Communication method and related equipment

Abstract

The invention provides a communication method and related equipment, which are used for improving the decoding performance of a signal receiver through first information obtained by source coding and channel coding processing and second information obtained by combining source and channel coding processing so as to improve the accuracy of service data obtained by the signal receiver and hope to improve the communication efficiency. In the method, first equipment determines first information, and the first information is obtained by carrying out information source coding processing and channel coding processing on service data; the first device determines second information, wherein the second information is obtained by carrying out joint source channel coding processing on the service data; and the first device sends the first information and the second information.

Description

Technical Field

[0001] The present application relates to the field of communication technology, and in particular to a communication method and related equipment. Background Art

[0002] Wireless communication may refer to the transmission communication between multiple communication nodes without propagation through conductors or cables. Generally, network devices and terminal devices may be used as different communication nodes to communicate based on wireless communication.

[0003] At present, in a wireless communication system, a communication node as a signal sender can send a signal after source coding and channel coding the service data to be sent; correspondingly, a communication node as a signal receiver can perform channel decoding and source decoding on the received signal to restore the service data (or obtain an estimate of the service data). Among them, the signal sender needs to perform channel coding based on the channel estimation result (or channel prediction result), and correspondingly, the signal receiver needs to perform channel decoding based on the channel estimation result.

[0004] However, in actual scenarios, when the actual channel noise is different from the predicted channel noise, it is very likely that the above channel estimation result will be inaccurate, which will in turn reduce the accuracy of the service data obtained by the signal receiver and affect the communication efficiency. Summary of the invention

[0005] The present application provides a communication method and related equipment, which are used to improve the decoding performance of a signal receiver by obtaining first information through source coding and channel coding, and second information obtained through joint source-channel coding, so as to improve the accuracy of business data obtained by the signal receiver, so as to improve communication efficiency.

[0006] In a first aspect, the present application provides a communication method, which is executed by a first device, or the method is executed by some components in the first device (such as a processor, a chip or a chip system, etc.), or the method can also be implemented by a logic module or software that can realize all or part of the functions of the first device. In the first aspect and its possible implementation, the communication method is described as being executed by a first device, and the first device can be a terminal device or a network device. In this method, the first device determines first information, and the first information is obtained by processing the service data through source coding and channel coding; the first device determines second information, and the second information is obtained by processing the service data through joint source channel coding; the first device sends the first information and the second information.

[0007] Based on the above technical solution, the first device acts as a signal sender, and the first information sent by the first device is obtained by processing the service data through source coding and channel coding, and the second information sent by the first device is obtained by processing the service data through joint source channel coding. In other words, the signal receiver can not only obtain the first information obtained by processing the service data through source coding and channel coding, but also obtain the second information obtained by processing the service data through joint source channel coding. The subsequent signal receiver can receive the estimate of the first information and the estimate of the second information through the wireless channel, and determine the service data based on the estimate of the first information and the estimate of the second information. Thus, the first information obtained through source coding and channel coding, and the second information obtained through joint source channel coding, can improve the decoding performance of the signal receiver, so as to improve the accuracy of the service data obtained by the signal receiver, in order to improve the communication efficiency.

[0008] In a possible implementation manner of the first aspect, the first device determines the second information including: the first device uses the business data as an input of a first neural network, and obtains the second information through processing by the first neural network.

[0009] Based on the above technical solution, when the first device determines the second information, the first device can obtain the second information through processing by the first neural network. Thus, through the processing of the neural network, data processing efficiency can be improved and processing delay can be reduced.

[0010] In a possible implementation manner of the first aspect, the network type of the first neural network includes a U-net network and / or a transformer network.

[0011] Based on the above technical solution, the first neural network used to obtain the second information may include a U-type network and / or a transformer network to provide a specific implementation method of the first neural network.

[0012] In a possible implementation manner of the first aspect, when the network type of the first neural network includes a U-type network, the first neural network includes a downsampling part of the U-type network.

[0013] Optionally, the upsampling portion of the U-type network is used by the second device to parse the estimate of the first information and the estimate of the second information.

[0014] Based on the above technical solution, when the network type of the first neural network includes a U-type network, the first neural network used to obtain the second information may include a downsampling part of the U-type network, and the business data is subjected to feature extraction and other processing through the downsampling part to obtain the second information.

[0015] Optionally, the downsampling part of the U-type network includes N convolutional layers and N embedding layers, where N is a positive integer; wherein each of the N embedding layers includes a linear layer, a normalization layer, and a Sigmoid layer. The downsampling part of the U-type network can also be implemented in other ways, which are only described as examples here. For example, the number of convolutional layers and embedding layers may not be equal; for another example, the embedding layer may not include one or two of the linear layer, the normalization layer, and the Sigmoid layer.

[0016] In a possible implementation of the first aspect, the first device determines the second information including: the first device uses the business data as input of a second neural network, and obtains semantic information of the business data through processing by the second neural network; the first device uses the semantic information of the business data as input of a third neural network, and obtains the second information through processing by the third neural network.

[0017] Based on the above technical solution, when the first device determines the second information, the first device can obtain the second information through the processing of the second neural network and the third neural network in sequence. Therefore, through the processing of the neural network, the data processing efficiency can be improved and the processing delay can be reduced.

[0018] In a possible implementation manner of the first aspect, the second neural network includes a generative adversarial network (GAN), or the second neural network includes a diffusion model and / or a variational auto encoder (VAE).

[0019] Based on the above technical solution, the second neural network used to obtain semantic information may include GAN and / or diffusion model, VAE, etc., to provide a specific implementation method of the second neural network.

[0020] In a possible implementation of the first aspect, the second neural network includes a diffusion model; the first device uses the business data as input of the second neural network, and obtains semantic information of the business data after processing by the second neural network, including: the first device uses the business data as input of the diffusion model, and obtains semantic information of the business data after processing by the diffusion model with T steps of Gaussian noise, where T is an integer greater than or equal to 1.

[0021] Optionally, T is an integer greater than or equal to 50.

[0022] Based on the above technical solution, when the second neural network includes a diffusion model, the first device can obtain the semantic information through a forward diffusion processing process.

[0023] In a possible implementation manner of the first aspect, the third neural network includes a deep-learning neural network (DNN).

[0024] Based on the above technical solution, the third neural network used to obtain the second information through semantic information may include a DNN to provide a specific implementation method of the third neural network.

[0025] In a possible implementation manner of the first aspect, a similarity between a probability distribution of an information sequence of the second information and a Gaussian distribution is greater than a threshold.

[0026] Optionally, the threshold is 0.8, 0.9 or 0.95 etc.

[0027] Based on the above technical solution, the similarity between the probability distribution of the information sequence of the second information obtained by the joint source channel coding processing of the business data and the Gaussian distribution is greater than the threshold, that is, the probability distribution of the information sequence of the second information is approximately Gaussian distribution, and transmission through a wireless channel in the form of an analog signal can effectively improve the bandwidth compression rate and save bandwidth resources.

[0028] In addition, when the probability distribution of the information sequence of the second information is approximately Gaussian distribution, the theoretical optimum can be achieved under the additive white Gaussian noise (AWGN) channel, thereby improving communication efficiency. Moreover, when the probability distribution of the information sequence of the second information is approximately Gaussian distribution, the privacy leakage problem generated during the transmission of semantic information can also be avoided, thereby improving communication security.

[0029] In a possible implementation manner of the first aspect, the first information and the second information are carried on different communication resources, and the communication resources include at least one of time domain resources, frequency domain resources, code domain resources and space domain resources.

[0030] Based on the above technical solution, the first information and the second information can be carried by a variety of different communication resources, thereby improving the flexibility of the solution implementation.

[0031] In a possible implementation manner of the first aspect, the business data includes at least one of picture data, video data, audio data, and text data.

[0032] Based on the above technical solution, the business data carried by the first information and the second information can be implemented in the above-mentioned multiple ways, thereby improving the flexibility of the solution implementation.

[0033] The second aspect of the present application provides a communication method, which is executed by a second device, or the method is executed by some components in the second device (such as a processor, a chip or a chip system, etc.), or the method can also be implemented by a logic module or software that can realize all or part of the functions of the second device. In the second aspect and its possible implementation, the communication method is described as being executed by a second device, and the second device can be a terminal device or a network device. In this method, the second device receives an estimate of first information and an estimate of second information, the first information is obtained by source coding and channel coding of the service data, and the second information is obtained by joint source channel coding of the service data; the second device determines the service data based on the estimate of the first information and the estimate of the second information.

[0034] Based on the above technical solution, the second device acts as a signal receiver. After the first information obtained by the service data through source coding and channel coding is transmitted through the channel, the second device can receive an estimate of the first information. After the second information obtained by the service data through joint source channel coding is transmitted through the channel, the second device can receive an estimate of the second information. In other words, the signal receiver can not only obtain the first information obtained by the service data through source coding and channel coding, but also obtain the second information obtained by the service data through joint source channel coding. In addition, the second device, as a signal receiver, can determine the service data based on the estimate of the first information and the estimate of the second information after receiving the estimate of the first information and the estimate of the second information. Thus, the first information obtained through source coding and channel coding, and the second information obtained through joint source channel coding, can improve the decoding performance of the signal receiver, so as to improve the accuracy of the service data obtained by the signal receiver, in order to improve the communication efficiency.

[0035] It should be understood that due to the inevitable signal interference / actual channel path loss / signal attenuation in the wireless communication system, the signal received by the signal receiver may not be completely identical to the signal sent by the signal sender. For this reason, the first device acts as the signal sender and the second device acts as the signal receiver. The received signal of the second device can be referred to as an estimate of the signal, a prediction of the signal, a simulated recovery of the signal, etc. Correspondingly, the information carried by the received signal can be referred to as an estimate of the information, a prediction of the information, a simulated recovery of the information, etc., that is, the above-mentioned estimate of the first information (and / or the estimate of the second information) can also be referred to as a prediction of the first information (and / or the prediction of the second information), a simulated recovery of the first information (and / or the simulated recovery of the second information), etc.

[0036] In a possible implementation of the second aspect, the second device determines the business data based on the estimate of the first information and the estimate of the second information, including: the second device uses the estimate of the first information and the estimate of the second information as inputs of a fourth neural network, and obtains the business data through processing by the fourth neural network.

[0037] Based on the above technical solution, in the process of determining the service data based on the estimation of the first information and the estimation of the second information, the second device can obtain the service data through processing by the fourth neural network. Thus, through the processing of the neural network, the data processing efficiency can be improved and the processing delay can be reduced.

[0038] In a possible implementation manner of the second aspect, the network type of the fourth neural network includes a U-type network or a transformer network.

[0039] Based on the above technical solution, the fourth neural network used to obtain business data may include a U-type network and / or a transformer network to provide a specific implementation method of the first neural network.

[0040] In a possible implementation manner of the second aspect, when the network type of the fourth neural network includes a U-type network, the fourth neural network includes an upsampling part of the U-type network.

[0041] Optionally, the downsampling part of the U-type network is used by the sender (eg, the first device) to generate the second information.

[0042] Based on the above technical solution, when the network type of the fourth neural network includes a U-type network, the fourth neural network for obtaining business data may include an upsampling part of the U-type network, and the features extracted by the downsampling part of the U-type network are decoded by the upsampling part to obtain business data.

[0043] Optionally, the upsampling part of the U-type network includes N convolutional layers and N embedding layers, where N is a positive integer; wherein each of the N embedding layers includes a linear layer, a normalization layer and a Sigmoid layer.

[0044] In a possible implementation of the second aspect, the second device determines the business data based on the estimate of the first information and the estimate of the second information, including: taking the estimate of the second information as input to a fifth neural network, and obtaining semantic information of the business data through processing by the fifth neural network; taking the semantic information of the business data and the estimate of the first information as input to a sixth neural network, and obtaining the business data through processing by the sixth neural network.

[0045] Based on the above technical solution, when the second device determines the service data, the second device can obtain the service data through the processing of the fifth neural network and the sixth neural network in sequence. Therefore, through the processing of the neural network, the data processing efficiency can be improved and the processing delay can be reduced.

[0046] In a possible implementation of the second aspect, the fifth neural network includes a generative adversarial network GAN, or the sixth neural network includes a diffusion model and / or a variational autoencoder VAE.

[0047] Based on the above technical solution, the fifth neural network for obtaining semantic information may include GAN and / or diffusion model, VAE, etc., to provide a specific implementation method of the fifth neural network.

[0048] In a possible implementation of the second aspect, the fifth neural network includes a diffusion model; the second device uses the estimate of the second information as the input of the fifth neural network, and obtains the semantic information of the business data after processing by the fifth neural network, including: using the second information as the input of the diffusion model, and processing the second information with T steps of Gaussian noise by the diffusion model to obtain the semantic information of the business data, where T is an integer greater than or equal to 1.

[0049] Optionally, T is an integer greater than or equal to 50.

[0050] Based on the above technical solution, when the fifth neural network includes a diffusion model, the second device can obtain the semantic information through a reverse diffusion processing process.

[0051] In a possible implementation manner of the second aspect, a similarity between a probability distribution of an information sequence of the semantic information of the service data and a Gaussian distribution is greater than a threshold.

[0052] Optionally, the threshold is 0.8, 0.9 or 0.95 etc.

[0053] Based on the above technical solution, the similarity between the probability distribution of the information sequence of the second information obtained by the joint source channel coding processing of the business data and the Gaussian distribution is greater than the threshold, that is, the probability distribution of the information sequence of the second information is approximately Gaussian distribution, and transmission through a wireless channel in the form of an analog signal can effectively improve the bandwidth compression rate and save bandwidth resources.

[0054] In addition, when the probability distribution of the information sequence of the second information is approximately Gaussian distribution, the theoretical optimum can be achieved under the additive white Gaussian noise (AWGN) channel, thereby improving communication efficiency. Moreover, when the probability distribution of the information sequence of the second information is approximately Gaussian distribution, the privacy leakage problem generated during the transmission of semantic information can also be avoided, thereby improving communication security.

[0055] In a possible implementation manner of the second aspect, the first information and the second information are carried on different communication resources, and the communication resources include at least one of time domain resources, frequency domain resources, code domain resources and space domain resources.

[0056] Based on the above technical solution, the first information and the second information can be carried by a variety of different communication resources, thereby improving the flexibility of the solution implementation.

[0057] In a possible implementation manner of the second aspect, the business data includes at least one of picture data, video data, audio data, and text data.

[0058] Based on the above technical solution, the business data carried by the first information and the second information can be implemented in the above-mentioned multiple ways, thereby improving the flexibility of the solution implementation.

[0059] The third aspect of the present application provides a communication device, which can implement the method in the first aspect or any possible implementation of the first aspect. The communication device includes a corresponding unit or module for executing the above method. The unit or module included in the communication device can be implemented by software and / or hardware. For example, the device can be a first device, or the device can be a component in the first device (such as a processor, a chip or a chip system, etc.), or the device can also be a logic module or software that can implement all or part of the functions of the first device.

[0060] Among them, the communication device includes a processing unit and a transceiver unit; the processing unit is used to determine the first information, which is obtained by processing the service data through source coding and channel coding; the processing unit is also used to determine the second information, which is obtained by processing the service data through joint source channel coding; the transceiver unit is used to send the first information and the second information.

[0061] In a possible implementation of the third aspect, the processing unit is used to determine the second information, including: the processing unit uses the business data as an input of the first neural network, and obtains the second information through processing by the first neural network.

[0062] In a possible implementation manner of the third aspect, the network type of the first neural network includes a U-type network and / or a transformer network.

[0063] In a possible implementation manner of the third aspect, when the network type of the first neural network includes a U-type network, the first neural network includes a downsampling part of the U-type network, and an upsampling part of the U-type network is used to parse the first information and the second information.

[0064] In a possible implementation of the third aspect, the downsampling part of the U-type network includes N convolutional layers and N embedding layers, where N is a positive integer; wherein each of the N embedding layers includes a linear layer, a normalization layer, and a Sigmoid layer.

[0065] In a possible implementation of the third aspect, the processing unit is used to determine the second information, including: the processing unit uses the business data as input of the second neural network, and obtains semantic information of the business data through processing by the second neural network; the processing unit uses the semantic information of the business data as input of the third neural network, and obtains the second information through processing by the third neural network.

[0066] In a possible implementation of the third aspect, the second neural network includes a generative adversarial network GAN, or the second neural network includes a diffusion model and / or a variational autoencoder VAE.

[0067] In a possible implementation of the third aspect, the second neural network includes a diffusion model; the first device uses the business data as input to the second neural network, and obtains semantic information of the business data after processing by the second neural network, including: the first device uses the business data as input to the diffusion model, and obtains semantic information of the business data after processing by the diffusion model with T steps of Gaussian noise, where T is an integer greater than or equal to 1.

[0068] In a possible implementation manner of the third aspect, the third neural network includes a deep neural network DNN.

[0069] In a possible implementation manner of the third aspect, a similarity between a probability distribution of an information sequence of the second information and a Gaussian distribution is greater than a threshold.

[0070] In a possible implementation manner of the third aspect, the first information and the second information are carried on different communication resources, and the communication resources include at least one of time domain resources, frequency domain resources, code domain resources and space domain resources.

[0071] In a possible implementation manner of the third aspect, the business data includes at least one of picture data, video data, audio data, and text data.

[0072] In a fourth aspect of the present application, a communication device is provided, which can implement the method in the second aspect or any possible implementation of the second aspect. The communication device includes a corresponding unit or module for executing the above method. The unit or module included in the communication device can be implemented by software and / or hardware. For example, the device can be a second device, or the device can be a component in the second device (such as a processor, a chip or a chip system, etc.), or the device can also be a logic module or software that can implement all or part of the functions of the second device.

[0073] The communication device includes a transceiver unit and a processing unit; the transceiver unit is used to receive an estimate of first information and an estimate of second information, the first information is obtained by processing the service data through source coding and channel coding, and the second information is obtained by processing the service data through joint source channel coding; the processing unit is used to determine the service data based on the estimate of the first information and the estimate of the second information.

[0074] In a possible implementation of the fourth aspect, the processing unit is used to determine the business data based on the estimate of the first information and the estimate of the second information, including: the processing unit uses the estimate of the first information and the estimate of the second information as inputs of a fourth neural network, and obtains the business data through processing by the fourth neural network.

[0075] In a possible implementation manner of the fourth aspect, the network type of the fourth neural network includes a U-type network or a transformer network.

[0076] In a possible implementation manner of the fourth aspect, when the network type of the fourth neural network includes a U-type network, the fourth neural network includes an upsampling part of the U-type network, and a downsampling part of the U-type network is used to generate the second information.

[0077] In a possible implementation of the fourth aspect, the upsampling part of the U-type network includes N convolutional layers and N embedding layers, where N is a positive integer; wherein each of the N embedding layers includes a linear layer, a normalization layer, and a Sigmoid layer.

[0078] In a possible implementation of the fourth aspect, the processing unit is used to determine the business data based on the estimate of the first information and the estimate of the second information, including: the processing unit uses the estimate of the second information as the input of a fifth neural network, and obtains semantic information of the business data after processing by the fifth neural network; the processing unit uses the semantic information of the business data and the estimate of the first information as the input of a sixth neural network, and obtains the business data after processing by the sixth neural network.

[0079] In a possible implementation of the fourth aspect, the fifth neural network includes a generative adversarial network GAN, or the sixth neural network includes a diffusion model and / or a variational autoencoder VAE.

[0080] In a possible implementation of the fourth aspect, the fifth neural network includes a diffusion model; the second device uses the estimate of the second information as the input of the fifth neural network, and obtains the semantic information of the business data after processing by the fifth neural network, including: using the estimate of the second information as the input of the diffusion model, and processing the second information by the diffusion model with T steps of Gaussian noise to obtain the semantic information of the business data, where T is an integer greater than or equal to 1.

[0081] In a possible implementation manner of the fourth aspect, a similarity between a probability distribution of an information sequence of the semantic information of the business data and a Gaussian distribution is greater than a threshold.

[0082] In a possible implementation manner of the fourth aspect, the first information and the second information are carried on different communication resources, and the communication resources include at least one of time domain resources, frequency domain resources, code domain resources and space domain resources.

[0083] In a possible implementation manner of the fourth aspect, the business data includes at least one of picture data, video data, audio data, and text data.

[0084] In a fifth aspect, the present application provides a communication device, which may be the first device in the above method embodiment, or a chip set in the first device. The communication device includes a communication interface and a processor, and optionally, also includes a memory. The memory is used to store computer programs or instructions, and the processor is coupled to the memory and the communication interface. When the processor executes the computer program or instruction, the communication device executes the method executed by the first device in the above method embodiment.

[0085] In a sixth aspect of the present application, a communication device is provided, which may be the second device in the above method embodiment, or a chip set in the second device. The communication device includes a communication interface and a processor, and optionally, also includes a memory. The memory is used to store computer programs or instructions, and the processor is coupled to the memory and the communication interface. When the processor executes the computer program or instruction, the communication device executes the method executed by the second device in the above method embodiment.

[0086] A seventh aspect of an embodiment of the present application provides a computer-readable storage medium, which is used to store one or more computer-executable instructions. When the computer-executable instructions are executed by a processor, the processor executes the method described in the first aspect or any possible implementation of the first aspect, or the processor executes the method described in the second aspect or any possible implementation of the second aspect.

[0087] An eighth aspect of an embodiment of the present application provides a computer program product (or computer program). When the computer program product is executed by the processor, the processor executes the method of the above-mentioned first aspect or any possible implementation of the first aspect, or the processor executes the above-mentioned second aspect or any possible implementation of the second aspect.

[0088] A ninth aspect of an embodiment of the present application provides a chip system, which includes at least one processor for supporting a communication device to implement the functions involved in the above-mentioned first aspect or any possible implementation of the first aspect, or for supporting a communication device to implement the functions involved in the above-mentioned second aspect or any possible implementation of the second aspect.

[0089] In a possible design, the chip system may also include a memory for storing program instructions and data necessary for the communication device. The chip system may be composed of a chip, or may include a chip and other discrete devices. Optionally, the chip system also includes an interface circuit, which provides program instructions and / or data for the at least one processor.

[0090] A tenth aspect of an embodiment of the present application provides a communication system, which includes the communication device of the third aspect and the communication device of the fourth aspect, and / or the communication system includes the communication device of the fifth aspect and the communication device of the sixth aspect.

[0091] Among them, the technical effects brought about by any design method in the third aspect to the tenth aspect can refer to the technical effects brought about by different implementation methods in the above-mentioned first aspect to the second aspect, and will not be repeated here.

[0092] It should be understood that, for components in a device, the “sending” mentioned above may be referred to as “output” and the “receiving” may be referred to as “input”. BRIEF DESCRIPTION OF THE DRAWINGS

[0093] Figure 1 A schematic diagram of a communication system provided for this application;

[0094] Figure 2 Another schematic diagram of a communication system provided for the present application;

[0095] Figure 3 Another schematic diagram of a communication system provided for the present application;

[0096] Figure 4 Another schematic diagram of a communication system provided for the present application;

[0097] Figure 5 A schematic diagram of the communication method provided by the present application;

[0098] Figure 6a Another schematic diagram of the communication method provided by the present application;

[0099] Figure 6b Another schematic diagram of the communication method provided by the present application;

[0100] Figure 6c Another schematic diagram of the communication method provided by the present application;

[0101] Figure 7 A schematic diagram of a communication device provided for this application;

[0102] Figure 8 Another schematic diagram of a communication device provided by the present application;

[0103] Fig. 9 Another schematic diagram of a communication device provided by the present application;

[0104] Fig.10 Another schematic diagram of a communication device provided by the present application. DETAILED DESCRIPTION

[0105] First, some terms in the embodiments of the present application are explained to facilitate understanding by those skilled in the art.

[0106] (1) Terminal device (or terminal, user, user terminal, terminal user, user equipment, etc.): It can be a wireless terminal device that can communicate with network equipment. The wireless terminal device can be a device that provides voice and / or data to the user, or a handheld device with wireless connection function, or other processing equipment connected to a wireless modem.

[0107] The terminal can communicate with one or more core networks or the Internet via a radio access network (RAN). The terminal can be a mobile terminal device, such as a mobile phone (or "cellular" phone, mobile phone), a computer or a data card, for example, a portable, pocket-sized, handheld, computer-built-in or vehicle-mounted mobile device. For example, the terminal can be a personal communication service (PCS) phone, a cordless phone, a session initiation protocol (SIP) phone, a wireless local loop (WLL) station, a personal digital assistant (PDA), a tablet computer (Pad), a computer with wireless transceiver function and other devices. The terminal may also be referred to as a system, a subscription unit, a subscriber station, a mobile station, a mobile station (MS), a remote station, an access point (AP), a remote terminal, an access terminal, a user terminal, a user agent, a customer premises equipment (CPE), a terminal, a user equipment (UE), a mobile terminal (MT), etc. The terminal device may also be a wearable device and a next generation communication system, for example, a terminal device in a fifth generation (5G) communication system or a terminal device in a future evolved network, etc.

[0108] In addition, the terminal involved in the present application can be widely used in various scenarios, for example, device to device (D2D), vehicle to everything (V2X) communication, machine-type communication (MTC), Internet of Things (IOT), virtual reality, augmented reality, industrial control, automatic driving, telemedicine, smart grid, smart furniture, smart office, smart wear, smart transportation, smart city, etc. The terminal can be a mobile phone, a tablet computer, a computer with wireless transceiver function, a wearable device, a vehicle, a drone, a helicopter, an airplane, a ship, a robot, a mechanical arm, a smart home device, etc. The embodiments of the present application do not limit the specific technology and specific device form adopted by the terminal.

[0109] (2) Network equipment: It can be equipment in a wireless network. For example, the network equipment can be a radio access network (RAN) node (or equipment) that connects a terminal device to a wireless network. It can be called a radio access network device, and can also be generally called a base station. At present, some examples of RAN equipment are: a new generation base station (gNodeB), a transmission reception point (TRP), an evolved Node B (eNB), a radio network controller (RNC), a Node B (NB), a base station controller (BSC), a base transceiver station (BTS), a home base station (e.g., home evolved Node B, or home Node B, HNB), a base band unit (BBU), or a wireless fidelity (Wi-Fi) access point (AP). In addition, in a network structure, the network equipment can include a centralized unit (CU) node and / or a distributed unit (DU) node.

[0110] In addition, the network device may also include a core network device, which may include, for example, an access and mobility management function (AMF), a user plane function (UPF) or a session management function (SMF).

[0111] It is understandable that the core network equipment can also be referred to as the core network element. Among them, the core network element can be used to complete functions such as registration, connection, and session management. The core network element mainly includes a network exposure function (NEF) element, a policy control function (PCF) element, an application function (AF) element, an access and mobility management function (AMF) element, a session management function module (SMF) element, and a user plane function (UPF) element.

[0112] For example, in Figure 1 In this paper, the terminal device in the communication network is UE, and the access network device is access network (AN) as an example to introduce the implementation method of the core network element mentioned above in the communication network. Figure 1 As shown, the core network element may include one or more of the following network elements:

[0113] NEF network element: used to expose the services and capabilities of the third generation partnership project (3GPP) network functions to the AF network element, and also allows the AF to provide information to the 3GPP network functions. The corresponding interface is the N33 interface;

[0114] PCF network element: used for policy management of charging and QoS policies;

[0115] AF network element, used to transmit the requirements of the application side to the network side;

[0116] AMF network element: used for mobility management, access authentication / authorization and other functions. In addition, it is also responsible for transmitting user policies between terminal devices and PCF; N1 interface is the signaling interface between UE and AMF. Since terminal devices cannot interact directly with the core network, non-access stratum (NAS) information must be transparently transmitted through AN; N2 interface is the signaling interface for AMF to request AN to allocate resources for PDU sessions, etc.;

[0117] SMF network element: used to complete UE's Internet protocol (IP) address allocation, UPF selection, billing and QoS policy control and other session management functions;

[0118] UPF network element: As the interface with the data network (DN), it is used to complete user plane data forwarding, session / flow-level billing statistics, bandwidth limitation and other functions. The N3 interface is the interface between RAN and UPF, mainly used to transfer uplink and downlink user plane data between RAN and UPF.

[0119] Optionally, Figure 1 The DN and server shown are generally not considered as core network elements. The DN is used to provide services such as operator services, Internet access services or third-party services. Generally, the server can be regarded as a part of the DN, and the server includes a service entity for implementing data processing (such as neural network data processing, video data processing, etc. involved in the embodiments below).

[0120] It is understandable that the network device may also be other devices that provide wireless communication functions for the terminal device. The embodiments of the present application do not limit the specific technology and specific device form used by the network device.

[0121] In the present application, the device for implementing the function of the network device can be a network device, or a device that can support the network device to implement the function, such as a processor, chip or chip system, or a logical module or software implementation that can implement all or part of the network device function.

[0122] (3) Configuration and pre-configuration:

[0123] In this application, configuration refers to a communication node sending configuration information or configuration parameters to other communication nodes via messages or signaling.

[0124] Pre-configuration refers to configuration information or configuration parameters that a communication node has negotiated with other communication nodes in advance. It can also refer to configuration information or configuration parameters pre-specified by a standard protocol, or it can be configuration information or configuration parameters pre-stored in a communication node. This application does not limit this.

[0125] Optionally, the communication node may be a terminal, and other communication nodes may be network devices (eg, access network devices, core network devices, servers, etc.).

[0126] Furthermore, these values ​​and parameters can be changed or updated.

[0127] (4) The terms "system" and "network" in this application may be used interchangeably. "At least one" means one or more, and "plurality" means two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B may represent: A exists alone, A and B exist at the same time, and B exists alone, where A and B may be singular or plural. The character " / " generally indicates that the associated objects before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, "at least one of A, B or C" includes A, B, C, AB, AC, BC or ABC. And, unless otherwise specified, the ordinal numbers such as "first" and "second" mentioned in the embodiments of this application are used to distinguish multiple objects, and are not used to limit the order, timing, priority or importance of multiple objects.

[0128] The following will describe the technical solutions in the embodiments of the present application in conjunction with the drawings in the embodiments of the present application. The described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0129] Figure 2 It is a schematic diagram of the architecture of the communication system 1000 applied in the embodiment of the present application.

[0130] like Figure 2 As shown, the communication system includes a wireless access network 100 and a core network 200. Optionally, the communication system 1000 may also include the Internet 300. The wireless access network 100 may include at least one wireless access network device (which may also be understood as a network device as described above, such as Figure 2 110a and 110b), and may also include at least one terminal (which may also be understood as the terminal device described above, such as Figure 2 In addition, the wireless access network device may be a macro base station (eg, Figure 2 110a), or a micro base station or an indoor station (such as Figure 2110b), it can also be a relay node or a donor node, etc. It can be understood that all or part of the functions of the wireless access network device in this application can also be implemented by software functions running on hardware, or by virtualization functions instantiated on a platform (such as a cloud platform). The embodiments of this application do not limit the specific technology and specific device form adopted by the wireless access network device. For the convenience of description, a base station is used as an example of a wireless access network device for description.

[0131] In this application, the base station and the terminal can be fixed or movable. The base station and the terminal can be deployed on land, including indoors or outdoors, handheld or vehicle-mounted, or on the water surface, or on aircraft, balloons, and artificial satellites in the air. The embodiments of this application do not limit the application scenarios of the base station and the terminal.

[0132] The roles of the base station and the terminal can be relative, for example, Figure 2 The helicopter or drone 120i in the figure can be configured as a mobile base station. For the terminal 120j that accesses the wireless access network 100 through 120i, the terminal 120i is a base station. But for the base station 110a, 120i is a terminal, that is, 110a and 120i communicate through the wireless air interface protocol. Of course, 110a and 120i can also communicate through the interface protocol between base stations. In this case, relative to 110a, 120i is also a base station. Therefore, base stations and terminals can be collectively referred to as communication devices. Figure 2 110a and 110b in the figure may be referred to as communication devices having base station functions. Figure 2 120a-120j in the figure can be called communication devices with terminal functions.

[0133] Communication between base stations and terminals, between base stations and base stations, and between terminals and terminals can be carried out through authorized spectrum, through unlicensed spectrum, or through both authorized spectrum and unlicensed spectrum. Communication can be carried out through spectrum below 6 gigahertz (GHz), through spectrum above 6 GHz, or through spectrum below 6 GHz and spectrum above 6 GHz at the same time. The embodiments of the present application do not limit the spectrum resources used for wireless communication.

[0134] In the embodiments of the present application, the functions of the base station may also be performed by a module (such as a chip) in the base station, or by a control subsystem including the base station function. The control subsystem including the base station function here may be a control center in the application scenarios of the above-mentioned terminals such as smart grid, industrial control, smart transportation, smart city, etc. The functions of the terminal may also be performed by a module (such as a chip or a modem) in the terminal, or by a device including the terminal function.

[0135] In this application, the base station sends a downlink signal (or downlink information) to the terminal, and the downlink signal (or downlink information) is carried on a downlink channel. The terminal sends an uplink signal (or uplink information) to the base station, and the uplink signal (or uplink information) is carried on an uplink channel.

[0136] In recent years, with the continuous advancement and improvement of XR technology, related industries have been booming. Today, extended reality technology has entered various fields closely related to people's production and life, such as education, entertainment, military, medical, environmental protection, transportation, public health, etc. Extended reality is a general term for various reality-related technologies, including virtual reality (VR), augmented reality (AR) and mixed reality (MR).

[0137] Optionally, VR technology mainly refers to the rendering of visual and audio scenes to simulate the visual and audio sensory stimulation of the user in the real world as much as possible. VR technology usually requires the user to wear a head mounted display (HMD) to completely replace the user's field of view with simulated visual components, and requires the user to wear headphones to provide the user with accompanying audio. In addition, it is usually necessary to perform some kind of head and movement tracking on the user in VR, so as to update the simulated visual and audio content in a timely manner, so that the visual and audio content of the user experience is consistent with the user's movements.

[0138] Optionally, AR technology mainly refers to providing visual or auditory additional information or artificially generated content in the real environment perceived by the user, wherein the user's acquisition of the real environment can be direct, that is, without intermediate sensing, processing and rendering, or indirect, that is, transmitted through sensors and other means, and further enhanced processing is performed.

[0139] Alternatively, MR mixed reality technology is an advanced form of AR, one of the ways of implementing it is to insert some virtual elements into the physical scene, with the aim of providing users with an immersive experience that these elements are part of the real scene.

[0140] The communication method and related devices provided in this application may involve the communication process of data related to XR technology, and can be applied to Figure 2 The system shown can also be applied to Figure 1 , Figure 3 and Figure 4 Or other systems, not limited here.

[0141] In the following, the terminal device is taken as UE. Figure 1 , Figure 3 and Figure 4 The implementation process of transmitting relevant data of XR technology by systems is illustrated by way of example.

[0142] exist Figure 1 In the system shown, it can be understood that the system is a server-network-UE architecture scenario. Among them, the server is used for video source encoding and decoding, rendering and other processing. Network transmission includes DN (such as fixed network), LTE / 5G and the core network of the next-generation air interface 6G (such as UPF network element) and AN. The UE can be a head-mounted XR glasses, video player, holographic projector and other devices.

[0143] exist Figure 3 In the system shown, it can be understood that the system is a UE-network-UE architecture scenario. Among them, the scenario can be an application of the tactile Internet. For example, one terminal is a tactile user in the main domain and an interface with the artificial system, and the other end is a remote control robot or remote operator in the controlled domain. The network transmission core network and access network include LTE / 5G and the next generation air interface 6G. The main domain receives audio / video feedback signals from the controlled domain. The main domain and the controlled domain are connected through a two-way communication link on the network domain with the help of various commands and feedback signals, thereby forming a global control loop.

[0144] exist Figure 4 In the system shown, it can be understood that the system is a Wi-Fi scenario, in which the cloud server transmits XR media data or ordinary video to the UE (ie, XR device) through the fixed network, WiFi router / AP / set-top box.

[0145] At present, in a wireless communication system, a communication node as a signal sender can send a signal after source coding and channel coding the service data to be sent; correspondingly, a communication node as a signal receiver can perform channel decoding and source decoding on the received signal to restore the service data (or obtain an estimate of the service data). The process of source coding and channel coding by the signal sender and channel decoding and source decoding by the signal receiver can be called digital communication.

[0146] It should be understood that the signal sender and the signal receiver can be applied to a variety of communication systems. For example, the signal sender can be a network device, and the signal receiver can be a terminal device. For another example, the signal sender can be a network device, and the signal receiver can also be a network device. For another example, the signal sender can be a terminal device, and the signal receiver can also be a terminal device. Exemplarily, the network device may include Figure 1 or Figure 3 AN in Figure 2 The wireless access network device in the wireless access network 100, Figure 4 WiFi router / WiFi AP, etc.; terminal devices may include Figure 1 UE in Figure 2 A terminal in a wireless access network 100 in Figure 3 UE1 / UE2 in Figure 4 UE in, etc.

[0147] During the communication process, the signal sender needs to perform channel coding processing based on the channel estimation result (or channel prediction result), and correspondingly, the signal receiver needs to perform channel decoding processing based on the channel estimation result.

[0148] As an implementation example, the above digital communication process is committed to improving communication capacity by increasing bandwidth, mainly focusing on the reliable transmission of data (such as XR data, including at least one of image data, video data, audio data, and text data). Among them, the digital communication system can be characterized by separately designing source coding and channel coding, and the signal sender performs (1) image compression (source coding), (2) channel coding, and (3) modulation respectively; after the signal is transmitted through the channel, the signal receiver performs corresponding demodulation, channel decoding, and decompression (i.e., source decoding).

[0149] (1) Image compression: The image is first compressed using standard codecs such as Joint Photographic Experts Group (JPEG), JPEG2000, and portable network graphics (PNG), applying discrete cosine transform (DCT) / discrete wavelet transform (DWT), quantization, and entropy coding.

[0150] (2) Channel Coding: Source coding alone cannot combat channel noise. Therefore, wireless communications require channel coding, such as Turbo codes, low-density parity check codes (LDPC codes), and polar codes.

[0151] (3) Modulation: Modulation methods are selected based on channel coding, such as binary phase shift keying (BPSK), 4-quadrature amplitude modulation (QAM), 16-QAM, and 64-QAM.

[0152] However, in actual scenarios, when the actual channel noise is different from the predicted channel noise, it is very likely that the above channel estimation result will be inaccurate, which will in turn reduce the accuracy of the service data obtained by the signal receiver, affecting the communication efficiency. For example, the problem of reduced accuracy can be called the cliff effect, that is, when the channel quality decreases, the communication quality of the digital communication scheme based on the separation of source coding and channel coding suddenly drops sharply.

[0153] In addition, in actual scenarios, due to practical constraints such as non-independent and identically distributed sources, limited encoding code length, and inability to accurately estimate channel noise, the process based on the above digital communication cannot achieve theoretical optimization. In addition, for the transmission process of image data (or video data), the quality of the reconstructed image (or video) at the decoding end is often measured by traditional reliability metrics such as peak signal-to-noise ratio (PSNR) to measure the restoration quality, without considering the user experience-oriented image (or video) perception metrics, such as Frechet inception distance (FID), learned perceptual image patch similarity (LPIPS), etc. The communication scheme based on this design wastes bandwidth resources (recent studies have shown that end-to-end image transmission design using perceptual metrics can save bandwidth).

[0154] In recent years, joint source-channel coding (JSCC) has attracted much attention as an alternative approach to improve end-to-end performance. In fact, JSCC predates digital transmission methods based on the separation theorem. For example, analog and frequency modulation (AM / FM) directly modulates the continuous-time input signal onto the carrier waveform, which is a JSCC scheme. Later, also framed in discrete-time communication, JSCC has been shown to outperform individual methods in pure image and video transmission tasks, especially in bandwidth-constrained scenarios, where JSCC is more resilient to channel noise variations. More recently, in the context of semantic communication, deep learning-based JSCC methods, such as DeepJSCC, have been shown to bring significant bandwidth reductions due to their ability to directly learn the mapping. Unlike digital transmission schemes based on separation, JSCC-based methods directly map image pixel values ​​to wireless channel input symbols. Through end-to-end neural network training, the encoder and decoder learn various channel conditions and make optimizations.

[0155] However, in the communication process based on JSCC, semantic decoding simply uses the generative model to reconstruct the image (video) without considering the architecture of reliable transmission, which may cause serious distortion. It is necessary to introduce an external semantic knowledge base to correct it, so that the distance between the reconstructed image and the natural image given in the semantic knowledge base is less than a given threshold to reduce the distortion. The construction, update, and storage of the semantic knowledge base will cause a waste of computing and bandwidth resources. In addition, the semantic information in the source image (video) may not be accurately described by language or by extracting key features. Extracting and transmitting semantic information expressed in natural language from user data is prone to privacy and security leakage of communication.

[0156] In order to solve the above problems, the present application provides a communication method and related equipment, which are used to improve the decoding performance of the signal receiver by obtaining the first information through source coding and channel coding, and the second information obtained through joint source channel coding, so as to improve the accuracy of the service data obtained by the signal receiver, so as to improve the communication efficiency.

[0157] See also Figure 5 , is a schematic diagram of an implementation of the communication method provided in this application, and the method includes the following steps.

[0158] It should be noted that Figure 5 The method is illustrated by taking the first device and the second device as the execution subjects of the interaction indication as an example, but the present application does not limit the execution subjects of the interaction indication. Figure 5In the corresponding implementation method, the execution subject in S501-S502 is the first device, and the execution subject may also be a chip, a chip system, or a processor that supports the first device to implement the method, or a logic module or software that can implement all or part of the functions of the first device. Figure 5 The second device in S501-S502 in the corresponding implementation can also be replaced by a chip, a chip system, or a processor that supports the second device to implement the method, and can also be replaced by a logic module or software that can implement all or part of the functions of the second device. The first device can be a terminal device or a network device, and the second device can also be a terminal device or a network device.

[0159] S501. A first device sends first information and second information, and correspondingly, a second device receives an estimate of the first information and an estimate of the second information, wherein the first information is service data obtained by source coding and channel coding, and the second information is service data obtained by joint source channel coding.

[0160] It should be understood that due to the inevitable signal interference / actual channel path loss / signal attenuation in the wireless communication system, the signal received by the signal receiver may not be completely identical to the signal sent by the signal sender. For this reason, the first device acts as the signal sender and the second device acts as the signal receiver. The received signal of the second device can be referred to as an estimate of the signal, a prediction of the signal, a simulated recovery of the signal, etc. Correspondingly, the information carried by the received signal can be referred to as an estimate of the information, a prediction of the information, a simulated recovery of the information, etc., that is, the above-mentioned estimate of the first information (and / or the estimate of the second information) can also be referred to as a prediction of the first information (and / or the prediction of the second information), a simulated recovery of the first information (and / or the simulated recovery of the second information), etc.

[0161] Optionally, the first information is obtained by subjecting the service data to source coding processing and channel coding processing. It can be understood that the first information is obtained by subjecting the service data to source coding processing and channel coding processing in sequence, that is, after the service data is processed by source coding to obtain a source coding result, the source coding result is processed by channel coding to obtain the first information.

[0162] In a possible implementation, before the first device sends the second information in step S501, the first device may determine the second information in advance. The first device may determine the second information in a variety of ways, which will be described below.

[0163] Method 1: The process of the first device determining the second information includes: the first device uses the business data as the input of the first neural network, and obtains the second information through processing by the first neural network.

[0164] Specifically, when the first device determines the second information, the first device can obtain the second information through processing by the first neural network. Thus, through the processing of the neural network, data processing efficiency can be improved and processing delay can be reduced.

[0165] In a possible implementation of the first method, the network type of the first neural network includes a U-net network and / or a transformer network.

[0166] Optionally, when the network type of the first neural network includes a U-type network, the first neural network includes a downsampling part of the U-type network. The first neural network used to obtain the second information may include the downsampling part of the U-type network, and the service data is processed by feature extraction and the like through the downsampling part to obtain the second information.

[0167] Optionally, the upsampling part of the U-type network is used to parse the first information and the second information.

[0168] Optionally, the downsampling part of the U-type network includes N convolutional layers and N embedding layers, where N is a positive integer; wherein each of the N embedding layers includes a linear layer, a normalization layer and a Sigmoid layer.

[0169] Method 2, the process of the first device determining the second information includes: the first device uses the business data as the input of the second neural network, and obtains the semantic information of the business data through processing by the second neural network; the first device uses the semantic information of the business data as the input of the third neural network, and obtains the second information through processing by the third neural network.

[0170] Specifically, when the first device determines the second information, the first device can obtain the second information through processing by the second neural network and the third neural network in sequence. Thus, through the processing of the neural network, the data processing efficiency can be improved and the processing delay can be reduced.

[0171] Optionally, the second neural network includes a generative adversarial network (GAN), or the second neural network includes a diffusion model and / or a variational autoencoder (VAE).

[0172] Optionally, the second neural network includes a diffusion model; the first device uses the service data as input of the second neural network, and obtains semantic information of the service data after processing by the second neural network, including: the first device uses the service data as input of the diffusion model, and obtains semantic information of the service data after processing by the diffusion model through T steps of Gaussian noise, where T is an integer greater than or equal to 1. Wherein, in the case where the second neural network includes a diffusion model, the first device can obtain the semantic information through a forward diffusion processing process.

[0173] Optionally, T is an integer greater than or equal to 50.

[0174] Optionally, the third neural network includes a deep-learning neural network (DNN).

[0175] In a possible implementation manner, the similarity between the probability distribution of the information sequence of the second information sent by the first device in step S501 and the Gaussian distribution is greater than a threshold.

[0176] Optionally, the threshold is 0.8, 0.9 or 0.95 etc.

[0177] Specifically, the similarity between the probability distribution of the information sequence of the second information obtained by the joint source channel coding processing of the business data and the Gaussian distribution is greater than a threshold, that is, the probability distribution of the information sequence of the second information is approximately Gaussian distribution, and transmission through a wireless channel in the form of an analog signal can effectively improve the bandwidth compression rate and save bandwidth resources.

[0178] In addition, when the probability distribution of the information sequence of the second information is approximately Gaussian distribution, the theoretical optimum can be achieved under the additive white Gaussian noise (AWGN) channel, thereby improving communication efficiency. Moreover, when the probability distribution of the information sequence of the second information is approximately Gaussian distribution, the privacy leakage problem generated during the transmission of semantic information can also be avoided, thereby improving communication security.

[0179] In a possible implementation, the first information and the second information are carried on different communication resources, and the communication resources include at least one of time domain resources, frequency domain resources, code domain resources, and space domain resources. In other words, the first information and the second information can be carried by a variety of different communication resources, thereby improving the flexibility of the solution implementation.

[0180] In a possible implementation, the service data includes at least one of picture data, video data, audio data, and text data. In other words, the service data carried by the first information and the second information can be implemented in the above-mentioned multiple ways, thereby improving the flexibility of the solution implementation to adapt to multiple service transmission scenarios.

[0181] S502. The second device determines business data based on the estimation of the first information and the estimation of the second information.

[0182] Optionally, the business data determined by the second device in step S502 may also be referred to as estimation of business data, prediction of business data, simulated recovery of business data, etc.

[0183] In a possible implementation manner, in step S502, the second device may determine the service data based on multiple implementation manners, which will be introduced below respectively.

[0184] Implementation method 1: In step S502, the second device uses the estimate of the first information and the estimate of the second information as inputs of the fourth neural network, and obtains the business data through processing by the fourth neural network.

[0185] Specifically, in the process of determining the service data based on the estimation of the first information and the estimation of the second information, the second device can obtain the service data through processing by the fourth neural network. Thus, through the processing of the neural network, the data processing efficiency can be improved and the processing delay can be reduced.

[0186] Optionally, the network type of the fourth neural network includes a U-type network or a transformer network.

[0187] Optionally, when the network type of the fourth neural network includes a U-type network, the fourth neural network includes an upsampling part of the U-type network.

[0188] Optionally, the downsampling part of the U-type network is used by the sender (eg, the first device) to generate the second information.

[0189] Specifically, when the network type of the fourth neural network includes a U-type network, the fourth neural network for obtaining business data may include an upsampling part of the U-type network, and the features extracted by the downsampling part of the U-type network are decoded by the upsampling part to obtain business data.

[0190] Optionally, the upsampling part of the U-type network includes N convolutional layers and N embedding layers, where N is a positive integer; wherein each of the N embedding layers includes a linear layer, a normalization layer and a Sigmoid layer.

[0191] Implementation method two, in step S502, the second device uses the estimate of the second information as the input of the fifth neural network, and obtains the semantic information of the business data through processing by the fifth neural network; and uses the semantic information of the business data and the estimate of the first information as the input of the sixth neural network, and obtains the business data through processing by the sixth neural network.

[0192] Specifically, when the second device determines the service data, the second device can obtain the service data through the processing of the fifth neural network and the sixth neural network in sequence. Thus, through the processing of the neural network, the data processing efficiency can be improved and the processing delay can be reduced.

[0193] Optionally, the fifth neural network includes a generative adversarial network (GAN), or the sixth neural network includes a diffusion model and / or a variational autoencoder (VAE).

[0194] Optionally, the fifth neural network includes a diffusion model; the second device uses the estimate of the second information as the input of the fifth neural network, and obtains the semantic information of the business data after processing by the fifth neural network, including: using the estimate of the second information as the input of the diffusion model, and processing the second information by the diffusion model with T steps of Gaussian noise to obtain the semantic information of the business data, where T is an integer greater than or equal to 1.

[0195] Optionally, T is an integer greater than or equal to 50.

[0196] Specifically, when the fifth neural network includes a diffusion model, the second device can obtain the semantic information through a reverse diffusion process.

[0197] In a possible implementation, the first information and the second information are carried on different communication resources, and the communication resources include at least one of time domain resources, frequency domain resources, code domain resources, and space domain resources. Specifically, the first information and the second information can be carried by a variety of different communication resources, thereby improving the flexibility of the solution implementation.

[0198] In a possible implementation, the service data includes at least one of picture data, video data, audio data, and text data. Specifically, the service data carried by the first information and the second information can be implemented in the above-mentioned multiple ways, thereby improving the flexibility of the solution implementation.

[0199] based on Figure 5In the technical solution shown, the second device acts as a signal receiver, and the first information received by the second device in step S501 is obtained by processing the service data through source coding and channel coding, and the second information received by the second device in step S501 is obtained by processing the service data through joint source channel coding. In other words, the signal receiver can not only obtain the first information obtained by processing the service data through source coding and channel coding, but also obtain the second information obtained by processing the service data through joint source channel coding. In addition, the second device acts as a signal receiver, and after receiving the estimate of the first information and the estimate of the second information, can determine the service data based on the estimate of the first information and the estimate of the second information in step S502. Thus, the first information obtained through source coding and channel coding, and the second information obtained through joint source channel coding, can improve the decoding performance of the signal receiver, so as to improve the accuracy of the service data obtained by the signal receiver, so as to improve the communication efficiency.

[0200] As an implementation example, the following will combine Figure 6a , Figure 6b and Figure 6c right Figure 5 The communication method shown is described as an example.

[0201] like Figure 6a As shown, in the digital communication process, the first device acts as a signal sender, and the service data that the first device intends to send can be recorded as X. X is processed by source coding (wherein the source coding can be understood as compression) to obtain Z. Thereafter, Z can be processed by channel coding and modulation to obtain the first information, which can be recorded as V. d , after being transmitted through a wireless channel (denoted as a noisy channel in the figure); due to the inevitable signal interference / actual channel path loss / signal attenuation in the wireless communication system, the signal received by the signal receiving party may not be completely identical to the signal sent by the signal sending party. The second device is the signal receiving party, and the first information received by the second device in step S501 can be denoted as

[0202] like Figure 6a As shown, in the process of semantic transmission, the first device acts as a signal sender, and the service data that the first device intends to send can be recorded as X. X is processed by a neural network (for example, the neural network can be the first neural network mentioned above, or the neural network can include the second neural network and the third neural network mentioned above) to obtain X.T , after which, X T The first information obtained by the joint source-channel encoding process can be recorded as V s , after being transmitted through a wireless channel (denoted as a noisy channel in the figure); due to the inevitable signal interference / actual channel path loss / signal attenuation in the wireless communication system, the signal received by the signal receiver may not be completely identical to the signal sent by the signal sender. The second device is the signal receiver, and the estimation of the first information received by the second device in step S501 can be recorded as

[0203] Thereafter, the second device may perform demodulation processing, channel decoding processing, etc. based on the estimation of the received first information, and the processed result may be recorded as Furthermore, the second device may perform a joint source-channel decoding process based on the estimation of the received second information, and the result after the processing may be recorded as Thereafter, the second device is processed by a neural network (for example, the neural network may be the fourth neural network mentioned above, or the neural network may include the fifth neural network and the sixth neural network mentioned above) to obtain the business data in step S502, which is recorded as Y.

[0204] Optionally, the related devices of the above processing process include a source-channel joint encoder located in the first device and a source-channel joint decoder located in the second device, and the first device implements source-channel joint encoding through its built-in algorithm (such as a neural network). After the encoded signal passes through the wireless channel, it is combined with the channel-decoded signal obtained by the digital communication system at the receiving end, and is embedded as a condition into the neural network of the diffusion process to perform source-channel joint decoding, thereby restoring the original signal.

[0205] For example, Figure 6b As shown, taking the process in which the first device uses the first neural network for processing and the second device uses the fourth neural network for processing as an example, the neural network types of the first neural network and the fourth neural network may be U-Net. The first neural network may include a downsampling part of the U-type network for performing feature extraction with the business data X, and the fourth neural network may include an upsampling part of the U-type network for decoding the features contained in the received information to restore the business data Y.

[0206] In one instance, the first device acts as a signal sender. The first device can first use standard JPEG2000 (or other compression algorithms such as JPEG, HEIF, BPG, etc.) to encode the source image (real vector), and then perform channel coding (for example, LDPC code, Turbo code, polarization code, etc.) after compression, and then perform modulation (for example, BPSK, 4-QAM, 16-QAM, 64-QAM, etc.). After the signal passes through the wireless channel, it is demodulated, channel decoded, and decompressed accordingly to obtain the decoding information of the digital communication.

[0207] Furthermore, in the first device, the source image is forward diffused according to the process ( Figure 6b The neural network encoding in the left dashed box) obtains a signal (real number vector) that approximates a Gaussian variable. After pairing each two real numbers, they are converted into complex numbers to form a complex-valued channel input, which is transmitted through the wireless channel. Figure 6b The neural network decoding in the dotted box on the right decodes the channel output through the embedding layer combined with the decoding information of the digital communication to restore the original image (real vector).

[0208] Optionally, Figure 6b The neural network parameters shown are obtained through pre-training. For the diffusion model, the general U-net architecture can be used, which consists of several 2D convolutional layers. The embedding layer uses position embedding to encode the time step t. Figure 6c As shown, each embedding layer may include a linear layer, a normalization layer, and a Sigmoid layer to combine information of time t (t is a positive integer representing the number of diffusion steps less than or equal to T) and a real vector Z from digital communication (Z is the processing result of the first information after source coding).

[0209] For example, business data Figure 6a For example, the sample space of image X can be composed of images (real-valued vectors) with height H, width W, and C color channels. Let N = H·W·C be the dimension of the input vector. Signal X is decomposed into a pair of signals (Z, X T )=fθ(X), where Z represents the compression of X, fθ represents the encoding neural network (which may include neural network parameters obtained by pre-training the data (e.g., gradient descent)), and θ represents the neural network parameters obtained by training. The other component X T The semantic information extracted by the representation is obtained by the input X according to the forward diffusion process (or processed by VAE, GAN, etc.), where X T It is obtained by adding t=T (T is a parameter in the neural network model, representing the number of diffusion steps) steps of Gaussian noise from t=0, T Approximately a Gaussian random variable. In general, T can be an integer greater than 50. The diffusion step satisfies:

[0210]

[0211] Its physical meaning is that from X T-1 To X T Step by step increase t I is the random Gaussian noise with variance, I is the identity matrix, β t is an adjustable parameter, which can be a real number between (0,1). When the step size T is large enough, X T Approximately obeys Gaussian distribution. Among them, X T Sampling can be done in the following ways:

[0212]

[0213] Then, the decomposed signal (Z,X T ) are transmitted through digital communication and semantic communication respectively. Therefore, upon receiving the first information (e.g. or ) and second information (e.g. or ), the decoder generates the reconstruction Y = gφ through the reverse diffusion process Among them, gφ is the neural network obtained by end-to-end training and learning the back diffusion process, and φ represents the neural network parameters.

[0214] From the above description, it can be seen that the technical solution provided by this application has the following characteristics:

[0215] In some embodiments, the source data is divided into two data streams, which are transmitted through classical digital communication and source-channel joint coding, respectively. Thus, the original image is processed into two data streams, the semantic information is converted into an approximate Gaussian distribution, and the compression rate is effectively improved and bandwidth resources are saved through source-channel joint coding.

[0216] In other embodiments, the source image (e.g., video I frame) is compressed by a certain unspecified lossy compression scheme to achieve the target bandwidth of digital communication. Thus, the standardized source compression scheme has been highly optimized, and the accuracy of the reconstructed image can be guaranteed by using standardized source coding combined with the existing digital communication system.

[0217] In other embodiments, the semantic information that follows the approximate Gaussian distribution is extracted by training the diffusion denoising model. Therefore, the semantic information in many images may not be accurately described by language or semantic features, and the diffusion model does not require natural language or feature extraction and is not subject to this limitation.

[0218] In other embodiments, a diffusion model is used to extract semantic information at the encoding end; conditional diffusion is used to integrate two sets of data streams at the decoding end through a neural network embedding layer to restore the image information. Thus, when the image is transmitted wirelessly, the compressed image is transmitted through a wireless channel through channel coding. When the actual channel noise is different from the predicted channel noise, a "cliff effect" will occur. This solution reduces the impact of the "cliff effect".

[0219] In other embodiments, the source data is converted into data that approximately obeys Gaussian distribution and transmitted through source-channel joint coding. Thus, semantic information is converted into a signal that approximately obeys Gaussian distribution, and the JSCC transmission method reaches the theoretical optimum under the AWGN channel, thus saving bandwidth and not causing security and privacy leakage problems caused by natural language.

[0220] In other embodiments, the conditional diffusion model is used to restore the original image using the neural network architecture of U-Net and JSCC. Thus, the image is reconstructed using the generative model, and the reconstruction effect of image compression and transmission using standards such as JPEG is improved, achieving an improvement in indicators such as SSIM and LPIPS that are more in line with user experience.

[0221] In order to implement the functions of the method provided by the present application, the device executing the method may include a hardware structure and / or a software module, and implement the functions in the form of a hardware structure, a software module, or a hardware structure plus a software module. Whether one of the functions is executed in the form of a hardware structure, a software module, or a hardware structure plus a software module depends on the specific application and design constraints of the technical solution.

[0222] See also Figure 7 The present application provides a communication device 700 , which includes a processing unit 701 and a transceiver unit 702 .

[0223] As an implementation example, the communication device 700 can implement the functions of the first device in the above method, and thus can also achieve the beneficial effects of the above method. In the present application, the communication device 700 can be a first device, or a software module, an integrated circuit or a component, such as a chip, inside the first device, without limitation. The following description takes the communication device 700 as the first device as an example.

[0224] Specifically, the processing unit 701 is used to determine the first information, which is obtained by processing the service data through source coding and channel coding; the processing unit 701 is also used to determine the second information, which is obtained by processing the service data through joint source channel coding; the transceiver unit 702 is used to send the first information and the second information.

[0225] In a possible implementation, the processing unit 701 is used to determine the second information, including: the processing unit 701 uses the business data as an input of the first neural network, and obtains the second information through processing by the first neural network.

[0226] In a possible implementation, the network type of the first neural network includes a U-type network and / or a transformer network.

[0227] In a possible implementation, when the network type of the first neural network includes a U-type network, the first neural network includes a downsampling part of the U-type network, and an upsampling part of the U-type network is used to parse the first information and the second information.

[0228] In a possible implementation, the downsampling part of the U-type network includes N convolutional layers and N embedding layers, where N is a positive integer; wherein each of the N embedding layers includes a linear layer, a normalization layer, and a Sigmoid layer.

[0229] In one possible implementation, the processing unit 701 is used to determine the second information including: the processing unit 701 uses the business data as input of the second neural network, and obtains semantic information of the business data through processing by the second neural network; the processing unit 701 uses the semantic information of the business data as input of the third neural network, and obtains the second information through processing by the third neural network.

[0230] In one possible implementation, the second neural network includes a generative adversarial network (GAN), or the second neural network includes a diffusion model and / or a variational autoencoder (VAE).

[0231] In one possible implementation, the second neural network includes a diffusion model; the first device uses the business data as input to the second neural network, and obtains semantic information of the business data after processing by the second neural network, including: the first device uses the business data as input to the diffusion model, and obtains semantic information of the business data after processing by the diffusion model with T steps of Gaussian noise, where T is an integer greater than or equal to 1.

[0232] In one possible implementation, the third neural network includes a deep neural network DNN.

[0233] In a possible implementation manner, the similarity between the probability distribution of the information sequence of the second information and the Gaussian distribution is greater than a threshold.

[0234] In a possible implementation manner, the first information and the second information are carried on different communication resources, and the communication resources include at least one of time domain resources, frequency domain resources, code domain resources and space domain resources.

[0235] In a possible implementation, the business data includes at least one of picture data, video data, audio data, and text data.

[0236] As another implementation example, the communication device 700 can implement the function of the second device in the above method, and thus can also achieve the beneficial effects of the above method. In the present application, the communication device 700 can be a second device, or a software module, an integrated circuit or a component, such as a chip, inside the second device, without limitation. The following description is made by taking the communication device 700 as the second device as an example.

[0237] Specifically, the transceiver unit 702 is used to receive an estimate of first information and an estimate of second information, the first information being obtained by processing the service data through source coding and channel coding, and the second information being obtained by processing the service data through joint source channel coding; the processing unit 701 is used to determine the service data based on the estimate of the first information and the estimate of the second information.

[0238] In one possible implementation, the processing unit 701 is used to determine the business data based on the estimate of the first information and the estimate of the second information, including: the processing unit 701 uses the estimate of the first information and the estimate of the second information as inputs of a fourth neural network, and obtains the business data after processing by the fourth neural network.

[0239] In a possible implementation, the network type of the fourth neural network includes a U-type network or a transformer network.

[0240] In a possible implementation, when the network type of the fourth neural network includes a U-type network, the fourth neural network includes an upsampling part of the U-type network, and a downsampling part of the U-type network is used to generate the second information.

[0241] In a possible implementation, the upsampling part of the U-type network includes N convolutional layers and N embedding layers, where N is a positive integer; wherein each of the N embedding layers includes a linear layer, a normalization layer, and a Sigmoid layer.

[0242] In one possible implementation, the processing unit 701 is used to determine the business data based on the estimate of the first information and the estimate of the second information, including: the processing unit 701 uses the estimate of the second information as the input of a fifth neural network, and obtains semantic information of the business data after processing by the fifth neural network; the processing unit 701 uses the semantic information of the business data and the estimate of the first information as the input of a sixth neural network, and obtains the business data after processing by the sixth neural network.

[0243] In one possible implementation, the fifth neural network includes a generative adversarial network (GAN), or the sixth neural network includes a diffusion model and / or a variational autoencoder (VAE).

[0244] In one possible implementation, the second neural network includes a diffusion model; the second device uses the estimate of the second information as the input of the fifth neural network, and obtains the semantic information of the business data after processing by the fifth neural network, including: using the estimate of the second information as the input of the diffusion model, and obtaining the semantic information of the business data after processing by the diffusion model with T steps of Gaussian noise, where T is an integer greater than or equal to 1.

[0245] In a possible implementation manner, the similarity between the probability distribution of the information sequence of the semantic information of the business data and the Gaussian distribution is greater than a threshold.

[0246] In a possible implementation manner, the first information and the second information are carried on different communication resources, and the communication resources include at least one of time domain resources, frequency domain resources, code domain resources and space domain resources.

[0247] In a possible implementation, the business data includes at least one of picture data, video data, audio data, and text data.

[0248] It should be noted that the information execution process and other contents of the units of the above-mentioned communication device 700 can be specifically referred to the description in the method embodiment shown in the above-mentioned application, and will not be repeated here.

[0249] See also Figure 8 , is another schematic structural diagram of a communication device 800 provided in the present application, and the communication device 800 at least includes an input and output interface 802. The communication device 800 may be a chip or an integrated circuit.

[0250] Optionally, the communication device also includes a logic circuit 801.

[0251] in, Figure 8 The transceiver unit 802 shown may be a communication interface, which may be Figure 8 The input / output interface 802 in the communication interface may include an input interface and an output interface. Alternatively, the communication interface may also be a transceiver circuit, which may include an input interface circuit and an output interface circuit.

[0252] Optionally, the logic circuit 801 is used to determine the first information, which is obtained by processing the service data through source coding and channel coding; the logic circuit 801 is also used to determine the second information, which is obtained by processing the service data through joint source channel coding; the input-output interface 802 is used to send the first information and the second information. The logic circuit 801 and the input-output interface 802 can also perform other steps performed by the first device in any embodiment and achieve corresponding beneficial effects, which will not be repeated here.

[0253] Optionally, the input-output interface 802 is used to receive an estimate of first information and an estimate of second information, the first information being obtained by processing the service data through source coding and channel coding, and the second information being obtained by processing the service data through joint source-channel coding; the logic circuit 801 is used to determine the service data based on the estimate of the first information and the estimate of the second information.

[0254] The logic circuit 801 and the input / output interface 802 may also execute other steps executed by the second device in any embodiment and achieve corresponding beneficial effects, which will not be described in detail here.

[0255] In one possible implementation, Figure 8 The processing unit 801 shown may be Figure 8 The logic circuit 801 in.

[0256] Optionally, the logic circuit 801 may be a processing device, and the functions of the processing device may be partially or completely implemented by software. The functions of the processing device may be partially or completely implemented by software.

[0257] Optionally, the processing device may include a memory and a processor, wherein the memory is used to store a computer program, and the processor reads and executes the computer program stored in the memory to perform corresponding processing and / or steps in any one of the method embodiments.

[0258] Alternatively, the processing device may include only a processor. A memory for storing a computer program is located outside the processing device, and the processor is connected to the memory via a circuit / wire to read and execute the computer program stored in the memory. The memory and the processor may be integrated together, or may be physically independent of each other.

[0259] Optionally, the processing device may be one or more chips, or one or more integrated circuits. For example, the processing device may be one or more field-programmable gate arrays (FPGA), application specific integrated circuits (ASIC), system on chip (SoC), central processor unit (CPU), network processor (NP), digital signal processor (DSP), microcontroller unit (MCU), programmable logic device (PLD) or other integrated chips, or any combination of the above chips or processors.

[0260] See also Fig. 9 , is a schematic diagram of the structure of the communication device involved in the above embodiments provided in the embodiments of the present application, wherein the communication device may specifically be the network device in the above embodiments, and the structure of the communication device may refer to Fig. 9 The structure shown.

[0261] The communication device includes at least one processor 911, at least one memory 912, at least one transceiver 913, and one or more antennas 914. The processor 911, the memory 912, and the transceiver 913 are connected, for example, through a bus. In the embodiment of the present application, the connection may include various interfaces, transmission lines, or buses, etc., which are not limited in this embodiment. The antenna 914 is connected to the transceiver 913.

[0262] As an implementation example, Fig. 9 When the communication device shown is the first device in the aforementioned embodiment, the processor 911 is used to determine the first information, which is obtained by processing the service data through source coding and channel coding; the processor 911 is also used to determine the second information, which is obtained by processing the service data through joint source channel coding; the transceiver 913 is used to send the first information and the second information.

[0263] As an implementation example, Fig. 9When the communication device shown is the second device in the aforementioned embodiment, the transceiver 913 is used to receive first information and second information, the first information being obtained by processing the service data through source coding and channel coding, and the second information being obtained by processing the service data through joint source channel coding; the processor 911 is used to determine the service data based on the first information and the second information.

[0264] It should be noted that the above Fig. 9 For details on the execution process of each device in the communication device, please refer to the description of the method embodiment shown in the previous section of the present application, which will not be repeated here.

[0265] The processor 911 is mainly used to process the communication protocol and communication data, and to control the entire communication device, execute the software program, and process the data of the software program, for example, to support the communication device to perform the actions described in the embodiment. The communication device may include a baseband processor and a central processing unit. The baseband processor is mainly used to process the communication protocol and communication data, and the central processing unit is mainly used to control the entire network device, execute the software program, and process the data of the software program. Fig. 9 The processor 911 in the embodiment can integrate the functions of the baseband processor and the central processor. It can be understood by those skilled in the art that the baseband processor and the central processor can also be independent processors, which are interconnected through technologies such as buses. It can be understood by those skilled in the art that the network device can include multiple baseband processors to adapt to different network formats, and the network device can include multiple central processors to enhance its processing capabilities. The various components of the network device can be connected through various buses. The baseband processor can also be described as a baseband processing circuit or a baseband processing chip. The central processor can also be described as a central processing circuit or a central processing chip. The function of processing the communication protocol and the communication data can be built into the processor, or it can be stored in the memory in the form of a software program, and the processor executes the software program to realize the baseband processing function.

[0266] The memory is mainly used to store software programs and data. The memory 912 can exist independently and be connected to the processor 911. Optionally, the memory 912 can be integrated with the processor 911, for example, integrated into a chip. Among them, the memory 912 can store program codes for executing the technical solutions of the embodiments of the present application, and the execution is controlled by the processor 911. The various types of computer program codes executed can also be regarded as drivers of the processor 911.

[0267] Fig. 9Only one memory and one processor are shown. In an actual network device, there may be multiple processors and multiple memories. The memory may also be referred to as a storage medium or a storage device, etc. The memory may be a storage element on the same chip as the processor, i.e., an on-chip storage element, or an independent storage element, which is not limited in the embodiments of the present application.

[0268] The transceiver 913 can be used to support the reception or transmission of radio frequency signals between the communication device and the terminal, and the transceiver 913 can be connected to the antenna 914. The transceiver 913 includes a transmitter Tx and a receiver Rx. Specifically, one or more antennas 914 can receive radio frequency signals, and the receiver Rx of the transceiver 913 is used to receive the radio frequency signal from the antenna, and convert the radio frequency signal into a digital baseband signal or a digital intermediate frequency signal, and provide the digital baseband signal or the digital intermediate frequency signal to the processor 911, so that the processor 911 further processes the digital baseband signal or the digital intermediate frequency signal, such as demodulation processing and decoding processing. In addition, the transmitter Tx in the transceiver 913 is also used to receive a modulated digital baseband signal or a digital intermediate frequency signal from the processor 911, and convert the modulated digital baseband signal or the digital intermediate frequency signal into a radio frequency signal, and send the radio frequency signal through one or more antennas 914. Specifically, the receiver Rx can selectively perform one or more stages of down-mixing and analog-to-digital conversion processing on the RF signal to obtain a digital baseband signal or a digital intermediate frequency signal, and the order of the down-mixing and analog-to-digital conversion processing is adjustable. The transmitter Tx can selectively perform one or more stages of up-mixing and digital-to-analog conversion processing on the modulated digital baseband signal or digital intermediate frequency signal to obtain a RF signal, and the order of the up-mixing and digital-to-analog conversion processing is adjustable. The digital baseband signal and the digital intermediate frequency signal can be collectively referred to as a digital signal.

[0269] The transceiver may also be referred to as a transceiver unit, a transceiver, a transceiver device, etc. Optionally, a device in a transceiver unit for implementing a receiving function may be regarded as a receiving unit, and a device in a transceiver unit for implementing a sending function may be regarded as a sending unit, that is, the transceiver unit includes a receiving unit and a sending unit, and the receiving unit may also be referred to as a receiver, an input port, a receiving circuit, etc., and the sending unit may be referred to as a transmitter, a transmitter, or a transmitting circuit, etc.

[0270] It should be noted that Fig. 9 The communication device shown can be specifically used to implement the steps implemented by the network device in any of the aforementioned method embodiments, and achieve the corresponding technical effects of the network device. Fig. 9 The specific implementation methods of the communication device shown can refer to the description in any of the aforementioned method embodiments, and will not be repeated here.

[0271] See also Fig.10, is a possible logical structure diagram of the communication device 1000 involved in the above-mentioned embodiments provided in an embodiment of the present application. The communication device can specifically be the terminal device in the aforementioned embodiment. The communication device 1000 can include but is not limited to a processor 1001, a communication port 1002, a memory 1003, and a bus 1004. In the embodiment of the present application, the processor 1001 is used to control and process the actions of the communication device 1000.

[0272] As an implementation example, Fig.10 The communication device shown is the aforementioned Figure 3 In the case of the first device in the related embodiments, the processor 1001 is used to determine the first information, which is obtained by processing the service data through source coding and channel coding; the processor 1001 is also used to determine the second information, which is obtained by processing the service data through joint source channel coding; the communication port 1002 is used to send the first information and the second information.

[0273] As another implementation example, Fig.10 The communication device shown is the aforementioned Figure 3 In the case of the second device in the related embodiments, the communication port 1002 is used to receive an estimate of the first information and an estimate of the second information, the first information being obtained by the service data after source coding and channel coding, and the second information being obtained by the service data after joint source channel coding; the processor 1001 is used to determine the service data based on the estimate of the first information and the estimate of the second information.

[0274] In addition, the processor 1001 can be a central processing unit, a general-purpose processor, a digital signal processor, an application-specific integrated circuit, a field programmable gate array or other programmable logic device, a transistor logic device, a hardware component or any combination thereof. It can implement or execute various exemplary logic blocks, modules and circuits described in conjunction with the disclosure of this application. The processor can also be a combination that implements a computing function, such as a combination of one or more microprocessors, a combination of a digital signal processor and a microprocessor, and the like. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working process of the above-described system, device and unit can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0275] It should be noted that Fig.10 The communication device shown can be specifically used to implement the steps implemented by the terminal device (such as the first device or the second device) in any of the aforementioned method embodiments, and achieve the corresponding technical effects of the terminal device. Fig.10The specific implementation methods of the communication device shown can refer to the description in any of the aforementioned method embodiments, and will not be repeated here.

[0276] An embodiment of the present application also provides a computer-readable storage medium. When the computer execution instructions in the storage medium are executed by a processor, the processor executes the method described in the possible implementation method of the communication device in the aforementioned embodiment, wherein the communication device can specifically be the first device or the second device in the aforementioned embodiment.

[0277] An embodiment of the present application also provides a computer program product (or computer program). When the computer program product in the program product is executed by the processor, the processor executes the method of the possible implementation method of the above-mentioned communication device, wherein the communication device can specifically be the first device or the second device in the aforementioned embodiment.

[0278] The embodiment of the present application also provides a chip system, which includes a processor for supporting a communication device to implement the functions involved in the possible implementation of the above communication device. In one possible design, the chip system may also include a memory, which is used to store the necessary program instructions and data of the communication device. The chip system may be composed of a chip, or may include a chip and other discrete devices, wherein the communication device may specifically be the first device or the second device in the above embodiment.

[0279] An embodiment of the present application further provides a communication system, which includes the above-mentioned communication device, and the communication device can specifically be the first device and the second device in any one of the above-mentioned embodiments.

[0280] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0281] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0282] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0283] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the part that essentially contributes to the technical solution of the present application or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0284] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A communication method, characterized in that: include: Determine first information, where the first information is obtained by subjecting service data to source coding and channel coding; Determine second information, where the second information is obtained by processing the service data through joint source-channel coding; The first information and the second information are sent.

2. The method according to claim 1, characterized in that The determining of the second information comprises: The business data is used as input of the first neural network, and is processed by the first neural network to obtain the second information.

3. The method according to claim 2, characterized in that The network type of the first neural network includes a U-type network and / or a transformer network.

4. The method according to claim 3, characterized in that When the network type of the first neural network includes a U-type network, the first neural network includes a downsampling part of the U-type network.

5. The method according to claim 4, characterized in that The downsampling part of the U-type network includes N convolutional layers and N embedding layers, where N is a positive integer; Each of the N embedding layers includes a linear layer, a normalization layer and a Sigmoid layer.

6. The method according to claim 1, characterized in that The determining of the second information comprises: Using the business data as input of a second neural network, and obtaining semantic information of the business data after processing by the second neural network; The semantic information of the business data is used as input of a third neural network, and the second information is obtained after being processed by the third neural network.

7. The method according to claim 6, characterized in that The second neural network includes a generative adversarial network (GAN), or the second neural network includes a diffusion model and / or a variational autoencoder (VAE).

8. The method according to claim 7, characterized in that The second neural network includes a diffusion model; The step of using the business data as input of the second neural network and obtaining semantic information of the business data after processing by the second neural network includes: The business data is used as the input of the diffusion model, and is processed by the diffusion model with T steps of Gaussian noise to obtain semantic information of the business data, where T is an integer greater than or equal to 1.

9. The method according to any one of claims 1 to 8, characterized in that: The similarity between the probability distribution of the information sequence of the second information and the Gaussian distribution is greater than a threshold.

10. The method according to any one of claims 1 to 9, characterized in that: The first information and the second information are carried on different communication resources, and the communication resources include at least one of time domain resources, frequency domain resources, code domain resources and space domain resources.

11. The method according to any one of claims 1 to 10, characterized in that: The business data includes at least one of picture data, video data, audio data, and text data.

12. A communication method, characterized in that: include: receiving an estimate of first information and an estimate of second information, wherein the first information is service data obtained by subjecting the service data to source coding and channel coding, and the second information is the service data obtained by subjecting the service data to joint source channel coding; The traffic data is determined based on the estimate of the first information and the estimate of the second information.

13. The method according to claim 12, characterized in that The determining the service data based on the estimation of the first information and the estimation of the second information comprises: The estimation of the first information and the estimation of the second information are used as inputs of a fourth neural network, and the business data is obtained after being processed by the fourth neural network.

14. The method according to claim 13, characterized in that The network type of the fourth neural network includes a U-type network or a transformer network.

15. The method according to claim 14, characterized in that When the network type of the fourth neural network includes a U-type network, the fourth neural network includes an upsampling part of the U-type network.

16. The method according to claim 15, characterized in that The upsampling part of the U-type network includes N convolutional layers and N embedding layers, where N is a positive integer; Each of the N embedding layers includes a linear layer, a normalization layer and a Sigmoid layer.

17. The method according to claim 12, characterized in that The determining the service data based on the first information and the second information comprises: Using the estimation of the second information as input of a fifth neural network, and obtaining semantic information of the business data through processing by the fifth neural network; The semantic information of the business data and the estimation of the first information are used as inputs of a sixth neural network, and the business data are obtained after being processed by the sixth neural network.

18. The method according to claim 17, characterized in that The fifth neural network includes a generative adversarial network GAN, or the sixth neural network includes a diffusion model and / or a variational autoencoder VAE.

19. The method according to claim 18, characterized in that The fifth neural network includes a diffusion model; The step of taking the estimation of the second information as the input of the fifth neural network and obtaining the semantic information of the business data after processing by the fifth neural network includes: The estimation of the second information is used as the input of the diffusion model, and the diffusion model is processed with T steps of Gaussian noise to obtain the semantic information of the business data, where T is an integer greater than or equal to 1.

20. The method according to any one of claims 12 to 19, characterized in that The similarity between the probability distribution of the information sequence of the semantic information of the business data and the Gaussian distribution is greater than a threshold.

21. The method according to any one of claims 12 to 20, characterized in that The first information and the second information are carried on different communication resources, and the communication resources include at least one of time domain resources, frequency domain resources, code domain resources and space domain resources.

22. The method according to any one of claims 12 to 21, characterized in that The business data includes at least one of picture data, video data, audio data, and text data.

23. A communication device, characterized in that: including a processing unit and a transceiver unit; The processing unit and the transceiver unit are configured to execute the method according to any one of claims 1 to 11.

24. A communication device, characterized in that: including a processing unit and a transceiver unit; The processing unit and the transceiver unit are configured to execute the method according to any one of claims 12 to 22.

25. A communication device, characterized in that: comprising at least one processor coupled to a memory; The at least one processor is used to execute the method as claimed in any one of claims 1 to 11; or, the processor is used to execute the method as claimed in any one of claims 12 to 22.

26. A communication system, characterized in that: The invention comprises a communication device for executing the method according to any one of claims 1 to 11, and a communication device for executing the method according to any one of claims 12 to 22.

27. A computer-readable storage medium, characterized in that: The medium stores instructions, and when the instructions are executed by a computer, the method according to any one of claims 1 to 22 is implemented.

28. A computer program product, characterized in that The method comprises instructions which, when executed on a computer, cause the computer to perform the method according to any one of claims 1 to 22.

Images