation, the data information comprises one or more of the following: the type of data to be processed, the dimension of the data to be processed or the precision of the data to be processed.

[0110] In a possible implementation, the input parameters of the third processing entity include one or more of the type, dimension or precision of the input data of the third processing entity, and the output parameters of the second entity include one or more of the type, dimension or precision of the output data of the second entity.

[0111] According to a ninth aspect, an embodiment of the present application further provides a communication device. The communication device includes a processor and is configured to implement the method of the first aspect or any one of the possible implementation manners of the first aspect. In one possible implementation manner, the processor implements the method by using a logic circuit. In another possible implementation manner, the processor implements the method by executing instructions.

[0112] Specifically, the processor is configured to output a first registration request, the first registration request including a first identifier, the first identifier indicating one or more of model information or data information of the communication device, and model parameters of the second entity are adjusted based on the real-time environmental parameters. The processor is further configured to receive a first registration response sent by the second entity, the first registration response indicating input parameters of the second entity. The processor is further configured to send a first response confirmation response, the first response confirmation response indicating whether the output parameters of the first processing entity match the input parameters of the second entity.

[0113] In a possible implementation, the processor is further configured to perform a first operation on the first input data to obtain first output data, and to transmit the first output data to the second entity through the first interface.

[0114] According to a tenth aspect, an embodiment of the present application further provides a communication device. The communication device includes a processor and is configured to implement the method of the second aspect or any one of the possible implementation manners of the second aspect. In one possible implementation manner, the processor implements the method by using a logic circuit. In another possible implementation manner, the processor implements the method by executing instructions.

[0115] Specifically, the processor is configured to receive a first registration request sent by a first processing entity, the first registration request including a first identifier, the first identifier indicating one or more of the following information of the first processing entity: model information or data information, and model parameters of the communication device are adjusted based on the real-time environmental parameters. The processor outputs a first registration response, the first registration response indicating input parameters of the communication device, and the input parameters of the communication device are used to determine whether the output parameters of the first processing entity match the input parameters of the communication device.

[0116] According to an eleventh aspect, an embodiment of the present application further provides a communication device. The communication device includes a processor and is configured to implement the method of the third aspect or any one of the possible implementation manners of the third aspect. In a possible implementation manner, the processor implements the method by using a logic circuit. In another possible implementation manner, the processor implements the method by executing instructions. Optionally, the communication device further includes a memory configured to store instructions.

[0117] Specifically, the processor is configured to output a second registration request, the second registration request including a second identifier, the second identifier indicating one or more of model information or data information of the communication device, and the model parameters of the second entity are adjusted based on the real-time environmental parameters. The processor is further configured to receive a second registration response sent by the second entity, the second registration response indicating output parameters of the second entity. The processor is further configured to output a second response confirmation response, the second response confirmation response indicating whether the input parameters of the communication device match the output parameters of the second entity.

[0118] According to a twelfth aspect, an embodiment of the present application further provides a communication device. The communication device includes a processor and is configured to implement the method of the fourth aspect or any one of the possible implementation manners of the fourth aspect. In a possible implementation manner, the processor implements the method by using a logic circuit. In another possible implementation manner, the processor implements the method by executing instructions. Optionally, the communication device further includes a memory configured to store instructions.

[0119] Specifically, the processor is configured to receive a second registration request sent by a third processing entity, the second registration request including a second identifier, the second identifier indicating one or more of model information or data information of the third processing entity, and model parameters of the communication device are adjusted based on the real-time environmental parameters. The processor is further configured to output a second registration response, the second registration response indicating output parameters of the communication device, and the output parameters of the communication device are used to determine whether input parameters of the third processing entity match the output parameters of the communication device.

[0120] According to a thirteenth aspect, an embodiment of the present application further provides a communications device. The communications device includes a processor and is configured to execute a computer program (or computer-executable instructions) stored in a memory. When the computer program (or computer-executable instructions) is executed, the device is enabled to perform the method of the first aspect or any one of possible implementations of the first aspect, the method of the second aspect or any one of possible implementations of the second aspect, the method of the third aspect or any one of possible implementations of the third aspect, or the method of the fourth aspect or any one of possible implementations of the fourth aspect.

[0121] In a possible implementation, the processor and memory are integrated together.

[0122] In another possible implementation, the memory is located external to the communication device.

[0123] The communication device further includes a communication interface used by the communication device to communicate with other devices, e.g., to send or receive data and / or signals. For example, the communication interface may be a transceiver, a circuit, a bus, a module, or another type of communication interface.

[0124] According to a fourteenth aspect, an embodiment of the present application further provides a communication device, wherein the communication device is configured to perform the method of the first aspect or any one of the possible implementation manners of the first aspect.

[0125] According to a fifteenth aspect, an embodiment of the present application further provides a communication device, the communication device being configured to perform the method of the second aspect or any one of the possible implementation manners of the second aspect, and / or the method of the fourth aspect or any one of the possible implementation manners of the fourth aspect.

[0126] According to a sixteenth aspect, an embodiment of the present application further provides a communication device, wherein the communication device is configured to perform the method of the third aspect or any one of the possible implementation manners of the third aspect.

[0127] The beneficial effects of the communication devices in the fifth to sixteenth aspects and possible implementations of the fifth to sixteenth aspects are described in the beneficial effects of the method in the first aspect and possible implementations of the first aspect, and the details will not be described again here.

[0128] According to a seventeenth aspect, an embodiment of the present application further provides a computer-readable storage medium that stores a computer program (or computer-executable instructions), which, when executed by a processor, performs some or all of the steps of the method in any one of the possible implementations of the first aspect, the second aspect and any one of the possible implementations of the second aspect, the third aspect and any one of the possible implementations of the third aspect, or the fourth aspect and any one of the possible implementations of the fourth aspect.

[0129] According to an eighteenth aspect, an embodiment of the present application further provides a computer program product including computer-executable instructions, which, when executed, performs some or all of the steps of the method in the first aspect and any possible implementation manner of the first aspect, the second aspect and any possible implementation manner of the second aspect, the third aspect and any possible implementation manner of the third aspect, or the fourth aspect and any possible implementation manner of the fourth aspect.

[0130] According to a nineteenth aspect, an embodiment of the present application further provides a computer program comprising computer-executable instructions, which, when executed, performs some or all of the steps of the method in the first aspect and any possible implementation manner of the first aspect, the second aspect and any possible implementation manner of the second aspect, the third aspect and any possible implementation manner of the third aspect, or the fourth aspect and any possible implementation manner of the fourth aspect.

[0131] According to a twentieth aspect, an embodiment of the present application further provides a chip system. The chip system includes a processor, and may further include a memory, and is configured to implement the method in any possible implementation manner of the first aspect and the first aspect, the second aspect and the second aspect, the third aspect and the third aspect, or the fourth aspect and the fourth aspect. The chip system may include a chip, or may include a chip and other discrete components.

[0132] According to a twenty-first aspect, an embodiment of the present application further provides a communication system, the communication system including a communication device provided in the fifth aspect or any one of possible implementations of the fifth aspect, a communication device provided in the sixth aspect or the eighth aspect or any one of possible implementations of the sixth aspect or the eighth aspect, and a communication device provided in the seventh aspect or any one of possible implementations of the seventh aspect. [Brief explanation of the drawings]

[0133] The following describes some accompanying drawings in embodiments of this application. [Figure 1] FIG. 1 is a diagram of the structure of an autoencoder. [Figure 2] 1 is a diagram of the architecture of a communication system to which embodiments of the present application are applicable; [Figure 3] FIG. 1 is a diagram of an artificial intelligence communication transceiver. [Figure 4] FIG. 1 is a diagram of a communication architecture according to an embodiment of the present application. [Figure 5] 5 is a diagram of interactions in a communication method 500 according to an embodiment of the present application. [Figure 6] 6 is a diagram of interactions in a communication method 600 according to an embodiment of the present application. [Figure 7] 7 is a diagram of interactions in a communication method 700 according to an embodiment of the present application. [Figure 8] 8 is a diagram of interactions in a communication method 800 according to an embodiment of the present application. [Figure 9] 9 is a diagram of interactions in a communication method 900 according to an embodiment of the present application. [Figure 10] 9 is a diagram of interactions in a communication method 900 according to an embodiment of the present application. [Figure 11] 1 is a schematic diagram of the structure of a processing entity according to an embodiment of the present application; [Figure 12] FIG. 2 is a diagram of the structure of a self-adaptive entity according to an embodiment of the present application; [Figure 13] An example is given in which the method provided in the embodiments of this application is realized by using a neural network and a self-adaptive iterative algorithm. [Figure 14] 1 is a diagram of the structure of a communication device according to an embodiment of the present application; [Figure 15] 1 is a diagram of the structure of a communication device according to an embodiment of the present application; [Figure 16] 1 is a diagram of the structure of a communication device according to an embodiment of the present application; DETAILED DESCRIPTION OF THE INVENTION

[0134] SUMMARY OF THE INVENTION Embodiments of this application provide a communication method and related apparatus for adapting to environmental changes and supporting efficient data reception and transmission.

[0135] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings in which: FIG.

[0136] First, terms used in the embodiments of this application will be explained.

[0137] The autoencoder in the embodiment of this application is an AI technology. Figure 1 is a diagram of the structure of the autoencoder. As shown in Figure 1, input data x is mapped / compressed into variables z in a latent space by using the f function (encoder), and then the data

number

number

[0138] In this application, " / " means "or." The term "and / or" only describes an association relationship for describing related objects and indicates that three relationships may exist. For example, B and / or C may represent the following three cases: only B is present, both B and C are present, and only C is present. In the description and claims of the embodiments of this application, terms such as "first" and "second" are intended to distinguish between different objects and are not intended to describe a specific order of objects unless otherwise specified. For example, a first communication device and a second communication device are used to distinguish between different communication devices and are not intended to describe a specific order of target objects. In the embodiments of this application, words such as "example" or "for example" are used to indicate an example, illustration, or explanation, and any embodiment or solution described as an "example" or "for example" in the embodiments of this application should not be construed as being preferred or having more advantages than other embodiments or design solutions. In the description of the embodiments of this application, "plurality" means two or more unless otherwise specified. For example, a plurality of processing units means two or more processing units.

[0139] The technical solutions of this application may be applied to cellular systems related to the 3rd generation partnership project (3GPP), for example, 4th generation (4G) communication systems such as long term evolution (LTE) systems, 5th generation (5G) communication systems such as new radio (NR) systems, communication systems supporting a collection of multiple wireless technologies, or 5G and beyond evolutionary communication systems.

[0140] 1 shows an example of a communication system to which embodiments of this application are applicable. Referring to FIG. 1, the communication system 100 includes at least one terminal device 110 and at least one access network device 120.

[0141] The terminal device 110 referred to in this embodiment of this application may be a device having a wireless transceiver function and may communicate with the access network device 120 or other terminal devices. The terminal device 110 may specifically be a user equipment (UE), an access terminal, a subscriber unit, a subscriber station, a mobile station, a remote station, a remote terminal, a mobile device, a user terminal, a wireless communication device, a user agent, or a user equipment. The terminal device may alternatively be a satellite phone, a mobile phone, a smartphone, a wireless data card, a wireless modem, a machine type communication device, a cordless phone, a session initiation protocol (SIP) phone, a wireless local loop (WLL) station, a personal digital assistant (PDA), a handheld device with wireless communication capabilities, a computing device, or other processing device connected to a wireless modem, an in-vehicle device, a communication device mounted on a high altitude aircraft, a wearable device, an unmanned aerial vehicle, a robot, a terminal in device-to-device (D2D) communication, a terminal in vehicle-to-everything (V2X), a virtual reality (VR) terminal device, an augmented reality (AR) terminal device, a wireless terminal in industrial control, a wireless terminal in self driving, a wireless terminal in remote medical, a wireless terminal in a smart grid, a wireless terminal in transportation safety, or the like. The wireless terminal may be a wireless terminal in smart safety, a wireless terminal in a smart city, a wireless terminal in a smart home, or a terminal device in an evolved communication network after 5G.This is not a limitation in this application.

[0142] The access network device 120 is a device having a wireless transceiver function and configured to communicate with the terminal device 110. The access network device may be a node in a radio access network (RAN) and may be referred to as a RAN node or a network device, or may be referred to as a base station. The access network device may be an evolved NodeB (eNB or eNodeB) in LTE, a base station in a 5G network such as a gNodeB (gNB), a base station in a post-5G evolved public land mobile network (PLMN), or a base station in a non-3G partnership project (3G) network. rd The network device may be a Third Generation Partnership Project (3GPP) access device, etc. Optionally, the network device in this embodiment of the present application may include various types of base stations, such as a macro base station, a micro base station (also called a small cell), a relay station, a transmitting and receiving point (TRP), a transmitting point (TP), and a device that functions as a base station in a mobile switching center, a device-to-device (D2D), a vehicle-to-everything (V2X), and a machine-to-machine (M2M) communication. The network device may also include a central unit (CU) and a distributed unit (DU) in a cloud radio access network (C-RAN) system, and a network device in a non-terrestrial network (NTN) communication system, which is not particularly limited in this embodiment of the present application.

[0143] 2 may further include a core network device (not shown in the drawing). The access network device 120 may interact with the core network device to provide communication services to terminals. For example, the core network device may provide communication connectivity, authentication, management and policy control for terminals, and may be responsible for data services, etc.

[0144] In this embodiment of the present application, the apparatus configured to realize the functions of the terminal device may be the terminal device itself, or may be an apparatus capable of supporting the terminal device in realizing the functions, such as a chip system. The apparatus may be mounted on the terminal device or used together with the terminal device. In this embodiment of the present application, the chip system may include a chip, or may include a chip and other discrete components.

[0145] In this embodiment of the present application, the apparatus configured to realize the functions of an access network device may be an access network device, or may be an apparatus, such as a chip system, that can support the access network device in realizing the functions. The apparatus may be mounted on the access network device or used together with the access network device.

[0146] In this embodiment of the present application, the terminal device may further have AI processing capability, and the access network device may also have AI processing capability. This is applicable to scenarios where AI and communication are combined. A possible scenario where AI and communication are combined is an AI communication transceiver. As shown in Figure 3, one or more modules of the transmitter and receiver are realized through an AI model (e.g., a neural network) to obtain excellent global performance.

[0147] The current combination of AI and communication is mainly to design an AI model or training method for given system parameters in a specific scenario. When the environment, such as device capabilities, scenarios (e.g., channel conditions), or system parameters, changes, the corresponding model needs to be redesigned and retrained, making it difficult to deploy and use the combination of AI and communication in a universal communication system.

[0148] To solve the above problems, embodiments of the present application provide a communication method, related apparatus and system for adapting to environmental changes such as channel conditions and user requirements and realizing effective data reception and transmission.

[0149] 4 is a diagram of the architecture of a communication system according to an embodiment of the present application. As shown in FIG. 4, the architecture of the communication system includes a first processing module 410, a second processing module 420, a third processing module 430, a first interface 440, and a second interface 450. The first processing module 410 acquires and processes data 1 to acquire data 2, and transmits data 2 to the second processing module 420 through the first interface 440. The second processing module 420 acquires and processes data 2 to acquire data 3, and transmits data 3 to the third processing module 430 through the second interface 450. The third processing module 430 acquires and processes data 2 to acquire data 4.

[0150] The first processing module 410 may be located at a transmitting end, and the data 1 may be a bit stream or symbols. For example, the first processing module 410 may perform one or more operations such as source coding, channel coding, modulation, and multiple-input multiple-output (MIMO) precoding on the data 1, or an operation equivalent to one or more of the above operations. The third processing module 430 may be located at a receiving end. For example, the third processing module 430 may perform one or more operations such as demodulation, channel decoding, and source decoding on the obtained data 3, or an operation equivalent to one or more of the above operations, to obtain data 4, and the data 4 may be a bit stream or symbols obtained by restoring the data 1.

[0151] The first processing module 410 and / or the third processing module 430 may be realized via a neural network model. Optionally, the first processing module 410 and / or the third processing module 430 may be configured through offline pre-training by using a large amount of environmental data (such as channel data), and the training may no longer be performed during actual use or inference.

[0152] The first interface 440 defines the interaction procedures and / or signaling between the first processing module 410 and the second processing module 420. The second interface 450 defines the interaction procedures and / or signaling between the second processing module 420 and the third processing module 430. The first interface 440 and / or the second interface 450 may be logical interfaces or physical interfaces.

[0153] The second processing module 420 may also be referred to as a self-adaptive module. In the communication process, adjustments may be performed based on environmental data. The "adjustment" may be achieved through neural network training or by solving analytical equations.

[0154] In the above communication architecture, the first processing module 410 and the third processing module 430 may be adjusted by using offline training, and the second processing module 420 may be adjusted based on actual environmental changes. In this way, adaptation to environmental changes can be achieved with low complexity, and effective data reception and transmission can be achieved.

[0155] The first processing module and the third processing module are obtained through offline training using a large amount of environmental data. The communication device may pre-store one or more first processing modules and / or one or more third processing modules. Optionally, the one or more first processing modules and / or one or more third processing modules may be downloaded from a model server or provided by a third party. During actual use, the communication device may select the first processing module or the third processing module based on the computing power, memory capacity, power consumption, and task type of the communication device. The second processing module is a module that is locally configured in the communication device and can be adjusted based on real-time environmental data. As the communication environment changes, the input / output parameters, model parameters, etc. of the second processing module may change. This may cause incompatibility between the first processing module and the second processing module, or between the third processing module and the second processing module. Therefore, when selecting the first processing module / third processing module and the second processing module, the communication device needs to consider adaptation between the first processing module / third processing module and the second processing module. Furthermore, the communication devices at the transmitting end and the receiving end may respectively select a first processing module and a third processing module to jointly realize data reception and transmission, and adaptation is also required for the selection of the first processing module and the selection of the third processing module to realize effective communication.

[0156] To address the above-mentioned problem, an embodiment of this application provides a registration procedure. A first processing module in a transmitting communication device may adapt to a second processing module in the communication device through the registration procedure, or may adapt to a third processing module in a receiving end by performing registration with the second processing module. The second processing module may be adjusted based on the actual transmission environment. This allows each processing module to complete data processing, adapt to changing communication environments, and achieve effective data reception and transmission.

[0157] In a possible implementation, interaction procedures and / or signaling between the processing entity and the self-adaptive entity may be defined through an interface between the processing entity and the self-adaptive entity, such as the above-mentioned first and second interfaces. The interaction procedures may include a registration process, a data transmission process, etc. Parameters related to the signaling may include identification information, data type, data precision, input dimension, output dimension, etc. transmitted between the processing entity and the self-adaptive entity.

[0158] The values ​​of the parameters related to the signaling defined by the first interface and the second interface may be predefined in a standard, or may be semi-statically or dynamically configured. In a possible implementation, the identification information may be associated with a task type of a specific implementation. Optionally, the task type may be indicated by using radio resource control (RRC) signaling. For example, a task type or task identifier field may be added to the RRC. In a possible implementation, the data type and data precision may be associated with performance requirements of the task. Optionally, the performance requirements of the task may be indicated by using RRC signaling. For example, a performance requirement field may be added to the RRC. In a possible implementation, the input dimension and the output dimension may be associated with actual transmission resources, actual transmission policy, etc., and the transmission resources or transmission policy may be indicated by using downlink control information (DCI) signaling. The transmission policy may be a modulation and coding scheme (MCS). For example, parameters such as the number of information bits, the number of coded bits and the number of modulated modulation symbols are determined based on the transmission resource and the MCS level, and these parameters are used to determine the input dimension and / or the output dimension.

[0159] As shown in FIG. 5 , an embodiment of the present application provides a communication method 500. The method is applicable to a transmitting end and relates to a first control entity, a first processing entity, and a second entity. The first control entity is configured to control or manage the first processing entity, the second entity, and the interaction between the first processing entity and the second entity. The first processing entity and the second entity correspond to the first processing module 410 and the second processing module 420, respectively, in the communication architecture shown in FIG. 4 . A model corresponding to the first processing entity may be obtained through offline training, and a model corresponding to the second entity may be adjusted based on environmental information. The first control entity, the first processing entity, and the second entity may be physically located in the same communication device or different communication devices. The communication device may be the above-mentioned terminal device or access network device.

[0160] The specific contents of the communication method 500 will be described below.

[0161] S500 (optional step): The first control entity sends a first registration instruction to the first processing entity, and in response, the first processing entity receives a registration instruction, where the registration instruction includes identification information of the second entity.

[0162] In a possible implementation manner, the first control entity selects a target processing entity, i.e., the first processing entity, from the one or more processing entities, selects a target self-adaptation entity, i.e., the second entity, from the one or more self-adaptation entities, and sends a registration instruction including identification information of the second entity to the first processing entity.

[0163] Optionally, the first control entity selects the target processing entity and the target self-adaptation entity based on a task type, which may correspond to one or more of the following: quantization, source coding, channel coding, modulation, and MIMO precoding.

[0164] In a possible implementation, the first control entity is a protocol layer entity above the physical layer, which may also be referred to as an upper layer entity, for example, the first control entity is a media access control (MAC) layer entity, and the registration indication is MAC layer signaling.

[0165] S501: A first processing entity sends a first registration request to a second entity, and in response, the second entity receives the first registration request, where the first registration request includes a first identifier.

[0166] In a possible implementation manner, the first processing entity sends a first registration request to the second entity based on the first registration instruction received in S500.

[0167] In another possible implementation manner, the first processing entity selects the second entity based on a task type corresponding to the first processing entity, and sends the first registration request to the second entity.

[0168] Specifically, the first identifier indicates one or more of the following information of the first processing entity: model information or data information.

[0169] The model information may be referred to as a model description and may include one or more of the following: model purpose, model type, model scale, model accuracy, or model performance.

[0170] The model objective indicates a task type corresponding to the first processing entity. For example, the model objective may include one or more of the following: quantization, source coding, channel coding, modulation, and MIMO precoding. The model type indicates a type of AI model used by the first processing entity. The model type may be indicated in a level-based manner. For example, the model type includes a category and a subcategory. For example, the category may include a fully connected network (FCN), a convolutional neural network (CNN), a recurrent neural network (RNN), a transformer, etc. The subcategories of CNN may include a spatial utilization-based CNN, a depth-based CNN, a multi-pass CNN, an attention mechanism-based CNN, etc. The subcategories of RNN may include a gated recurrent unit (GRU), a long short-term memory (LSTM), etc. The model scale indicates the size of the model. When the model is a neural network, the model scale may include the number of neural network layers, neurons in each layer, and one or more of the following parameters: weights and biases. Model precision indicates the precision used by the model parameters. Module precision may include binary type, integer type, floating point type, etc., and the floating point type may further include float16, float32, etc. Model performance indicates the inference performance of the model. The inference performance of the model may be represented by the accuracy of inference performed by the model on a reference dataset.

[0171] The data information may be referred to as a data description. The data information may include one or more of the following: data type, data dimension, and data precision.

[0172] The data type indicates the type of data that can be processed by the model. The data type may include bit, integer, real number, complex number, etc. The data dimension indicates the input dimension and / or output dimension of the model. The data dimension may include the number of elements in an input data vector or the number of elements in an output data vector. The data precision indicates the precision of the data that can be processed by the model. The data precision may include binary type, integer type, floating point type, etc., and the floating point type may further include float16, float32, etc.

[0173] S502: The second entity sends a first registration response to the first processing entity, and in response, the first processing entity receives the first registration response, where the first registration response includes input parameters of the second entity.

[0174] Specifically, the second entity determines whether the model information and / or data information of the first processing entity meets the requirements of the second entity based on the first identifier in the received first registration request, and when determining that the model information and / or data information of the first processing entity meets the requirements of the second entity, sends a first registration response to the first processing entity.

[0175] Alternatively, when the first processing entity determines that the model information and / or data information does not meet the requirements of the second entity, the second entity may feed back a registration failure indication to the first processing entity, and optionally, the first processing entity reports the registration failure indication to the first control entity.

[0176] Alternatively, when the first processing entity determines that the model information and / or data information does not meet the requirements of the second entity, the second entity may not feedback a registration failure indication but may implicitly indicate a registration failure. When sending the first registration request, the first processing entity starts a first registration response time measurement. If a first registration response is received before the time measurement expires, the time measurement is reset. If the time measurement expires, a first registration response timeout indication is sent to the first control entity.

[0177] In a possible implementation, the input parameters of the second entity include one or more of the following aspects of the input data of the second entity: type, dimension or precision.

[0178] S503 (not shown separately in the drawings): The first processing entity sends a first response confirmation response based on the received first registration response, and the first response confirmation response indicates whether the output parameters of the first processing entity match the input parameters of the second entity.

[0179] The output parameters of the first processing entity include one or more of the following aspects of the output data of the first processing entity: type, dimension or precision.

[0180] Specifically, before sending the first response acknowledgment, the first processing entity determines whether the output parameters of the first processing entity are compatible with the input parameters of the second entity, for example, whether the type of output data of the first processing entity is compatible with the type of input data of the second entity, whether the dimension of the output data of the first processing entity is compatible with the dimension of the input data of the second entity, and / or whether the precision of the output data of the first processing entity is compatible with the precision of the input data of the second entity.

[0181] When the first processing entity determines that the output parameters of the first processing entity are successfully matched with the input parameters of the second entity, the first processing entity sends a first response confirmation response to the second entity (S503a). If the match fails, the first processing entity sends a first response confirmation response to the first control entity (S503b).

[0182] S503a: The first processing entity sends a first response confirmation response to the second entity, where the first response confirmation response indicates that the output parameters of the first processing entity successfully match the input parameters of the second entity, and in response, the second entity receives the first response confirmation response.

[0183] In a possible implementation, after determining that the output parameters of the first processing entity successfully match the input parameters of the second entity, the first processing entity performs a first processing on the first input data to obtain first output data, and transmits the first output data to the second entity through the first interface.

[0184] For details of the data processing and data transmission processes of the first processing entity, the second entity and the third processing entity, please refer to the related description below in FIG.

[0185] S503b: The first processing entity sends a first response acknowledgment to the first control entity, where the first response acknowledgment indicates that the output parameters of the first processing entity are incompatible with the input parameters of the second entity, and in response, the first control entity receives the first response acknowledgment.

[0186] Optionally, when the first processing entity determines that the output parameters of the first processing entity are incompatible with the input parameters of the second entity, the first processing entity may feed back a parameter compatibility failure indication to the second entity.

[0187] Alternatively, after determining that the output parameters of the first processing entity are incompatible with the input parameters of the second entity, the first processing entity may not feed back a parameter match failure indication. When sending the first registration response, the second entity starts a first response acknowledgment time measurement. If the first response acknowledgment is received before the time measurement expires, the time measurement is reset. If the time measurement expires, a first response acknowledgment timeout indication is sent to the first control entity. The second entity learns of the parameter match failure in an implicit manner, which can reduce signaling overhead.

[0188] The parameters of the model corresponding to the first processing entity may be obtained through offline training, and the parameters of the model corresponding to the second entity may be adjusted based on the environment information. In the actual communication process, the first processing entity does not need to retrain its model, but performs registration with the appropriate second entity through the registration procedure shown in Figure 5 (model information and / or data information match, and input parameters match output parameters), and then performs data processing and transmission. This method can achieve adaptation to changes in the actual environment with low complexity and realize effective data reception and transmission.

[0189] As shown in FIG. 6 , an embodiment of the present application provides a communication method 600. The method is applicable to a receiving end and relates to a third control entity, a third processing entity, and a second entity. The third control entity is configured to control or manage the third processing entity, the second entity, and the interaction between the third processing entity and the second entity. The third processing entity and the second entity correspond to the third processing module 430 and the second processing module 420, respectively, in the communication architecture shown in FIG. 4 . A model corresponding to the third processing entity may be obtained through offline training, and a model corresponding to the second entity may be adjusted based on environmental information. The third control entity, the third processing entity, and the second entity may be physically located in the same communication device or different communication devices. The communication device may be the above-mentioned terminal device or access network device.

[0190] The specific contents of the communication method 600 will be described below.

[0191] S600 (optional step): The third control entity sends a second registration instruction to the third processing entity, and in response, the third processing entity receives the second registration instruction, where the second registration instruction includes identification information of the second entity.

[0192] In a possible implementation manner, the third control entity selects a third processing entity from the one or more processing entities, selects a second entity from the one or more self-adaptive entities, and sends a registration instruction including identification information of the second entity to the third processing entity.

[0193] Optionally, the third control entity selects the third processing entity and the second entity based on a task type, which may correspond to one or more of the following: demodulation, channel decoding, and source decoding.

[0194] In a possible implementation, the third control entity is a protocol layer entity above the physical layer, which may also be referred to as an upper layer entity. For example, the third control entity is a MAC layer entity, and the registration instruction is MAC layer signaling.

[0195] When the first processing entity and the third processing entity jointly realize data reception and transmission, the third control entity interacts with the first control entity through an interaction mechanism of a protocol layer (e.g., a MAC layer protocol) corresponding to the third control entity to ensure that the third processing entity and the first processing entity are registered with the same second entity.

[0196] S601: A third processing entity sends a second registration request to a second entity, and in response, the second entity receives the second registration request, where the second registration request includes a first identifier.

[0197] In a possible implementation manner, the third processing entity sends a second registration request to the second entity based on the second registration instruction received in S600.

[0198] In another possible implementation manner, the third processing entity selects the second entity based on a task type corresponding to the third processing entity, and sends a second registration request to the second entity.

[0199] Specifically, the first identifier indicates one or more of model information or data information of the third processing entity.

[0200] The model information may be referred to as a model description and may include one or more of the following: model purpose, model type, model scale, model accuracy, or model performance.

[0201] The model objective indicates the task type corresponding to the third processing entity. For example, the model objective includes one or more of the following: demodulation, channel decoding, and source decoding. For other model information, please refer to the related description in S501. The details will not be described again here.

[0202] The data information may be referred to as a data description. The data information may include one or more of the following: data type, data dimension, and data precision.

[0203] The data type indicates the type of data that can be processed by the model. The data dimension indicates the input dimension and / or output dimension of the model. The data dimension may include the number of elements in the input data vector or the number of elements in the output data vector. For other description parameters related to data information, please refer to the relevant description in S501. The details will not be described again here.

[0204] S602: The second entity sends a second registration response to the third processing entity, and in response, the third processing entity receives the second registration response, where the second registration response includes output parameters of the second entity.

[0205] In a possible implementation, the output parameters of the second entity include one or more of the type, dimension or precision of the output data of the second entity.

[0206] For other aspects of the second entity sending the second registration response to the third processing entity, please refer to the relevant description of the second entity sending the first registration response to the first processing entity in S502, and the details will not be described again here.

[0207] Optionally, when the third processing entity determines that the model information and / or data information does not meet the requirements of the second entity, the second entity may feed back a registration failure indication to the third processing entity, and optionally, the third processing entity reports the registration failure indication to the third control entity.

[0208] Alternatively, when the third processing entity determines that the model information and / or data information does not meet the requirements of the second entity, the second entity may not feedback a registration failure indication but may implicitly indicate a registration failure. When sending the second registration request, the third processing entity starts a second registration response time measurement. If a second registration response is received before the time measurement expires, the time measurement is reset. If the time measurement expires, a second registration response timeout indication is sent to the third control entity.

[0209] S603 (not shown separately in the drawings): The third processing entity sends a second response confirmation response based on the received second registration response, and the second response confirmation response indicates whether the output parameters of the third processing entity match the input parameters of the second entity.

[0210] The input parameters of the third processing entity include one or more of the type, dimension or precision of the input data of the third processing entity.

[0211] Specifically, before sending the second response acknowledgment, the third processing entity determines whether the input parameters of the third processing entity are compatible with the output parameters of the second entity, for example, whether the type of input data of the third processing entity is compatible with the type of output data of the second entity, whether the dimension of the input data of the third processing entity is compatible with the dimension of the output data of the second entity, and / or whether the precision of the input data of the third processing entity is compatible with the precision of the output data of the second entity.

[0212] When the third processing entity determines that the input parameters of the third processing entity are successfully matched with the output parameters of the second entity, the third processing entity sends a second response confirmation response to the second entity (S603a). If the match fails, the third processing entity sends a second response confirmation response to the third control entity (S603b).

[0213] S603a: The third processing entity sends a second response confirmation response to the second entity, where the second response confirmation response indicates that the input parameters of the third processing entity successfully match the output parameters of the second entity, and in response, the second entity receives the second response confirmation response.

[0214] In a possible implementation, after determining that the input parameters of the third processing entity are successfully matched with the output parameters of the second entity, the third processing entity receives the second output data transmitted by the second entity through the second interface, and performs second processing on the second output data to obtain third data. For specific details, please refer to the following related description in Figure 10.

[0215] S603b: The third processing entity sends a second response acknowledgment to the third control entity, where the second response acknowledgment indicates that the input parameters of the third processing entity are incompatible with the output parameters of the second entity, and in response, the third control entity receives the second response acknowledgment.

[0216] Optionally, when the third processing entity determines that the input parameters of the third processing entity are incompatible with the output parameters of the second entity, the third processing entity may feed back a parameter compatibility failure indication to the second entity.

[0217] Alternatively, when it determines that the input parameters of the third processing entity are incompatible with the output parameters of the second entity, the third processing entity may not feed back a parameter match failure indication. When sending the second registration response, the second entity starts a second response acknowledgment time measurement. If a second response acknowledgment is received before the time measurement expires, the time measurement is reset. If the time measurement expires, a second response acknowledgment timeout indication is sent to the third control entity. The second entity learns of the parameter match failure in an implicit manner, which can reduce signaling overhead.

[0218] The model corresponding to the third processing entity may be obtained through offline training, and the model corresponding to the second entity may be adjusted based on the environment information. In the actual communication process, the third processing entity does not need to retrain its model, but performs registration with the appropriate second entity through the registration procedure shown in Figure 6 (model information and / or data information match, and input parameters match output parameters), and then performs data processing and transmission. This method can achieve low complexity in adapting to changes in the actual environment and achieve effective data reception and transmission.

[0219] In communication method 500 and communication method 600, three steps related to the registration process, namely, registration request, registration response, and response acknowledgment, may each cause a registration failure. For example, a registration response timeout, a response acknowledgment timeout, or a mismatch between input and output parameters between the two entities may cause a processing entity to fail to register with the second entity. For a failure in any registration step, registration may be achieved by selecting a new first processing entity / third processing entity to initiate a registration procedure with the second entity, or by initiating a new registration procedure with the new second entity by the first processing entity / third processing entity. Figures 7-9 describe some possible procedures after a registration failure according to embodiments of this application.

[0220] As shown in Fig. 7, an embodiment of the present application provides a communication method 700. The method describes a processing method for when a registration failure occurs after a first processing entity sends a registration request to a second entity. The communication method 700 is described by using an example in which a first processing entity registers with a second entity, and is also applicable to a scenario in which a third processing entity registers with the second entity. The method involves a first control entity, a first processing entity, a second entity, and a fourth entity, where the fourth entity is similar to the second entity, and the model of the fourth entity may be modulated based on environment information.

[0221] The specific contents of the communication method 700 will be described below.

[0222] S700: The first control entity sends a first registration instruction to the first processing entity, and in response, the first processing entity receives the registration instruction, where the registration instruction includes identification information of the second entity.

[0223] S701: A first processing entity sends a first registration request to a second entity, and in response, the second entity receives the first registration request, where the first registration request includes a first identifier.

[0224] For the specific contents of S700 and S710, please refer to S500 and S501, and the details will not be described again here.

[0225] S702: When sending a first registration request, the first processing entity starts a first registration response time measurement.

[0226] When the first registration response sent by the second entity is received before the first registration response time measurement expires, the time measurement is stopped or reset. The time measurement period may be predefined.

[0227] When the first registration response time measurement expires, step S703 is executed.

[0228] S703: The first processing entity sends a first registration response timeout indication to the first control entity, and correspondingly, the first control entity receives the first registration response timeout indication.

[0229] S704: The first control entity sends a third registration instruction to the first processing entity, and the first processing entity receives the third registration instruction in response. The third registration instruction includes identification information of the fourth entity.

[0230] Specifically, after receiving the first registration response timeout instruction, the first control entity selects another entity, i.e., a fourth entity, from the one or more self-adaptive entities, and sends a third registration instruction to the first processing entity, including identification information of the fourth entity.

[0231] Optionally, the fourth entity may alternatively be an entity obtained by updating the second entity based on the output parameters of the first processing entity.

[0232] S705: The first processing entity sends a third registration request to the fourth entity, and the fourth entity correspondingly receives the third registration request, where the third registration request includes the first identifier. For the specific content of S705, please refer to S501. The details will not be described again here.

[0233] In method 700, after sending a first registration request, the first processing entity starts a first registration response time measurement. When the time measurement expires, the first processing entity reports a timeout indication to the first control entity. The first control entity indicates other self-adaptive entities to the first processing entity to avoid long waiting times for the first processing entity and improve efficiency.

[0234] As shown in FIG. 8 , an embodiment of the present application provides a communication method 800. The method describes a processing method for subsequent operations after a registration failure occurs after a second entity sends a registration response to a first processing entity, for example, when the registration failure is caused by a mismatch between an input parameter of the second entity and an output parameter of the first processing entity. The communication method 800 is described by using an example in which a first processing entity is registered with a second entity, but is also applicable to a scenario in which a third processing entity is registered with the second entity. The method involves a first control entity, a first processing entity, a second entity, and a fourth entity, where the fourth entity is similar to the second entity, and the model of the fourth entity may be modulated based on environmental information.

[0235] S800 (optional step): The first control entity sends a first registration instruction to the first processing entity, and in response, the first processing entity receives the registration instruction, where the registration instruction includes identification information of the second entity.

[0236] S801: A first processing entity sends a first registration request to a second entity, and in response, the second entity receives the first registration request, where the first registration request includes a first identifier.

[0237] S802: The second entity sends a first registration response to the first processing entity, and in response, the first processing entity receives the first registration response, where the first registration response includes input parameters of the second entity.

[0238] For a specific description of S800 to S802, please refer to S500 to S502, and the details will not be described again here.

[0239] S803: The first processing entity sends a first response acknowledgment to the first control entity, where the first response acknowledgment indicates that the output parameters of the first processing entity are incompatible with the input parameters of the second entity, and in response, the first control entity receives the first response acknowledgment.

[0240] The output parameters of the first processing entity include one or more of the following contents of the output data of the first processing entity: type, dimension, or precision. Specifically, before sending the first response acknowledgment, the first processing entity determines whether the output parameters of the first processing entity are compatible with the input parameters of the second entity, for example, whether the type of the output data of the first processing entity is compatible with the type of the input data of the second entity, whether the dimension of the output data of the first processing entity is compatible with the dimension of the input data of the second entity, and / or whether the precision of the output data of the first processing entity is compatible with the precision of the input data of the second entity.

[0241] When the first processing entity determines that the output parameters of the first processing entity are incompatible with the input parameters of the second entity, the first processing entity sends a first response acknowledgment to the first control entity.

[0242] S804: The first control entity sends a third registration instruction to the first processing entity, and the first processing entity receives the third registration instruction in response. The third registration instruction includes identification information of the fourth entity.

[0243] Specifically, after receiving a first acknowledgment indicating an adaptation failure, the first control entity selects another entity, i.e., a fourth entity, from the one or more self-adaptive entities and sends a third registration instruction to the first processing entity, the third registration instruction including identification information of the fourth entity.

[0244] S805: The first processing entity sends a third registration request to the fourth entity, and the fourth entity correspondingly receives the third registration request, where the third registration request includes the first identifier. For the specific content of S705, please refer to S501. The details will not be described again here.

[0245] In method 800, when the first processing entity determines that the output parameters of the first processing entity do not match the input parameters of the second entity indicated in the first registration response, the first processing entity sends a first acknowledgment response indicating the match failure to the first control entity. The first control entity indicates to the first processing entity other self-adaptive entities with which the first processing entity can register, thereby avoiding multiple registrations with the second entity. Thus, efficiency is improved.

[0246] As shown in FIG. 9 , an embodiment of the present application provides a communication method 900. The method describes subsequent operations after a registration failure caused by a timeout of a first registration response time measurement. The communication method 900 is described by using an example in which a first processing entity is registered with a second entity, and is also applicable to a scenario in which a third processing entity is registered with the second entity. The method involves a first control entity, a first processing entity, a second entity, and a fourth processing entity, where the fourth processing entity is similar to the first processing entity, and a model corresponding to the fourth processing entity is obtained through offline training.

[0247] S900 (optional step): The first control entity sends a first registration instruction to the first processing entity, and in response, the first processing entity receives the registration instruction, where the registration instruction includes identification information of the second entity.

[0248] S901: A first processing entity sends a first registration request to a second entity, and in response, the second entity receives the first registration request, where the first registration request includes a first identifier.

[0249] S902: The second entity sends a first registration response to the first processing entity, and in response, the first processing entity receives the first registration response, where the first registration response includes input parameters of the second entity.

[0250] For a specific description of S900 to S902, please refer to S500 to S502, and the details will not be described again here.

[0251] S903: When sending the first registration request, the second entity starts a first response acknowledgment time measurement.

[0252] When the first response acknowledgment sent by the second entity is received before the first response acknowledgment time measurement expires, the time measurement is stopped or reset. The time measurement period may be predefined.

[0253] When the first response confirmation response time measurement expires, step S703 is executed.

[0254] S904: The second entity sends a first response acknowledgement timeout indication to the first control entity, and in response, the first control entity receives the first response acknowledgement timeout indication.

[0255] S905: The first control entity sends a fourth registration instruction to the fourth processing entity, and in response, the fourth processing entity receives the fourth registration instruction, where the fourth registration instruction includes identification information of the second entity.

[0256] Specifically, after receiving the first response acknowledgment timeout instruction, the first control entity selects a target processing entity, i.e., a fourth processing entity, different from the first processing entity from the one or more processing entities, and sends a fourth registration instruction to the fourth processing entity, including the identifier of the second entity.

[0257] For the specific content of selecting a target processing entity from one or more processing entities by the first control entity, please refer to the relevant description in S500, and the details will not be described again here.

[0258] S906: The first processing entity sends a third registration request to the second entity, and in response, the second entity receives the third registration request, where the third registration request includes a third identifier.

[0259] Specifically, the fourth processing entity sends a third registration request to the second entity based on the received fourth registration instruction. The third registration request includes a third identifier. The third identifier indicates one or more of the model information or data information of the fourth processing entity. For specific details, please refer to the relevant description of the first identifier in S501. Details will not be described again here.

[0260] The subsequent steps of method 900 are the same as those after the second entity receives the first registration request of the first processing entity in method 500. Details will not be described again here. In method 900, when the first processing entity does not feed back the first response acknowledgment to the second entity before the first response acknowledgment time measurement expires, the second entity reports a first response acknowledgment timeout indication to the first control entity. The first control entity selects another target processing entity for registering with the second entity to avoid the first processing entity failing to register with the second entity multiple times. Therefore, efficiency is improved.

[0261] After the target processing entities at the transmitting end and the receiving end are registered with the target self-adaptive entity through any one of the above possible methods described in Figures 5 to 9, the target processing entities may perform data processing and data reception and transmission. As shown in Figure 10, an embodiment of this application provides a communication method 1000. The method is described by using a first processing entity, a second entity, and a third processing entity as examples, where the first processing entity and the second entity may be located at the transmitting end, and the third processing entity may be located at the receiving end. The output parameters of the first processing entity match with the input parameters of the second entity, and the output parameters of the second entity match with the input parameters of the third processing entity. The first processing entity performs a first processing on the data, and the third processing entity performs a third processing on the data, where the third processing may be an inverse operation of the first processing.

[0262] The specific contents of the communication method 1000 will be described below.

[0263] S1001: A first processing entity performs a first process on first input data to obtain first output data.

[0264] In a possible implementation, the first processing entity implements coding and modulation functions, the first processing including channel coding and modulation or equivalent processing of channel coding and modulation, the first input data is a bit stream obtained through source coding, and the first output data is a modulation symbol sequence.

[0265] For example, the structure of a first processing entity is shown in FIG. 11 , which includes an input adaptation module, a data space transformation module, an output adaptation module, and an optional parameter configuration module. The first input data corresponds to input 1101 in the figure and may be a bit stream obtained through source encoding. The input adaptation module is configured to perform feature embedding and extraction on the input 1101, so that the input 1101 adapts to the input dimension of the data space transformation module. The output of the input adaptation module is bit groups, and the amount of bits in each group corresponds to the input dimension of the data space transformation module. The data space transformation module is configured to perform a subspace transformation on the output of the input adaptation module to obtain a modulation symbol sequence and realize the coding and modulation function. The parameters of the data space transformation module (including the input dimension of the data space transformation module) may be determined based on one or more of the following: a channel coding code rate, a modulation order, and the number of transmission resources. Optionally, the parameter configuration module is configured to configure the parameters of the data space transformation module based on one or more of the following: a channel coding code rate, a modulation order, and the number of transmission resources. The output adaptation module is configured to adapt the output of the data space transformation module to a particular data type, a particular data dimension and a particular data precision to obtain first output data, ie, output 1102 .

[0266] For example, the input configuration module is configured to adapt the dimension of the first input data to the input dimension of the data space transformation module. The data space transformation module is configured to perform a first process on the first input data to obtain first intermediate output data. The output adaptation module is configured to perform a dimensional transformation on the first intermediate output data to obtain the first output data. The dimensional transformation on the first intermediate output data may include a change from a lower dimensionality to a higher dimensionality. For example, the dimensionality increase is achieved by replicating the first intermediate output data to deal with poor channel conditions.

[0267] In a possible implementation, the first processing entity implements quantization, source coding, channel coding and modulation functions. The first processing includes quantization, source coding, channel coding and modulation, or equivalent processes of quantization, source coding, channel coding and modulation, the first input data is a symbol sequence, and the first output data is a modulated symbol sequence. Optionally, the first input data is sensing data.

[0268] For example, the structure of a first processing entity is shown in FIG. 11 , which includes an input adaptation module, a data space transformation module, an output adaptation module, and an optional parameter configuration module. The first input data corresponds to input 1101 in the figure and may be a symbol sequence. The input adaptation module is configured to perform feature embedding and extraction on input 1101, so that input 1101 adapts to the input dimension of the data space transformation module. The output of the input adaptation module is symbol groups, and the number of symbols in each group corresponds to the input dimension of the data space transformation module. The data space transformation module is configured to perform a subspace transformation on the output of the input adaptation module to obtain a modulation symbol sequence and implement quantization, source coding, channel coding, and modulation functions. Parameters of the data space transformation module (including the input dimension of the data space transformation module) may be determined based on one or more of the quantization level, source coding scheme, channel coding code rate, modulation order, and number of transmission resources. Optionally, the parameter configuration module is configured to configure the parameters of the data space transformation module based on one or more of the following: channel coding code rate, modulation order, and number of transmission resources. The output adaptation module is configured to adapt the output of the data space transformation module to a particular data type, a particular data dimension and a particular data precision to obtain first output data, ie, output 1102 .

[0269] In a possible implementation, the data space transformation module is a neural network model, and the first processing entity is obtained through offline training.

[0270] S1002: A first processing entity transmits first output data through a first interface, and correspondingly, a second entity receives the first output data through the first interface.

[0271] S1003: The second entity executes a second process on the first output data to obtain second output data.

[0272] In a possible implementation, the second processing includes mapping the first output data to second output data, wherein the mapping from the first output data to the second output data is determined based on real-time environmental parameters.

[0273] For example, the second entity corresponds to the second processing module in Fig. 4. The structure of the second processing module is shown in Fig. 12, and includes input nodes, output nodes, and wireless channels. The number of input nodes corresponds to the input dimension of the second entity, and the number of output nodes corresponds to the output dimension of the second entity.

[0274] The possible types of mapping between input and output nodes may be expressed as follows:

number

[0275] x j is the jth input value, j∈J, where J is the number of input nodes, and y i is the i-th output value, i∈I, where I is the number of output nodes, and w ji is the weight between the jth input value and the ith output value. jimay be adjusted based on real-time environmental parameters, and may be updated by using, for example, a message passing algorithm (MPA), belief propagation (BP), minimum mean square error (MMSE), weighted minimum mean square error (WMMSE), maximum likelihood (ML), maximum a posteriori (MAP), or other algorithms. The above mapping formula is an example of a basic formula. In practical applications, the above mapping formula may be transformed, etc., which is not limited in this application.

[0276] S1004: The second entity transmits the second output data to the third processing entity through the second interface, and correspondingly, the third processing entity receives the second output data.

[0277] S1005: A third processing entity performs a third process on the second output data to obtain third data.

[0278] In a possible implementation, the first processing entity performs the coding and modulation functions, and the third processing entity performs the demodulation and channel decoding functions correspondingly. The third processing may be understood as the inverse process of the first processing. The input of the third processing entity, i.e., the second output data, is a symbol sequence, and the output of the third processing entity, i.e., the third data, is a bit sequence.

[0279] For example, the structure of a third processing entity is shown in FIG. 11 , which includes an input adaptation module, a data space transformation module, an output adaptation module, and an optional parameter configuration module. The second output data corresponds to input 1101 in the figure and may be a modulation symbol sequence. The input adaptation module is configured to perform feature embedding and extraction on the input 1101, so that the input 1101 adapts to the input dimension of the data space transformation module. The output of the input adaptation module is symbol groups, and the number of symbols in each group corresponds to the input dimension of the data space transformation module. The data space transformation module is configured to perform a subspace transformation on the output of the input adaptation module to obtain a modulation symbol sequence and realize demodulation and channel decoding functions. The parameters of the data space transformation module (including the input dimension of the data space transformation module) may be determined based on one or more of the following: a channel coding code rate, a modulation order, and the number of transmission resources. Optionally, the parameter configuration module is configured to configure the parameters of the data space transformation module based on one or more of the following: a channel coding code rate, a modulation order, and the number of transmission resources. The output adaptation module is configured to adapt the output of the data space transformation module to a subsequent module to obtain an output 1102 .

[0280] In a possible implementation, the first processing entity implements the quantization, source coding, channel coding and modulation functions, and correspondingly the third processing entity implements the demodulation, channel decoding and source decoding functions, the input of the third processing entity, i.e. the second output data, is a modulation symbol sequence, and the output of the third processing entity, i.e. the third data, is a symbol sequence.

[0281] For example, the structure of a third processing entity is shown in FIG. 11 , which includes an input adaptation module, a data space transformation module, an output adaptation module, and an optional parameter configuration module. The first output data corresponds to input 1101 in the figure and may be a modulation symbol sequence. The input adaptation module is configured to perform feature embedding and extraction on input 1101, so that input 1101 adapts to the input dimension of the data space transformation module. The output of the input adaptation module is symbol groups, and the number of symbols in each group corresponds to the input dimension of the data space transformation module. The data space transformation module is configured to perform a subspace transformation on the output of the input adaptation module to obtain a modulation symbol sequence and implement quantization, source coding, channel coding, and modulation functions. The parameters of the data space transformation module (including the input dimension of the data space transformation module) may be determined based on one or more of a quantization level, a source coding scheme, a channel coding code rate, a modulation order, and the number of transmission resources. Optionally, the parameter configuration module is configured to configure parameters of the data space transformation module based on one or more of the following: a channel coding code rate, a modulation order, and a number of transmission resources. The output adaptation module is configured to adapt an output of the data space transformation module to a subsequent module to obtain an output 1102.

[0282] In a possible implementation, the data space transformation module is a neural network model, and the third processing entity is obtained through offline training.

[0283] In a possible implementation, the first input data is sensing data, and the third data is sensing data restored by a third processing entity. Optionally, a communication device corresponding to the third processing entity may further process the sensing data to obtain a sensing result.

[0284] In the communication method 1000, data processing and data reception and transmission may be realized via a first processing entity, a second entity, and a third processing entity that are adapted to each other. The first processing entity and the third processing entity may be acquired through offline training, and the second entity may be adjusted based on real-time environmental parameters. In this way, adaptation to a changing transmission environment can be realized with low complexity, and effective data reception and transmission can be realized.

[0285] In this embodiment of the present application, the first processing entity and the third processing entity may be implemented on a neural network, and the second entity may be implemented via a self-adaptive algorithm or a self-adaptive iterative algorithm. These algorithms include, but are not limited to, MPA, BP, MMSE, WMMSE, maximum likelihood ML, MAP, and other algorithms. Figure 13 shows a specific example in which the method provided in the embodiment of the present application is implemented by using a neural network and a self-adaptive iterative algorithm. The first processing module 410 corresponds to the first processing entity, the second processing module 420 corresponds to the second entity, and the third processing module 430 corresponds to the third processing entity.

[0286] The first processing module 410 includes a neural network model NN Tx, and the third processing module 430 includes a neural network model NN Rx. NN Tx and NN Rx may be realized by using autoencoders or transformers, and are obtained by performing offline pre-training based on training samples generated through large-scale environmental parameter samples or modeling based on a mathematical model.

[0287] The second processing module 420 includes an iterative algorithm module, a wireless channel module, and optional modules such as beamforming and MIMO precoding. The iterative operation module may perform online adjustments based on real-time environmental parameters.

[0288] In a possible implementation, the real-time environmental parameters are determined based on a feedback instruction of the third processing module. The feedback instruction includes channel state information or a channel eigenvector, where the channel eigenvector is an eigenvalue vector extracted based on the channel state information. Optionally, the feedback instruction further includes capability switching information, where the capability switching information is used to enable or disable the first processing module 410 and / or the second processing module 420.

[0289] In a possible implementation, the iterative algorithm module may be implemented via MPA, BP, MMSE, WMMSE, maximum likelihood ML, MAP or other algorithms.

[0290] To realize the functions in the methods provided in the embodiments of this application, the first processing entity, the second entity, and the third processing entity each include a hardware structure and / or a software module, and may realize the above functions in the form of a hardware structure, a software module, or a hardware structure + a software module. Whether some of the above functions are performed by using a hardware structure, a software module, or a combination of a hardware structure and a software module depends on the specific application and design constraints of the technical solution.

[0291] As shown in FIG. 14 , an embodiment of the present application provides a communication device 1400. The communication device 1400 may be a terminal or a network device, a device within a terminal device or a network device, or a device usable with a terminal device or a network device. In a possible implementation, the communication device 1400 may include modules or units corresponding one-to-one to the methods / operations / steps / actions performed by the first processing entity, the second entity, and the third processing entity in the above-described method embodiments. The units may be hardware circuits or software, or may be realized by a combination of hardware circuits and software. In a possible implementation, the communication device 1400 may include a processing unit 1410 and a transceiver unit 1420. The processing unit 1410 may be configured to invoke the transceiver unit 1420 to perform receiving and / or transmitting functions.

[0292] In a possible implementation, the communications device 1400 is configured to perform operations of a first processing entity. The transceiver unit 1420 is configured to send a first registration request to a second entity, the first registration request including a first identifier, the first identifier indicating one or more of model information or data information of the communications device, and model parameters of the second entity are adjusted based on real-time environmental parameters. The transceiver unit 1420 is further configured to receive a first registration response sent by the second entity, the first registration response indicating input parameters of the second entity. The transceiver unit 1420 is further configured to send a first response confirmation response, the first response confirmation response indicating whether output parameters of the first processing entity match the input parameters of the second entity.

[0293] In a possible implementation, the communication device 1400 is configured to perform operations of a second entity. Specifically, the transceiver unit 1420 is configured to receive a first registration request sent by a first processing entity, the first registration request including a first identifier, the first identifier indicating one or more of the following information of the first processing entity: model information or data information, and model parameters of the communication device are adjusted based on real-time environmental parameters. The transceiver unit 1420 is configured to send a first registration response to the first processing entity, the first registration response indicating input parameters of the communication device, and the input parameters of the communication device are used to determine whether output parameters of the first processing entity are compatible with the input parameters of the communication device.

[0294] In a possible implementation, the first processing entity and the second entity belong to the same communications device, i.e., communications device 1400 may implement both the operations performed by the first processing entity and the operations performed by the second processing entity.

[0295] In a possible implementation, the communications device 1400 is configured to perform the operations of a third processing entity. The transceiver unit 1420 is configured to send a second registration request to the second entity, the second registration request including a second identifier, the second identifier indicating one or more of model information or data information of the communications device, and the model parameters of the second entity are adjusted based on real-time environmental parameters. The transceiver unit 1420 is further configured to receive a second registration response sent by the second entity, the second registration response indicating output parameters of the second entity. The transceiver unit 1420 is further configured to send a second response acknowledgment, the second response acknowledgment indicating whether the input parameters of the communications device match the output parameters of the second entity.

[0296] In a possible implementation, the communications device 1400 is configured to perform the operations of a second entity.

[0297] The functional modules or units in the embodiments of this application may be integrated into one processor, and each of the modules or units may exist physically independently, or at least two modules or units may be integrated into one module or unit. The integrated modules or units may be realized in the form of hardware or in the form of a software functional module. In a possible implementation, the processing unit 1410 may be a processor, and the transceiver unit 1420 may be a transceiver.

[0298] Referring to FIG. 15 , an embodiment of the present application further provides a communication device 1500. The communication device 1500 is configured to implement the functions of the first processing entity, the second entity, or the third processing entity in the above-mentioned method. The communication device may be a terminal, a network device (a source access network device, a target access network device, or a computing management function), a device (e.g., a chip or circuit) within a terminal or a network device, or a device usable with a terminal or a network device. The communication device 1500 includes at least one processor 1510, and may further include a communication interface 1520. In this embodiment of the present application, the communication interface may be a transceiver, a circuit, a bus, a module, or another type of communication interface, configured to communicate with other devices by using a transmission medium. The communication interface 1520 is, for example, used by a device within the communication device 1500 to communicate with other devices.

[0299] The processor 1510 may perform functions performed by the processing unit 1410 of the communications device 1400. The communications interface 1520 may be configured to perform functions performed by the transceiver unit 1420 within the communications device 1400.

[0300] In a possible implementation, when the communication device 1500 is configured to execute the operation performed by the first processing entity, the communication interface 1520 is configured to send a first registration request to the second entity, the first registration request including a first identifier, the first identifier indicating one or more of model information or data information of the communication device, and model parameters of the second entity are adjusted based on real-time environmental parameters. The communication interface 1520 is further configured to receive a first registration response sent by the second entity, the first registration response indicating input parameters of the second entity. The communication interface 1520 is further configured to send a first response confirmation response, the first response confirmation response indicating whether the output parameters of the first processing entity match the input parameters of the second entity.

[0301] In a possible implementation, when the communication device 1500 is configured to execute an operation performed by a second entity, the communication interface 1520 is configured to receive a first registration request sent by a first processing entity, the first registration request including a first identifier, the first identifier indicating one or more of the following information of the first processing entity: model information or data information, and model parameters of the communication device are adjusted based on real-time environmental parameters. The communication interface 1520 is further configured to send a first registration response to the first processing entity, the first registration response indicating input parameters of the communication device, and the input parameters of the communication device are used to determine whether output parameters of the first processing entity match the input parameters of the communication device.

[0302] In a possible implementation, when the communication device 1500 is configured to perform the operation performed by the second entity, the communication interface 1520 is configured to receive a second registration request sent by a third processing entity, the second registration request including a second identifier, the second identifier indicating one or more of model information or data information of the third processing entity, and model parameters of the communication device are adjusted based on the real-time environmental parameters. The communication interface 1520 is further configured to send a second registration response to the third processing entity, the second registration response indicating output parameters of the communication device, and the output parameters of the communication device are used to determine whether input parameters of the third processing entity match the output parameters of the communication device.

[0303] In a possible implementation, when the communication device 1500 is configured to execute the operation performed by the third processing entity, the communication interface 1520 is configured to send a second registration request to the second entity, the second registration request including a second identifier, the second identifier indicating one or more of model information or data information of the communication device, and the model parameters of the second entity are adjusted based on the real-time environmental parameters. The communication interface 1520 is further configured to receive a second registration response sent by the second entity, the second registration response indicating output parameters of the second entity. The communication interface 1520 is further configured to send a second response confirmation response, the second response confirmation response indicating whether the input parameters of the communication device match the output parameters of the second entity.

[0304] The communication device 1500 may further include at least one memory 1530 configured to store program instructions and / or data. The memory 1530 is coupled to the processor 1510. A coupling in this embodiment of the application is an indirect coupling or communication connection between devices, units, or modules, and may be an electrical, mechanical, or other form for information exchange between the devices, units, or modules. The processor 1510 may cooperate with the memory 1530. The processor 1510 may execute computer programs or instructions stored in the memory 1530. In a possible implementation, at least one of the at least one memory may be integrated with the processor. In another possible implementation, the memory 1530 is located external to the communication device 1500.

[0305] The specific connection medium between the communication interface 1520, the processor 1510, and the memory 1530 is not limited in this embodiment of the application. In this embodiment of the application, in FIG. 15, the memory 1530, the processor 1510, and the communication interface 1520 are connected to each other through a bus 1540. In FIG. 15, the bus is represented by a thick line. The connection manner between the other components is merely an example for explanation and does not impose limitations. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, in FIG. 15, the bus is represented by using only one thick line. However, this does not indicate that there is only one bus or only one type of bus.

[0306] In a possible implementation, the communication device 1500 may be a chip system, which in this embodiment of the application may include a chip, or may include a chip and other discrete components.

[0307] Refer to Figure 16. An embodiment of the present application further provides a communication device 1600. The communication device 1600 is configured to implement the functions of the first processing entity, the second entity, or the third processing entity in the above-mentioned method. The communication device may be a terminal or a network device, or may be a device (e.g., a chip or circuit) within the terminal or network device, or may be a device usable with the terminal or network device. The communication device includes a processor 1610. The processor is configured to implement some or all of the functions of the first communication device, the second communication device, or the terminal.

[0308] In a possible implementation, when the communication device 1600 is configured to execute the operation performed by the first processing entity, the processor 1610 is configured to send a first registration request to the second entity, the first registration request including a first identifier, the first identifier indicating one or more of model information or data information of the communication device, and model parameters of the second entity are adjusted based on real-time environmental parameters. The processor 1610 is further configured to receive a first registration response sent by the second entity, the first registration response indicating input parameters of the second entity. The processor 1610 is further configured to send a first response acknowledgment, the first response acknowledgment indicating whether the output parameters of the first processing entity match the input parameters of the second entity.

[0309] In a possible implementation, when the communication device 1600 is configured to execute the operation performed by a second entity, the processor 1610 is configured to receive a first registration request sent by the first processing entity, the first registration request including a first identifier, the first identifier indicating one or more of the following information of the first processing entity: model information or data information, and model parameters of the communication device are adjusted based on the real-time environmental parameters. The processor 1610 is further configured to send a first registration response to the first processing entity, the first registration response indicating input parameters of the communication device, and the input parameters of the communication device are used to determine whether output parameters of the first processing entity match the input parameters of the communication device.

[0310] In a possible implementation, when the communication device 1600 is configured to perform the operation performed by the second entity, the processor 1610 is configured to receive a second registration request sent by a third processing entity, the second registration request including a second identifier, the second identifier indicating one or more of model information or data information of the third processing entity, and model parameters of the communication device are adjusted based on the real-time environmental parameters. The processor 1610 is further configured to send a second registration response to the third processing entity, the second registration response indicating output parameters of the communication device, and the output parameters of the communication device are used to determine whether input parameters of the third processing entity match the output parameters of the communication device.

[0311] In a possible implementation, when the communication device 1600 is configured to execute the operations performed by the third processing entity, the processor 1610 is configured to send a second registration request to the second entity, the second registration request including a second identifier, the second identifier indicating one or more of model information or data information of the communication device, and the model parameters of the second entity are adjusted based on the real-time environmental parameters. The processor 1610 is further configured to receive a second registration response sent by the second entity, the second registration response indicating output parameters of the second entity. The processor 1610 is further configured to send a second response acknowledgment, the second response acknowledgment indicating whether the input parameters of the communication device match the output parameters of the second entity.

[0312] In a possible implementation, the processor 1610 executes instructions stored in the memory 1620 to implement functions provided by the first processing entity, the second entity, or the third processing entity. Optionally, the communications device further includes the memory 1620. Optionally, the processor 1610 and the memory 1620 are integrated together. Optionally, the memory 1620 is external to the communications device 1600.

[0313] In a possible implementation, the processor 1610 may be a logic circuit that inputs / outputs messages or signaling through an input / output interface (not shown). The logic circuit may be a signal processor, a chip, or other integrated circuit capable of implementing the methods in this application.

[0314] When the communication device is a chip used in a terminal, the chip in the terminal realizes the functions of the terminal in the above method embodiment. The chip in the terminal receives information from another module (e.g., a radio frequency module or an antenna) in the terminal, and the information is transmitted to the terminal by another terminal or a network device. Alternatively, the chip in the terminal outputs information to another module (e.g., a radio frequency module or an antenna) in the terminal, and the information is transmitted to the other terminal or a network device by the terminal.

[0315] When the communication device is a chip used in a network device, the chip of the network device realizes the functions of the network device in the above method embodiments. The chip in the network device receives information from other modules (e.g., radio frequency modules or antennas) in the network device, and the information is transmitted to the network device by a terminal or other network device. Alternatively, the chip in the network device outputs information to other modules (e.g., radio frequency modules or antennas) in the network device, and the information is transmitted to the terminal or other network device by the network device.

[0316] In the embodiments of this application, a processor (e.g., processor 1510 or processor 1610) may be one or more central processing units (CPUs). When a processor is a CPU, the CPU may be a single-core CPU or a multi-core CPU. The processor may be a general-purpose processor, a digital signal processor, an application-specific integrated circuit, a field programmable gate array, another programmable logic device, a discrete gate, a transistor logic device, or a discrete hardware component, and may implement or perform the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor may be a microprocessor, any conventional processor, etc. The steps of the methods disclosed with reference to the embodiments of this application may be performed and completed directly by a hardware processor, or may be performed and completed using a combination of hardware and software modules in the processor.

[0317] In embodiments of this application, memory (e.g., memory 1530 or memory 1620) may include, but is not limited to, non-volatile memory such as a hard disk drive (HDD) or solid-state drive (SSD), random access memory (RAM), erasable programmable read-only memory (EPROM), read-only memory (ROM), or compact disc read-only memory (CD-ROM). Memory is, but is not limited to, any other medium that can be used to carry or store a computer program in the form of instructions or data structures and that is accessible by a computer. Memory in embodiments of this application may alternatively be a circuit or any other device capable of performing a storage function and configured to store computer programs or instructions, and / or data.

[0318] An embodiment of this application further provides a computer-readable storage medium. The computer-readable storage medium stores a computer program or instructions. The computer program or instructions are executed by a computer (e.g., a processor) to implement some or all of the steps of any of the methods performed by any of the devices in the embodiments of this application.

[0319] When the apparatus provided in this application is implemented in the form of a software functional unit and sold or used as an independent product, the apparatus may be stored in a computer-readable storage medium. The computer-readable storage medium may be any available medium accessible by a computer, or a data storage device such as a server or data center that consolidates one or more available media. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, or a magnetic tape), an optical medium (e.g., an optical disk), a semiconductor medium (e.g., a solid-state disk), etc.

[0320] An embodiment of the present application further provides a computer program product including a computer program or a group of instructions, which, when run on a computer, performs some or all of the steps of any of the methods in the above aspects.

[0321] All or part of the methods in the above embodiments may be realized by using software, hardware, firmware, or any combination thereof. When software is used to realize the embodiments, all or part of the embodiments may be realized in the form of a computer program product. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. Computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another.

[0322] This application further provides a chip or a chip system. The chip may include a processor. The chip may further include a memory (or storage module) and / or a transceiver (or communication module), or the chip is coupled to the memory (or storage module) and / or the transceiver (or communication module). The transceiver (or communication module) may be configured to support the chip in performing wired and / or wireless communication. The memory (or storage module) may be configured to store a program or a group of instructions. The processor may call the program or the group of instructions to realize the operations performed by a terminal or a network device in any one of the above method embodiments or possible implementation manners of the method embodiments. The chip system may include the chip, or may include the chip and other discrete devices such as a memory (or storage module) and / or a transceiver (or communication module).

[0323] Based on the same concept as the above method embodiment, this application further provides a communication system. The communication system may include the above first processing entity, second entity, and third processing entity. The communication system may be configured to implement the operations performed by the first processing entity, second entity, and third processing entity in any one of the above method embodiments or possible implementation manners of the method embodiment. For example, the communication system may have a structure shown in FIG. 2.

[0324] In the above embodiments, the description of each embodiment has its own focus, and for the parts not described in detail in the embodiments, reference is made to the related descriptions in other embodiments.

[0325] The above descriptions are merely some specific implementation modes of this application and are not intended to limit the scope of protection of this application. Those skilled in the art may make other changes and modifications to these embodiments within the technical scope disclosed in this application. The appended claims are intended to be interpreted as including the above method embodiments and changes and modifications that fall within the scope of this application. Therefore, the scope of protection of this application shall be subject to the scope of protection of the claims.

Claims

1. A communication method for use in a first processing entity of a first communication device, comprising: sending, by the first processing entity, a first registration request to a second entity, the first registration request including a first identifier, the first identifier indicating one or more of the following information of the first processing entity: model information or data information; model parameters of the first processing entity are obtained through offline training; and model parameters of the second entity are adjusted based on real-time environmental parameters; receiving, by the first processing entity, a first registration response sent by the second entity, the first registration response indicating input parameters of the second entity; sending a first response acknowledgment by the first processing entity, the first response acknowledgment indicating whether the output parameters of the first processing entity match the input parameters of the second entity, the output parameters of the first processing entity including one or more of the following content of the output data of the first processing entity: type, dimension or precision; and the input parameters of the second entity including one or more of the following content of the input data of the second entity: type, dimension or precision. A method comprising:

2. The first response acknowledgment indicates that the output parameters of the first processing entity have successfully matched with the input parameters of the second entity, and the method further comprises: performing, by the first processing entity, a first operation on first input data to obtain first output data, the first output data being an output of the first processing entity; transmitting, by the first processing entity, the first output data to the second entity through a first interface; The method of claim 1 further comprising:

3. the first processing entity includes an input adaptation module, a data space transformation module, and an output adaptation module; the input adaptation module is configured to adapt a dimension of the first input data to an input dimension of the data space transformation module; the data space transformation module is configured to perform the first operation on the first input data to obtain first intermediate output data; The method of claim 2 , wherein the output adaptation module is configured to perform a dimensional transformation on the first intermediate output data to obtain the first output data.

4. the first processing includes encoding and modulation, the first input data is a bit stream, and the first output data is modulation symbols; or The method of claim 2 , wherein the first processing includes quantization, source coding, channel coding and modulation, the first input data is a symbol sequence, and the first output data is a modulated symbol sequence.

5. The method of claim 2 , wherein the first interface is further for transmission of one or more of the registration request, the registration response, or the response acknowledgment.

6. The method of claim 2 , wherein the first interface defines a procedure and / or signaling for interaction between the first processing entity and the second entity.

7. The first response acknowledgment indicates a match failure, and the method further comprises: sending, by the first processing entity, a second registration request to a fourth entity, the second registration request including the first identifier, the first identifier indicating one or more of the following information of the first processing entity: the model information or the data information; and model parameters of the fourth entity are adjusted based on the real-time environmental parameters; receiving, by the first processing entity, a second registration response sent by the fourth entity, the second registration response indicating input parameters of the fourth entity; sending, by the first processing entity, a second response acknowledgment, the second response acknowledgment indicating whether the output parameters of the first processing entity match the input parameters of the fourth entity; The method of claim 1 further comprising:

8. A communication method for use with a second entity of a second communication device, comprising: receiving, by the second entity, a first registration request sent by a first processing entity, the first registration request including a first identifier, the first identifier indicating one or more of the following information of the first processing entity: model information or data information; model parameters of the first processing entity are obtained through offline training; and model parameters of the second entity are adjusted based on real-time environmental parameters; sending, by the second entity, a first registration response to the first processing entity, the first registration response indicating input parameters of the second entity, the input parameters of the second entity being used to determine whether output parameters of the first processing entity match with the input parameters of the second entity, the output parameters of the first processing entity including one or more of the following features of output data of the first processing entity: type, dimension, or precision; and the input parameters of the second entity including one or more of the following features of input data of the second entity: type, dimension, or precision. A method comprising:

9. The method comprises: The method further includes receiving, by the second entity, a first response acknowledgment, the first response acknowledgment indicating that the output parameters of the first processing entity have successfully matched with the input parameters of the second entity, the method comprising: receiving, by the second entity through a first interface, first output data, the first output data being output data of the first processing entity; performing, by the second entity, a second process on the first output data to obtain second output data; outputting, by the second entity, the second output data; The method of claim 8 further comprising:

10. The step of outputting the second output data by the second entity includes: The method of claim 9 , comprising transmitting, by the second entity, the second output data to a third communication device.

11. The step of performing, by the second entity, a second process on the first output data to obtain second output data includes: The method of claim 9 , further comprising mapping, by the second entity, the first output data to the second output data.

12. 10. The method of claim 9, wherein the first interface is further for transmission of one or more of the following: the first registration request, the first registration response, or the first response acknowledgment.

13. 10. The method of claim 9, wherein the first interface defines a procedure and / or signaling for interaction between the first processing entity and the second entity.

14. 9. The method of claim 8, wherein when sending the first registration response, the second entity starts a first response acknowledgment time measurement, and when the response acknowledgment time measurement expires, the second entity sends a first response acknowledgment timeout indication to a first control entity.

15. 15. The method of any one of claims 1 to 14, wherein the model information comprises one or more of the following: model purpose, model type, model scale, model accuracy or model performance.

16. 15. The method of any one of claims 1 to 14, wherein the data information comprises one or more of the following: a type of data to be processed, a dimension of the data to be processed, or a precision of the data to be processed.

17. The method of any one of claims 1 to 14, wherein the model parameters of the second entity are adjusted based on the real-time environmental parameters.

18. The method of any one of claims 1 to 14, wherein the first processing entity and the second entity belong to the same communication device.

19. 1. A communication method comprising: sending, by a third processing entity, a second registration request to a second entity, the second registration request including a second identifier, the second identifier indicating one or more of model information or data information of the third processing entity, model parameters of the third processing entity being obtained through offline training, and the model parameters of the second entity being adjusted based on real-time environmental parameters; receiving, by the third processing entity, a second registration response sent by the second entity, the second registration response indicating output parameters of the second entity; sending, by the third processing entity, a second response acknowledgment, the second response acknowledgment indicating whether the input parameters of the third processing entity match the output parameters of the second entity, the input parameters of the third processing entity including one or more of a type, dimension, or precision of input data of the third processing entity, and the output parameters of the second entity including one or more of a type, dimension, or precision of output data of the second entity; A method comprising:

20. the second response acknowledgment indicating that the input parameters of the third processing entity match the output parameters of the second entity, the method comprising:

20. The method of claim 19, further comprising obtaining, by the third processing entity, second output data transmitted by the second entity, and performing third processing on the second output data to obtain third data.

21. the third processing entity includes an input adaptation module, a data space transformation module, and an output adaptation module; the input adaptation module is configured to adapt a dimension of the second output data to an input dimension of the data space transformation module; the data space transformation module is configured to perform the third processing on the second output data to obtain second intermediate output data; 21. The method of claim 20, wherein the output adaptation module is configured to perform a dimensional transformation on the second intermediate output data to obtain the third data.

22. the third processing includes demodulation and channel decoding, the second output data is modulation symbols, and the third data is a bit stream; or 21. The method of claim 20, wherein the third processing includes demodulation, channel decoding, and source decoding, and the second output data is a modulation symbol sequence, and the third data is a symbol sequence.

23. the second response acknowledgment indicating that the input parameters of the third processing entity are incompatible with the output parameters of the second entity, the method comprising: sending, by the third processing entity, a second registration request to a fourth entity, the second registration request including the second identifier, the second identifier indicating one or more of the model information or the data information of the third processing entity, and model parameters of the fourth entity being adjusted based on the real-time environmental parameters; receiving, by the third processing entity, a second registration response sent by the fourth entity, the second registration response indicating output parameters of the fourth entity; sending, by the third processing entity, a second response acknowledgment, the second response acknowledgment indicating whether the input parameters of the third processing entity match the input parameters of the fourth entity; 20. The method of claim 19 further comprising:

24. 1. A communication method comprising: receiving, by a second entity, a second registration request sent by a third processing entity, the second registration request including a second identifier, the second identifier indicating one or more of model information or data information of the third processing entity, model parameters of the third processing entity being obtained through offline training, and model parameters of the second entity being adjusted based on real-time environmental parameters; sending, by the second entity, a second registration response to the third processing entity, the second registration response indicating output parameters of the second entity, the output parameters of the second entity being used to determine whether input parameters of the third processing entity match the output parameters of the second entity, the input parameters of the third processing entity including one or more of a type, dimension, or precision of input data of the third processing entity, and the output parameters of the second entity including one or more of a type, dimension, or precision of output data of the second entity; A method comprising:

25. the second entity receives a second response acknowledgment, the second response acknowledgment indicating that the input parameters of the third processing entity have successfully matched with the output parameters of the second entity; and the method further comprises: performing, by the second entity, a second processing on the first output data to obtain second output data, the first output data being output data of a first processing entity; transmitting, by the second entity, the second output data through a second interface to the third processing entity; 25. The method of claim 24, further comprising:

26. The step of performing, by the second entity, a second process on the first output data to obtain second output data includes:

26. The method of claim 25, comprising: mapping, by the second entity, the first output data to the second output data, wherein a mapping scheme is adjusted based on the real-time environmental parameters.

27. 25. The method of claim 24, wherein when sending the second registration response, the second entity starts a second response acknowledgment time measurement, and when the response acknowledgment time measurement expires, the second entity sends a second response acknowledgment timeout indication to a third control entity.

28. 28. The method of any one of claims 19 to 27, wherein the model information comprises one or more of the following: model purpose, model type, model scale, model accuracy, or model performance.

29. 28. The method of any one of claims 19 to 27, wherein the data information comprises one or more of the following: a type of data to be processed, a dimension of the data to be processed, or a precision of the data to be processed.

30. A communication device comprising a module or unit configured to perform the method of any one of claims 1 to 7, a module or unit configured to perform the method of any one of claims 8 to 14, a module or unit configured to perform the method of any one of claims 19 to 23, or a module or unit configured to perform the method of any one of claims 24 to 27.

31. 1. A communication device including a processor, 26. A communications device, the processor coupled to a memory, the memory storing instructions, the processor configured to execute the instructions to enable the communications device to perform a method according to any one of claims 1 to 7, a method according to any one of claims 8 to 14, a method according to any one of claims 19 to 23, or a method according to any one of claims 24 to 27.

32. A computer-readable storage medium containing a computer program or instructions, A computer-readable storage medium, which, when the computer program or the instructions are executed on a computer, performs the method of any one of claims 1 to 14 and claims 19 to 27.

33. A computer program comprising instructions, A computer program which, when run on a computer, performs the method of any one of claims 1 to 14 and claims 19 to 27.

34. A communication system comprising an apparatus configured to perform the method of any one of claims 1 to 7 and an apparatus configured to perform the method of any one of claims 8 to 14.

35. A communication system comprising an apparatus configured to perform a method according to any one of claims 19 to 23 and an apparatus configured to perform a method according to any one of claims 24 to 27.

Images