Communication method, device, and storage medium

Abstract

Embodiments of the present application provide a communication method, device and storage medium, the method comprising: a second device sending first information to a first device, the first information comprising channel feedback information and second information, the channel feedback information comprising a channel feedback sequence corresponding to channel information of a plurality of bandwidth units, and the second information being used to indicate that a coding and decoding result of channel information of a first bandwidth unit in the plurality of bandwidth units is inaccurate. The reliability of communication is improved.

Description

Technical Field

[0001] This application relates to communication technology, and more particularly to a communication method, device and storage medium. Background Technology

[0002] Currently, the channel information feedback method in the 5th generation (5G) new radio (NR) standard is a codebook-based feedback method. This method selects the optimal feedback codebook matrix and corresponding feedback coefficients from the codebook based on the estimated channel information. This mapping process from channel information to the codebook is lossy in terms of quantization. On the other hand, a new approach has been proposed where the channel information feedback end (e.g., the terminal) encodes the channel information into an encoded sequence using a neural network model. The channel information receiving end (e.g., the base station) can then decode the encoded sequence using the neural network model to recover the recovered channel information, which can improve the accuracy of channel information feedback. However, the parameters of the neural network model are based on training data, which cannot cover all channel scenarios. Therefore, poor feedback performance may occur in some channel scenarios. Summary of the Invention

[0003] This application provides a communication method, device, and storage medium to improve the reliability of communication.

[0004] In a first aspect, embodiments of this application may provide a communication method applied to a communication device, the method comprising: Send first information to the first device. The first information includes channel feedback information and second information. The channel feedback information includes channel feedback sequences corresponding to channel information of multiple bandwidth units. The second information is used to indicate that the encoding and decoding results of the channel information of the first bandwidth unit among the multiple bandwidth units are inaccurate.

[0005] Secondly, embodiments of this application may provide a communication method applied to a communication device, the method comprising: Receive first information from the second device, the first information including channel feedback information and second information, the channel feedback information including channel feedback sequences corresponding to channel information of multiple bandwidth units, and the second information used to indicate that the encoding and decoding results of the channel information of the first bandwidth unit among the multiple bandwidth units are inaccurate; Based on this first information, the recovery channel information of the multiple bandwidth units is obtained.

[0006] Thirdly, embodiments of this application may also provide a communication device, including: The processing unit is used to determine that the encoding and decoding result of the channel information of the first bandwidth unit among multiple bandwidth units is inaccurate; The transceiver unit is used to send first information to the first device. The first information includes channel feedback information and second information. The channel feedback information includes the channel feedback sequence corresponding to the channel information of the plurality of bandwidth units. The second information is used to indicate that the encoding and decoding result of the channel information of the first bandwidth unit among the plurality of bandwidth units is inaccurate.

[0007] Fourthly, embodiments of this application may also provide a communication device, including: The transceiver unit is used to receive first information from the second device. The first information includes channel feedback information and second information. The channel feedback information includes channel feedback sequences corresponding to channel information of multiple bandwidth units. The second information is used to indicate that the encoding and decoding results of the channel information of the first bandwidth unit among the multiple bandwidth units are inaccurate. The processing unit is used to obtain the recovery channel information of the plurality of bandwidth units based on the first information.

[0008] Fifthly, embodiments of this application may also provide a communication device, including: Processor, memory, and interfaces for communicating with network devices; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory, causing the processor to perform a communication method as provided in either the first or second aspect.

[0009] In a sixth aspect, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the communication method as described in either the first or second aspect.

[0010] In a seventh aspect, embodiments of this application provide a program that, when executed by a processor, performs the communication method as described in either the first or second aspect above.

[0011] Alternatively, the processor described above can be a chip.

[0012] Eighthly, embodiments of this application provide a computer program product, including program instructions for implementing the communication method described in either the first or second aspect.

[0013] Ninthly, embodiments of this application provide a chip, including: a processing module and a communication interface, wherein the processing module is capable of executing the communication method described in either the first or second aspect.

[0014] Furthermore, the chip also includes a storage module (e.g., a memory) for storing instructions, a processing module for executing the instructions stored in the storage module, and the execution of the instructions stored in the storage module causes the processing module to perform the communication method described in either the first or second aspect. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of a communication system applicable to embodiments of this application; Figure 2 A schematic structural diagram of a neuron provided in an embodiment of this application; Figure 3 A schematic diagram of a neural network provided in an embodiment of this application; Figure 4 Another schematic diagram of the neural network provided in the embodiments of this application; Figure 5 A schematic diagram of the LSTM cell structure provided in the embodiments of this application; Figure 6 A schematic diagram of an autoencoder neural network provided in an embodiment of this application; Figure 7 A schematic diagram of the channel information feedback system provided in the embodiments of this application; Figure 8 A schematic flowchart illustrating the communication method provided in the embodiments of this application; Figure 9 A schematic diagram of the verification model provided in the embodiments of this application; Figure 10 Another schematic diagram of the communication method provided in the embodiments of this application; Figure 11 Another schematic diagram of the communication method provided in the embodiments of this application; Figure 12 Another schematic diagram of the communication method provided in the embodiments of this application; Figure 13 A schematic block diagram of the communication device provided in this application; Figure 14 A schematic structural diagram of the communication device provided in this application. Detailed Implementation

[0016] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0017] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0018] The technical solutions of this application can be applied to various communication systems, such as: Long Term Evolution (LTE) systems, LTE Frequency Division Duplex (FDD) systems, LTE Time Division Duplex (TDD) systems, Universal Mobile Telecommunications System (UMTS), Worldwide Interoperability for Microwave Access (WiMAX) systems, 5th Generation (5G) systems, new radio (NR) systems, and future communication systems, such as 6th Generation (6G) systems. This application does not limit these applications.

[0019] Figure 1 This is a schematic structural diagram of a communication system applicable to this application.

[0020] To facilitate understanding of the embodiments of this application, firstly, in conjunction with Figure 1 This application describes in detail a communication system applicable to the communication method provided in the embodiments of this application. As shown in the figure, the communication system 100 may include at least one network device, such as... Figure 1 The network device 101 in the communication system 100 may also include at least one terminal device, such as Figure 1The network device 101 and terminal devices 102 to 107 are described. These terminal devices 102 to 107 can be mobile or fixed. One or more of these devices can communicate via a wireless link. The network device and the terminal devices can communicate using the communication methods provided in this embodiment. Optionally, the terminal devices can communicate directly. For example, device-to-device (D2D) technology can be used to achieve direct communication between terminal devices. As shown in the figure, terminal devices 105 and 106, and terminal devices 105 and 107, can communicate directly using D2D technology. Terminal devices 106 and 107 can communicate with terminal device 105 individually or simultaneously. The communication methods provided in this embodiment can be used when the terminal devices communicate with each other.

[0021] It should be understood that Figure 1 Different communication devices for wireless communication systems are illustrated exemplarily. However, this application is not limited thereto, and the communication methods provided in the embodiments of this application can also be applied to wired communication systems.

[0022] The terminal device in this application embodiment can be referred to as a terminal, user equipment (UE), or an access terminal, user unit, user station, mobile station, mobile station, remote station, remote terminal, mobile device, user terminal, terminal equipment, wireless communication equipment, user agent, or user device. The terminal device can also be a cellular phone, cordless phone, session initiation protocol (SIP) phone, wireless local loop (WLL) station, personal digital assistant (PDA), handheld device with wireless communication capabilities, computing device, or other processing device connected to a wireless modem, vehicle-mounted device, wearable device, terminal device in a 5G network, or terminal device in a future evolved public land mobile network (PLMN), etc. This application embodiment does not limit the scope of these claims.

[0023] In this embodiment, the terminal device can also be a wearable device. Wearable devices, also known as wearable smart devices, are a general term for devices that utilize wearable technology to intelligently design and develop everyday wearables, such as glasses, gloves, watches, clothing, and shoes. Wearable devices are portable devices that are worn directly on the body or integrated into the user's clothing or accessories. Wearable devices are not merely hardware devices, but also achieve powerful functions through software support, data interaction, and cloud interaction. Broadly speaking, wearable smart devices include those that are feature-rich, large in size, and can achieve complete or partial functions without relying on a smartphone, such as smartwatches or smart glasses, as well as those that focus on a specific type of application function and require the use of other devices such as smartphones, such as various smart bracelets and smart jewelry for vital sign monitoring. Furthermore, in this embodiment, the terminal device can also be a terminal device in a vehicle networking system or an Internet of Things (IoT) system.

[0024] The network device in this application embodiment can be a device for communicating with terminal devices. The network device can be a base station (BTS) in a GSM or CDMA system, a base station (nodeB, NB) in a WCDMA system, an evolved NodeB (eNB or eNodeB) in an LTE system, a radio controller in a cloud radio access network (CRAN) scenario, or a relay station, access point, vehicle-mounted equipment, and network equipment in a 5G network or a network equipment in a future evolved PLMN network, etc. The embodiments of this application are not limited.

[0025] The relevant technologies and terms involved in the embodiments of this application are described below.

[0026] I. Feature Vector Feedback Scheme Based on Codebook In current communication systems, downlink channel state information (CSI) feedback is typically provided by the terminal equipment using a codebook-based feature vector feedback method, enabling the base station to acquire the downlink CSI. Specifically, the base station sends a downlink CSI-reference signal (CSI-RS) to the terminal equipment. The terminal equipment uses the CSI-RS to estimate the downlink channel's CSI and performs eigenvalue decomposition on the estimated downlink channel to obtain the corresponding feature vector. In NR systems, two codebook design schemes, Type 1 and Type 2, are provided. Type 1 codebooks are used for CSI feedback of normal precision and for single-user-multiple-input multiple-output (SU-MIMO) and multiple-user-MIMO (MU-MIMO) transmissions. Type 2 codebooks are used to improve the transmission performance of MU-MIMO. Both Type 1 and Type 2 codebooks employ... The two-level codebook feedback, in which Used to characterize the bandwidth and long-term characteristics of the channel, and to determine a set of L DFT beams. Used to characterize the subband and short-term characteristics of the channel. Specifically, for Type 1 codebooks, Its function is to select one beam from L DFT beams; for Type 2 codebooks, Its function is to The L DFT beams in the codebook are linearly combined and fed back in terms of amplitude and phase. Generally, the Type 2 codebook utilizes a higher number of feedback bits to achieve higher precision CSI feedback performance.

[0027] II. Neural Networks and Deep Learning A neural network is a computational model consisting of multiple interconnected neurons, where the connections between nodes represent the input signal. Weighted value of the output signal (referred to as weights), where, Each node performs a weighted summation of different input signals and outputs the result through a specific activation function. The neuron structure is as follows: Figure 2 As shown, f represents a non-linear activation function that performs non-linear activation on the weighted input, such as softmax, ReLU, sigmoid, etc. t is the output of the neuron. A simple neural network is shown below. Figure 3As shown, it includes an input layer, a hidden layer, and an output layer. Through different connection methods, weights, and activation functions of multiple neurons, different outputs can be generated, thereby fitting the mapping relationship from input to output.

[0028] Deep learning employs deep neural networks with multiple hidden layers, greatly enhancing the network's ability to learn features and fit complex nonlinear mappings from input to output. Therefore, it has found widespread application in speech and image processing. Besides deep neural networks, deep learning also includes commonly used basic structures such as convolutional neural networks (CNNs) and recurrent neural networks (RNNs) for different tasks.

[0029] The basic structure of a convolutional neural network includes: an input layer, multiple convolutional layers, multiple pooling layers, a fully connected layer, and an output layer, such as... Figure 4 As shown, each neuron in the convolutional kernel of the convolutional layer is locally connected to its input, and by introducing pooling layers to extract the local maximum or average features of a certain layer, the parameters of the network are effectively reduced, and local features are mined, enabling the convolutional neural network to converge quickly and achieve excellent performance.

[0030] RNNs (Recurrent Neural Networks) are neural networks that model sequential data and have achieved remarkable success in natural language processing applications such as machine translation and speech recognition. Specifically, the network memorizes information from past time steps and uses it in the calculation of the current output; that is, nodes in the hidden layers are no longer disconnected but connected, and the input to the hidden layers includes not only the input layer but also the output of the hidden layer from the previous time step. Commonly used RNN structures include Long Short-Term Memory (LSTM) artificial neural networks and gated recurrent units (GRUs). Figure 5 The diagram shows a basic LSTM unit structure, where f represents the hyperbolic tangent (tanh) activation function. It can refer to various non-linear activation functions. Unlike RNNs, which only consider the most recent state, LSTMs determine which states should be retained and which should be forgotten based on the cell state, thus solving the shortcomings of traditional RNNs in long-term memory.

[0031] III. Channel Information Feedback Method Based on Deep Learning Given the tremendous success of artificial intelligence (AI) technology, especially deep learning, in areas such as computer vision and natural language processing, the communications field has begun to explore its application to solve technical challenges that traditional communication methods struggle with. For example, the neural network architecture commonly used in deep learning is non-linear and data-driven, enabling feature extraction from actual channel matrix data and, at the base station, reconstructing the compressed channel matrix information from the UE (User Equipment) feedback as accurately as possible. This not only ensures accurate channel information reconstruction but also reduces CSI (Channel Signal Inquiry) feedback overhead at the UE. Deep learning-based CSI feedback treats channel information as an image to be compressed, using a deep learning autoencoder to compress and feed back the channel information. The compressed channel image is then reconstructed at the transmitting end, preserving channel information to a greater extent. For example… Figure 6 As shown, the autoencoder includes a neural network (NN) encoder at the transmitting end and an NN decoder at the receiving end. The NN encoder and decoder are obtained through joint learning (or joint training), as follows: Figure 6 As shown, an image with a pixel size of 28×28 (i.e., 784) is processed by an NN encoder to obtain compressed feedback information of the input target (i.e., the image). The size of this compressed feedback information is usually less than 784. After the transmitter sends this compressed feedback information to the receiver, the receiver inputs it into an NN decoder, which is able to reconstruct the original target (i.e., the image).

[0032] Figure 7 This is a schematic diagram of a channel information feedback system applicable to this application, provided as an embodiment of the present application. The entire feedback system is divided into an encoder and a decoder, deployed at the transmitting end and the receiving end, respectively. After obtaining channel information through channel estimation, the transmitting end compresses and encodes the channel information matrix through the neural network of the encoder, and feeds back the compressed bit stream to the receiving end through the air interface feedback link. The receiving end recovers the channel information based on the feedback bit stream using the decoder to obtain complete feedback channel information. Figure 7 The data receiver (Rx) in the diagram feeds back CSI to the data transmitter (Tx). The internal structure of Rx's CSI encoder, as shown within the encoder's dashed box, employs a stack of multiple fully connected layers. After the channel vector is input as an image, it outputs an M×1 codeword and sends it to Tx. The internal structure of Tx's CSI decoder, as shown within the decoder's dashed box, utilizes a design with convolutional layers and residual structures. Specifically, the channel vector input to the CSI encoder from Rx undergoes batch normalization after passing through a convolutional layer (where the alpha of the hysteresis-corrected linear unit (ReLU) is... After processing (e.g., 0.3) and reconstruction, an M×1 codeword is obtained. The CSI decoder of Tx undergoes linear activation processing by a fully connected (Dense) layer, followed by reconstruction and batch normalization through multiple convolutional layers before outputting to the next refineNet network. Finally, the recovered channel vector is output after processing by a sigmoid function (e.g., sigmoid). However, this application is not limited to this. The network model structure within the encoder and decoder can be flexibly designed while maintaining the encoding and decoding framework.

[0033] To address the inaccuracy issues in current channel information feedback methods (such as CSI feedback), this application proposes that the encoded channel information be verified at the channel information feedback end (e.g., the terminal) to determine whether the feedback information receiving end (e.g., the base station) can accurately recover the channel information. If the verification result indicates that the channel information cannot be accurately recovered, the feedback end can notify the receiving end to prevent the receiving end from transmitting data based on inaccurate channel information, thus avoiding communication failures and resource waste. This aims to improve communication reliability.

[0034] The communication method provided in this application will be described below with reference to the accompanying drawings.

[0035] Figure 8 This is a schematic flowchart of a communication method provided in an embodiment of this application.

[0036] S801, the second device acquires the coded feedback sequence corresponding to the channel information of multiple bandwidth units.

[0037] For example, the frequency domain bandwidth in which the first device communicates with the second device includes multiple bandwidth units (or a bandwidth unit can be called a sub-band of the frequency domain bandwidth). The second device can obtain the channel information of each of the multiple bandwidth units and feed it back to the first device so that the first device can process the data to be sent to the second device based on the channel information of the multiple bandwidth units, such as channel coding processing, precoding processing, etc., but this application is not limited to this.

[0038] Optionally, the first device may send a reference signal for channel information measurement to the second device, and correspondingly, the second device may receive the reference signal from the first device. The second device may obtain channel information between the first device and the second device based on the reference signal; specifically, the second device may obtain channel information regarding the direction in which the first device sends a signal to the second device based on the reference signal.

[0039] For example, the first device can be a network device, and the second device can be a terminal device. The network device sends a reference signal to the terminal device, and the terminal device receives the reference signal from the network device and can obtain the downlink channel information from the network device to the terminal device based on the reference signal.

[0040] By way of example and without limitation, the reference signal can be CSI-RS or a synchronization signal (SS).

[0041] The reference signal can be transmitted in multiple bandwidth units, and the second device can obtain the channel information of these multiple bandwidth units based on the reference signal.

[0042] The second device can encode the channel information of the multiple bandwidth units to obtain an encoded feedback sequence.

[0043] For example, the second device compresses and encodes the channel information of the multiple bandwidth units to obtain a compressed and encoded feedback sequence. The second device can reduce the overhead of feedback information by performing compression and encoding on the channel information and sending the encoded feedback sequence to the first device.

[0044] Optionally, the second device can input channel information of multiple bandwidth units into the coding model to obtain the coding feedback sequence output by the coding model.

[0045] In one embodiment, the coding model is used to encode the channel information of each of the multiple input bandwidth units. The second device can input the channel information of each of the multiple bandwidth units into the coding model to obtain the sub-coding sequence corresponding to the channel information of each bandwidth unit output by the coding model. That is, the coding feedback sequence includes the sub-coding sequence corresponding to the channel information of each of the multiple bandwidth units.

[0046] In other words, the second device independently encodes the channel information of each bandwidth unit using an encoding model, thereby obtaining a sub-encoded sequence corresponding to the channel information of each bandwidth unit that can be independently decoded.

[0047] For example, if the second device needs to feed back channel information for K bandwidth units, it inputs the channel information of bandwidth unit 1 into the coding model to obtain the sub-coded sequence corresponding to the channel information of bandwidth unit 1 output by the coding model. Then, it inputs the channel information of bandwidth unit 2 into the coding model to obtain the sub-coded sequence corresponding to the channel information of bandwidth unit 2 output by the coding model. The same method can be used to obtain the sub-coded sequence corresponding to the channel information of each of the K bandwidth units. The second device can then arrange these K sub-coded sequences in the order output by the coding model to obtain the coded feedback sequence.

[0048] In another embodiment, the coding model is used to jointly code the channel information of multiple input bandwidth units. The second device can input the channel information of multiple bandwidth units into the coding model to obtain the joint coding sequence output by the coding model. That is, the coding feedback sequence is the joint coding sequence corresponding to the channel information of the multiple bandwidth units.

[0049] The coding model can jointly encode channel information from multiple bandwidth units based on the frequency correlation of those units, resulting in a joint coded sequence. This improves the feedback performance of channel information feedback.

[0050] For example, if the second device needs to feed back channel information for K bandwidth units, the second device will input the channel information in the K bandwidth units into the coding model in sequence to obtain the joint coding sequence corresponding to the channel information of the K bandwidth units output by the coding model.

[0051] This encoding model can be an AI model. As an example and without limitation, this encoding model is the encoder model in an autoencoder neural network model.

[0052] S802, the second device determines that the encoding and decoding result of the channel information of the first bandwidth unit among the multiple bandwidth units is inaccurate.

[0053] The second device can decode and verify the acquired coded feedback sequence to obtain verification channel information for multiple bandwidth units. It then compares the differences between the verification channel information of these multiple bandwidth units and the channel information of the multiple bandwidth units to determine whether the channel information encoding and decoding for each bandwidth unit is accurate. The accuracy of the channel information encoding and decoding reflects the performance of the channel information feedback.

[0054] Optionally, the second device can input the coded feedback sequence into the verification model to obtain the verification channel information of multiple bandwidth units output by the verification model.

[0055] The verification model may include a decoding model corresponding to the encoding model of the second device. As an example and not a limitation, the encoding model is the encoder model in the autoencoder neural network model, and the verification model is the decoder model in the autoencoder neural network model.

[0056] In one embodiment, a first device and a second device jointly train an online encoding model and a decoding model for channel information transmission. The second device trains the encoding model to encode the channel information and obtain an encoded feedback sequence. The first device trains the decoding model to decode the channel information received from the second device and obtain decoded channel information. The first and second devices complete the online joint training, obtaining the trained parameter information for the decoding and encoding models, respectively. The first device can send third information to the second device, which indicates the configuration parameter information (i.e., the trained parameter information) of the first device's decoding model. Correspondingly, the second device receives the third information from the first device and configures a verification model according to the configuration parameter information indicated by the third information.

[0057] In another implementation, the encoding and decoding models can be trained offline. The first and second devices can exchange configuration parameter information for both the encoding and decoding models.

[0058] In one example, the first device pre-stores configuration parameter information for the encoding model and the decoding model. After the first device establishes a communication connection with the second device, the second device can obtain the configuration parameter information for the encoding model and the decoding model from the first device. The second device can configure the encoding model according to the configuration parameter information for encoding channel information to obtain the encoded feedback sequence, and configure the verification model according to the configuration parameter information for verifying the obtained encoded feedback sequence to obtain the verification channel information.

[0059] For example, the first device is a network device and the second device is a terminal device. After the terminal device accesses the network and establishes a communication connection with the network device, it can download the configuration parameter information of the encoding model and the configuration parameter information of the decoding model from the network, and use them to configure the encoding model and the verification model respectively.

[0060] In another example, the second device pre-stores configuration parameter information for the encoding and decoding models. After the first device establishes a communication connection with the second device, the first device can obtain the configuration parameter information of the decoding model from the second device. The first device can configure the decoding model according to the configuration parameter information of the decoding model, and use it to decode and obtain the decoded channel information based on the channel feedback information received from the second device.

[0061] The second device can compare the differences between the channel information of each bandwidth unit and the verification channel information in multiple bandwidth units to determine whether the channel information encoding and decoding results of each bandwidth unit are accurate.

[0062] Optionally, the second device can determine the difference between the channel information of the bandwidth unit and the verification channel information based on the normalized mean square error (NMSE) or cosine similarity (CS).

[0063] For example, the second device can periodically feed back channel information. In the nth feedback period, the channel information of bandwidth unit k can be expressed as... The second device encodes and decodes the channel information to obtain the verification channel information for the bandwidth unit k, which can be represented as follows: The second device can be determined using the CS method (or CS index). and The difference between them, namely and Cosine similarity between Can be used for characterization and The difference between them, the cosine similarity Satisfy the following formula:

[0064] The second device obtains the cosine similarity. The cosine similarity Compare with the verification threshold r, if This indicates that the channel information encoding and decoding results for bandwidth unit k differ significantly, and the encoding and decoding results are inaccurate.

[0065] The second device can compare the channel information of each bandwidth unit with the check channel information in multiple bandwidth units, and determine that the channel information encoding and decoding result of the first bandwidth unit is inaccurate based on the difference between the check channel information of the first bandwidth unit and the channel information of the first bandwidth unit.

[0066] For example, if the second device calculates the CS between the verification channel information of the first bandwidth and the channel information of the first bandwidth unit, and finds that the CS is less than the verification threshold r, it indicates that the similarity between the verification channel information of the first bandwidth and the channel information of the first bandwidth unit is low and the difference is large. Therefore, the second device determines that the encoding and decoding result of the channel information of the first bandwidth unit is inaccurate.

[0067] Optionally, the second device inputs the encoded feedback sequence and channel information of multiple bandwidth units into the verification model to obtain second information. This second information is used to indicate that the encoding and decoding result of the channel information of the first bandwidth unit is inaccurate.

[0068] This second piece of information may be called verification instruction information, but this application is not limited thereto.

[0069] For example, the validation model can be like Figure 9 As shown, the verification model may include a decoding model and a verification module. The second device inputs channel information of multiple bandwidth units into the verification model, specifically into the verification module of the verification model. The second device also inputs the encoded feedback sequence corresponding to the channel information of multiple bandwidth units output by the encoding model into the decoding model within the verification model. The decoding model decodes the encoded feedback sequence to obtain the verification channel information of multiple bandwidth units. The verification module in the verification model can compare the differences between the channel information of each bandwidth unit and the verification channel information based on the input channel information of the multiple bandwidth units and the verification channel information of those multiple bandwidth units, determine whether the encoding and decoding result of the channel information of each bandwidth unit is accurate, and output verification indication information. For example, if the verification module determines that the encoding and decoding result of the channel information of the first bandwidth unit is inaccurate, then the verification indication information is used to indicate that the encoding and decoding result of the channel information of the first bandwidth unit is inaccurate.

[0070] Optionally, the configuration parameter information of the decoding model in the verification model can be obtained from third information acquired from the first device. Specific implementation methods can be found in the preceding description, and will not be repeated here for brevity.

[0071] Optionally, the verification module can determine the difference between the channel information of the bandwidth unit and the verification channel information based on the normalized mean square error (NMSE) or cosine similarity (CS).

[0072] S803, the second device sends first information to the first device. The first information includes channel feedback information and second information. The channel feedback information includes the channel feedback sequence corresponding to the channel information of the plurality of bandwidth units. The second information is used to indicate that the channel information encoding and decoding result of the first bandwidth unit among the plurality of bandwidth units is inaccurate.

[0073] Accordingly, the first device receives the first information from the second device.

[0074] The following section first introduces the indication method for inaccurate channel information encoding and decoding results of the first bandwidth unit, including but not limited to one of the following indication methods.

[0075] In the first indication method, the second information may include the identification information of the first bandwidth unit.

[0076] For example, the frequency domain bandwidth of the second device includes K bandwidth units, which are identified as 1, 2, 3, ... K. The identification information of the first bandwidth unit is k. Therefore, the second information includes the identification information of the first bandwidth unit, i.e., k. If the second device determines in S802 that the channel encoding and decoding results of multiple bandwidth units are inaccurate, then the second information includes the identification information of the multiple bandwidth units whose channel encoding and decoding results are inaccurate. For instance, if the second device determines in S802 that the channel encoding and decoding results of two bandwidth units with identification information 1 and k are inaccurate, then the second information includes the identification information 1 and k of those two bandwidth units.

[0077] In the second indication method, the second information includes a bitmap, which includes multiple bits that correspond to multiple bandwidth units within the frequency domain bandwidth of the second device. One of the multiple bits is used to indicate whether the encoding and decoding result of the channel information of the corresponding bandwidth unit is accurate.

[0078] For example, if the frequency domain bandwidth of the second device includes 5 bandwidth units, then the bitmap includes 5 bits. If the second device determines that the encoding / decoding result of the channel information of the first bandwidth unit is inaccurate, while the encoding / decoding results of the channel information of other bandwidth units are accurate, then the first bit of the bitmap, corresponding to the first bandwidth unit, can indicate "1", indicating that the encoding / decoding result of the channel information of the first bandwidth unit is inaccurate, and the other bits of the bitmap can indicate "0", indicating that the encoding / decoding result of the channel information of the corresponding bandwidth unit is accurate. In this case, the bitmap is "10000". As another example, if the second device determines that the encoding / decoding results of the channel information of multiple bandwidth units out of the 5 bandwidth units are inaccurate, then the bits in the bitmap corresponding to the multiple bandwidth units with inaccurate channel information encoding / decoding results are all indicated as "1". However, this application is not limited to this. Bits in the bitmap can indicate "1" to indicate that the encoding / decoding result of the channel information of the corresponding bandwidth unit is accurate; bits indicating "0" indicate that the encoding / decoding result of the channel information of the corresponding bandwidth unit is inaccurate.

[0079] Instruction Method 3: If the frequency domain bandwidth of the second device includes K bandwidth units, then the second information may include... 1 bit, the One bit can indicate K values, and these K values ​​correspond to K bandwidth units. When this... When a bit indicates one of the K values, it means that the channel information encoding and decoding result of the bandwidth unit corresponding to that value is inaccurate.

[0080] For example, if the second device includes 5 bandwidth units, then the second information may include 3 bits, with each of the 3 bits taking a value from "000" to "101" to correspond to one of the 5 bandwidth units. For example, the first bandwidth unit may correspond to "000". If the second device determines that the encoding and decoding result of the channel information of the first bandwidth unit is inaccurate, then the 3 bits in the second information indicate "000".

[0081] Since the possibility of severe feedback performance loss occurring simultaneously in two bandwidth units is low in channel information feedback methods based on AI-based channel encoding and decoding (such as the encoder model and decoder model in an autoencoder neural network, respectively), this indication method three can be adopted when providing channel information feedback based on AI-based channel encoding and decoding, which can reduce the signaling overhead of the second information.

[0082] The following describes a specific implementation method for the second device to send channel feedback information to the first device, including but not limited to one of the following implementation methods.

[0083] In one embodiment, the channel feedback sequence corresponding to the channel information of the plurality of bandwidth units includes the coded feedback sequence corresponding to the channel information of the plurality of bandwidth units.

[0084] In other words, the channel feedback sequence includes the encoded feedback sequence obtained by the second device encoding the channel information of multiple bandwidth units.

[0085] Optionally, the coding feedback sequence may be a joint coding sequence corresponding to the channel information of the multiple bandwidth units output by the coding model, or the coding feedback sequence may include the coding sequence corresponding to the channel information of each bandwidth unit obtained by the coding model encoding the channel information of the multiple bandwidth units respectively.

[0086] For example Figure 10 As shown, the second device can be a terminal device, and the first device can be a network device. After the terminal device measures the channel information of multiple bandwidth units based on the reference signal, it inputs the information into the encoder model to obtain the coding feedback sequence output by the coding model. The terminal device inputs the channel information of the multiple bandwidth units and the coding feedback sequence output by the encoder model into the verification model to obtain the verification indication information (i.e., the second information) output by the verification model. This verification indication information indicates that the encoding and decoding results of the channel information of the first bandwidth unit are inaccurate. The terminal device sends the channel feedback information including the coding feedback sequence as CSI to the network device, and also sends the verification indication information to the network device.

[0087] It should be noted that the terminal device can also send both channel feedback information and verification indication information as CSI to the network device. In other words, the terminal device can send the channel feedback information and verification indication information to the network device in the same message or in separate messages. This application does not limit this.

[0088] After receiving the first information, the network device can decode the encoded feedback sequence in the channel feedback information to obtain the decoded channel information for multiple bandwidth units. Based on the second information, the network device can determine that the encoding and decoding result of the decoded channel information corresponding to the first bandwidth unit is inaccurate. The network device can avoid using the inaccurate channel information corresponding to the first bandwidth unit for data transmission in that first bandwidth unit, thereby reducing the probability of data transmission failure.

[0089] Optionally, the channel feedback information includes the codebook sequence corresponding to the channel information of the first bandwidth unit.

[0090] The second device may pre-store a codebook set corresponding to the channel information, which includes multiple codebook sequences. The second device can determine the codebook sequence corresponding to the channel information of the first bandwidth unit from the codebook set based on the channel information of the first bandwidth unit. The second device sends channel feedback information and second information to the first device. The channel feedback information includes the coded feedback sequence and the codebook sequence corresponding to the channel information of the first bandwidth unit.

[0091] According to this scheme, for bandwidth units with poor feedback performance when using AI models (such as encoding and decoding models) for channel feedback, a codebook-based feedback method is adopted to perform codebook sequence feedback, achieving hybrid CSI feedback of AI model encoding feedback and codebook feedback. This allows the first device to obtain the codebook sequence of the bandwidth unit with poor feedback performance, enabling it to process the data to be transmitted on the first bandwidth unit based on the codebook sequence. This improves resource utilization and communication reliability.

[0092] For example Figure 11 As shown, after the terminal device determines that the channel information encoding and decoding result of the first bandwidth unit is inaccurate through the verification model, the terminal device can select the corresponding codebook sequence in the codebook set according to the channel information of the first bandwidth unit, and send the channel feedback sequence, which includes the encoding feedback sequence and the codebook sequence corresponding to the channel information of the first bandwidth unit, as well as the verification indication information (i.e., the second information) to the network device.

[0093] After receiving the channel feedback sequence, the network device can input it into the decoder model and output recovered channel information. This recovered channel information can include the decoded recovered channel information of multiple bandwidth units obtained by the decoder model from decoding the encoded feedback sequence. This includes the decoded recovered channel information of the first bandwidth unit, and the codebook recovered channel information obtained by the decoder model from decoding the codebook sequence corresponding to the channel information of the first bandwidth unit. The network device can determine, based on the verification indication information, that the encoding and decoding result of the channel information of the first bandwidth unit is inaccurate, and therefore the decoded recovered channel information of the first bandwidth unit is inaccurate. The network device can then process the data to be transmitted on the first bandwidth unit based on this codebook recovered channel information (e.g., precoding).

[0094] In another embodiment, the second device encodes multiple bandwidth units to obtain a sub-coding sequence corresponding to the channel information of each bandwidth unit (i.e., the channel information of each bandwidth unit is independently encoded to obtain a sub-coding sequence that can be independently decoded). The channel feedback sequence may include the sub-coding sequence corresponding to the channel information of the second bandwidth unit, which is a bandwidth unit other than the first bandwidth unit among the multiple bandwidth units.

[0095] In other words, after the second device determines that the encoding and decoding result of the channel information of the first bandwidth unit is inaccurate, the second device does not feed back the sub-coded sequence corresponding to the channel information of the first bandwidth unit to the first device. The channel feedback information includes the sub-coded sequences corresponding to the channel information of multiple bandwidth units other than the first bandwidth unit.

[0096] Optionally, the second device encodes multiple bandwidth units to obtain a sub-coding sequence corresponding to the channel information of each bandwidth unit. The channel feedback sequence includes the sub-coding sequence corresponding to the second bandwidth unit and the codebook sequence corresponding to the first bandwidth unit.

[0097] For example, the second device encodes multiple bandwidth units to obtain a sub-coded sequence corresponding to the channel information of each bandwidth unit. The encoded feedback sequence composed of these sub-coded sequences corresponding to the channel information of multiple bandwidth units can be written as... If the first device determines, through verification, that the encoding and decoding result of bandwidth unit k (i.e., an example of the first bandwidth unit) is inaccurate, then the first device obtains the codebook sequence of the first bandwidth unit. The sub-coded sequence corresponding to the channel information of the first bandwidth unit in the encoded feedback sequence. Replace with The channel feedback sequence is obtained. The channel feedback sequence and the second information are then sent to the first device.

[0098] For example Figure 12As shown, the second device is a terminal device, and the first device is a network device. After determining that the channel information encoding and decoding result of the first bandwidth unit is inaccurate, the terminal device selects the corresponding codebook sequence from the codebook set based on the channel information of the first bandwidth unit, and replaces the sub-encoding sequence corresponding to the channel information of the first bandwidth unit in the encoding feedback sequence with the codebook sequence to obtain the channel feedback sequence. The terminal device sends the channel feedback information and the second information, including the channel feedback sequence, to the network device. After receiving the channel feedback information and the second information, the network device inputs the channel feedback sequence into the decoder model to obtain the recovered channel information output by the decoder model. Based on the second information, the network device can determine that the channel information encoding and decoding result of the first bandwidth unit is inaccurate, and thus can determine that the recovered channel information includes the decoded recovered channel information of multiple bandwidth units other than the first bandwidth unit, as well as the codebook recovered channel information of the first bandwidth unit.

[0099] According to the above scheme, after the second device determines that the channel information encoding and decoding result of the first bandwidth unit is inaccurate, it replaces the sub-coding sequence corresponding to the channel information of the first bandwidth unit with the corresponding codebook sequence. This avoids sending inaccurate feedback sequences and reduces transmission overhead. Furthermore, the first device can also obtain the codebook sequence corresponding to the channel information of the first bandwidth unit and process the data to be transmitted on the first bandwidth unit based on the codebook sequence. This reduces the probability of data transmission failure on the first bandwidth unit, improves resource utilization, and enhances communication reliability.

[0100] The above, combined with Figures 8 to 12 The methods provided in the embodiments of this application are described in detail. The communication apparatus and communication device provided in the embodiments of this application are described below.

[0101] Figure 13 This is a schematic block diagram of a communication device provided in an embodiment of this application. Figure 13 As shown, the communication device 1300 may include a processing unit 1310 and a transceiver unit 1320.

[0102] In one possible design, the communication device 1300 may correspond to the communication device (e.g., the first device or the second device) in the above method embodiments, or be a chip configured in (or used for) the first device or the second device.

[0103] It should be understood that the communication device 1300 may correspond to the first device or the second device in the method 800 according to the embodiments of this application, and the communication device 1300 may include tools for performing... Figure 8 The method 800 is a unit that executes the method by the first device or the second device. Furthermore, each unit in the communication device 1300 and the other operations and / or functions described above are respectively for implementing... Figure 8 The corresponding process of method 800 in the middle.

[0104] It should also be understood that when the communication device 1300 is a chip configured in (or used in) the first device or the second device, the transceiver unit 1320 in the communication device 1300 can be the input / output interface or circuit of the chip, and the processing unit 1310 in the communication device 1300 can be the processor in the chip.

[0105] Optionally, the processing unit 1310 of the communication device 1300 can be used to process instructions or data to implement corresponding operations.

[0106] Optionally, the communication device 1300 may further include a storage unit 1330, which can be used to store instructions or data. The processing unit 1310 can execute the instructions or data stored in the storage unit to enable the communication device to perform corresponding operations. The transceiver unit 1320 in the communication device 1300 is a transceiver unit that can correspond to... Figure 14 The transceiver 1410 and storage unit 1430 in the communication device 1400 shown can correspond to Figure 14 The memory 1430 in the communication device 1400 shown in the figure.

[0107] Optionally, the transceiver unit 1320 in the communication device 1300 can be implemented via a communication interface (such as a transceiver or input / output interface). And / or, the processing unit 1310 in the communication device 1300 can be implemented via at least one logic circuit.

[0108] It should be understood that the specific process of each unit performing the above-mentioned corresponding steps has been described in detail in the above method embodiments, and will not be repeated here for the sake of brevity.

[0109] Figure 14 This is a schematic diagram of the structure of the communication device 1400 provided in an embodiment of this application. The communication device 1400 can be applied to, for example... Figure 1 In the system shown, the functions of the first or second device in the above method embodiments are executed. As shown, the communication device 1400 includes a processor 1420 and a transceiver 1410. Optionally, the communication device 1400 also includes a memory 1430. The processor 1420, transceiver 1410, and memory 1430 can communicate with each other through internal connection paths to transmit control and / or data signals. The memory 1430 is used to store computer programs, and the processor 1420 is used to execute the computer programs in the memory 1430 to control the transceiver 1410 to transmit and receive signals.

[0110] The processor 1420 and memory 1430 can be combined into a single processing device. The processor 1420 executes the program code stored in the memory 1430 to achieve the aforementioned functions. In specific implementations, the memory 1430 can be integrated into the processor 1420 or independent of it. The processor 1420 can be combined with... Figure 13 The corresponding processing unit in the process.

[0111] The transceiver 1410 described above can be used with Figure 13 The transceiver unit 1320 corresponds to this. The transceiver 1410 may include a receiver (or receiver circuit) and a transmitter (or transmitter circuit). The receiver is used to receive signals, and the transmitter is used to transmit signals.

[0112] It should be understood that Figure 14 The communication device 1400 shown can achieve Figure 8 The various processes involved in the method 800 embodiments of the above method 1400 pertain to the first or second device. The operation and / or function of each module in the communication device 1400 are respectively for implementing the corresponding processes in the above method embodiments. For details, please refer to the description in the above method embodiments; to avoid repetition, detailed descriptions are appropriately omitted here.

[0113] The processor 1420 described above can be used to perform the actions implemented internally by the first or second device as described in the preceding method embodiments, while the transceiver 1410 can be used to perform the actions described in the preceding method embodiments of sending to or receiving from other communication devices. For details, please refer to the descriptions in the preceding method embodiments; they will not be repeated here.

[0114] Optionally, the communication device 1400 may also include a power supply for providing power to various devices or circuits in the terminal device.

[0115] Optionally, both the first device and the second device can be terminal devices, or the second device can be a terminal device and the first device can be a network device, or the first device can be a terminal device and the second device can be a network device.

[0116] This application also provides a processing apparatus, including a processor and an interface; the processor is used to execute the method in any of the above method embodiments.

[0117] It should be understood that the aforementioned processing device can be one or more chips. For example, the processing device can be a field-programmable gate array (FPGA), an application-specific integrated circuit (ASIC), a system-on-chip (SoC), a central processor unit (CPU), a network processor (NP), a digital signal processor (DSP), a microcontroller unit (MCU), a programmable logic device (PLD), or other integrated chips.

[0118] In implementation, each step of the above method can be completed by integrated logic circuits in the processor's hardware or by instructions in software. The steps of the method disclosed in the embodiments of this application can be directly implemented by a hardware processor, or by a combination of hardware and software modules in the processor. The software modules can reside in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. This storage medium is located in memory, and the processor reads information from the memory and, in conjunction with its hardware, completes the steps of the above method. To avoid repetition, detailed descriptions are omitted here.

[0119] It should be noted that the processor in the embodiments of this application can be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above method embodiments can be completed by the integrated logic circuitry in the processor's hardware or by instructions in software form. The processor can be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this application can be directly embodied as execution by a hardware decoding processor, or as a combination of hardware and software modules in the decoding processor. The software modules can be located in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. This storage medium is located in memory, and the processor reads the information in the memory and, in conjunction with its hardware, completes the steps of the above methods.

[0120] The present application provides a method in the embodiments of the present application. The present application also provides a computer program product, which includes computer program code, which, when executed by one or more processors, causes a device including the processor to perform the method in the above embodiments.

[0121] According to the method provided in the embodiments of this application, this application also provides a computer-readable storage medium storing program code that, when run by one or more processors, causes a device including the processor to perform the method in the above embodiments.

[0122] According to the method provided in the embodiments of this application, this application also provides a system that includes one or more of the aforementioned second devices. The system may further include one or more of the aforementioned first devices.

[0123] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between modules may be electrical, mechanical, or other forms.

[0124] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A communication method characterized by comprising: The method includes: Send first information to the first device. The first information includes channel feedback information and second information. The channel feedback information includes a channel feedback sequence corresponding to the channel information of multiple bandwidth units. The second information is used to indicate that the encoding and decoding result of the channel information of the first bandwidth unit among the multiple bandwidth units is inaccurate. Specifically, the channel information of the plurality of bandwidth units is input into an encoding model to obtain an encoding feedback sequence output by the encoding model; the encoding feedback sequence is input into a verification model to obtain verification channel information of the plurality of bandwidth units output by the verification model; based on the difference between the verification channel information of the first bandwidth unit and the channel information of the first bandwidth unit in the channel verification information of the plurality of bandwidth units, it is determined that the encoding and decoding result of the channel information of the first bandwidth unit is inaccurate; the encoding model includes an encoder model in an autoencoder neural network model, the encoder model is located on the second device side, and the verification model is located on the second device side; The channel feedback information further includes the codebook sequence corresponding to the channel information of the first bandwidth unit. Based on the channel information of the first bandwidth unit, the codebook sequence corresponding to the channel information of the first bandwidth unit is determined in the codebook set. The first device is used to input the codebook sequence into the decoding model to obtain the codebook recovery channel information of the first bandwidth unit, and to process the data to be transmitted on the first bandwidth unit based on the codebook recovery channel information. The decoding model includes a decoder model in an autoencoder neural network model, and the decoder model is located on the first device side.

2. The method according to claim 1, characterized in that, The method further includes: The channel information of the multiple bandwidth units is input into the coding model to obtain the coding feedback sequence output by the coding model; The encoded feedback sequence and the channel information of the plurality of bandwidth units are input into the verification model to obtain the second information output by the verification model.

3. The method according to claim 1 or 2, characterized in that, The coding model is used to jointly encode the channel information of the multiple bandwidth units, and the coding feedback sequence is a joint coding sequence corresponding to the channel information of the multiple bandwidth units, or... The coding model is used to encode the channel information of each of the plurality of bandwidth units, and the coding feedback sequence includes the sub-coding sequence corresponding to the channel information of each of the plurality of bandwidth units.

4. The method according to claim 3, characterized in that, The channel feedback sequence includes the coded feedback sequence, or, The coding feedback sequence includes a sub-coding sequence corresponding to the channel information of each of the plurality of bandwidth units, and the channel feedback sequence includes a sub-coding sequence corresponding to the channel information of a second bandwidth unit, wherein the second bandwidth unit is a bandwidth unit other than the first bandwidth unit among the plurality of bandwidth units.

5. The method according to claim 1, characterized in that, The verification model includes the decoder model in the autoencoder neural network model.

6. The method according to claim 2, characterized in that, Before inputting the encoded feedback sequence into the verification model, the method further includes: Receive third information from the first device, the third information being used to indicate the configuration parameter information of the decoding model of the first device, the decoding model being used to decode the channel feedback information; Configure the verification model based on the configuration parameter information.

7. A communication method, characterized in that, The method includes: The system receives first information from a second device. The first information includes channel feedback information and second information. The channel feedback information includes a channel feedback sequence corresponding to the channel information of multiple bandwidth units. The second information is used to indicate that the encoding and decoding result of the channel information of the first bandwidth unit among the multiple bandwidth units is inaccurate. Based on the first information, the recovered channel information of the plurality of bandwidth units is obtained, wherein the channel feedback information further includes the codebook sequence corresponding to the channel information of the first bandwidth unit, and the codebook sequence is input into the decoding model to obtain the codebook recovered channel information of the first bandwidth unit; the decoding model includes a decoder model in an autoencoder neural network model, and the decoder model is located on the first device side; The second device is configured to input the channel information of the plurality of bandwidth units into an encoding model to obtain an encoding feedback sequence output by the encoding model; input the encoding feedback sequence into a verification model to obtain verification channel information of the plurality of bandwidth units output by the verification model; determine that the encoding and decoding result of the channel information of the first bandwidth unit is inaccurate based on the difference between the verification channel information of the first bandwidth unit and the channel information of the first bandwidth unit in the channel verification information of the plurality of bandwidth units; the encoding model includes an encoder model in an autoencoder neural network model, the encoder model is located on the second device side, and the verification model is located on the second device side; the second device is further configured to determine the codebook sequence corresponding to the channel information of the first bandwidth unit in the codebook set.

8. The method according to claim 7, characterized in that, The channel feedback sequence is a joint coding sequence corresponding to the channel information of the plurality of bandwidth units, or the channel feedback sequence includes a sub-coding sequence corresponding to the channel information of each of the plurality of bandwidth units; And, obtaining the recovery channel information of the plurality of bandwidth units based on the first information includes: The channel feedback sequence is input into the decoding model to obtain the decoding recovery channel information for each of the plurality of bandwidth units.

9. The method according to claim 7, characterized in that, The channel feedback sequence includes a sub-coded sequence corresponding to the channel information of the second bandwidth unit, wherein the second bandwidth unit is a bandwidth unit other than the first bandwidth unit among the plurality of bandwidth units, and the channel feedback sequence also includes a codebook sequence corresponding to the channel information of the first bandwidth unit. And, obtaining the recovery channel information of the plurality of bandwidth units based on the first information includes: The channel feedback sequence is input into the decoding model to obtain the decoding recovery channel information of the second bandwidth unit and the codebook recovery channel information of the first bandwidth unit.

10. The method according to claim 7 or 9, characterized in that, The method further includes: Based on the codebook of the first bandwidth unit, channel information is recovered, and data is transmitted in the first bandwidth unit; and / or, Data is transmitted in the second bandwidth unit according to the channel recovery information of the second bandwidth unit, wherein the second bandwidth unit is a bandwidth unit other than the first bandwidth unit among the plurality of bandwidth units.

11. A communication device, characterized in that, include: A processing unit is configured to determine that the encoding and decoding result of the channel information of a first bandwidth unit among a plurality of bandwidth units is inaccurate. The unit inputs the channel information of the plurality of bandwidth units into an encoding model to obtain an encoding feedback sequence output by the encoding model; inputs the encoding feedback sequence into a verification model to obtain verification channel information of the plurality of bandwidth units output by the verification model; and determines that the encoding and decoding result of the channel information of the first bandwidth unit is inaccurate based on the difference between the verification channel information of the first bandwidth unit and the channel information of the first bandwidth unit in the channel verification information of the plurality of bandwidth units. The encoding model includes an encoder model in an autoencoder neural network model, the encoder model being located on the second device side, and the verification model being located on the second device side. A transceiver unit is configured to send first information to a first device. The first information includes channel feedback information and second information. The channel feedback information includes a channel feedback sequence corresponding to the channel information of the plurality of bandwidth units. The second information is used to indicate that the encoding and decoding result of the channel information of the first bandwidth unit among the plurality of bandwidth units is inaccurate. The channel feedback information also includes a codebook sequence corresponding to the channel information of the first bandwidth unit. The codebook sequence corresponding to the channel information of the first bandwidth unit is determined in the codebook set based on the channel information of the first bandwidth unit. The first device is used to input the codebook sequence into the decoding model to obtain the codebook recovery channel information of the first bandwidth unit, and to process the data to be transmitted on the first bandwidth unit based on the codebook recovery channel information. The decoding model includes a decoder model in an autoencoder neural network model, and the decoder model is located on the first device side.

12. The apparatus according to claim 11, characterized in that, The processing unit is also used for: The channel information of the multiple bandwidth units is input into the coding model to obtain the coding feedback sequence output by the coding model; The encoded feedback sequence and the channel information of the plurality of bandwidth units are input into the verification model to obtain the second information output by the verification model.

13. The apparatus according to claim 11 or 12, characterized in that, The coding model is used to jointly encode the channel information of the multiple bandwidth units, and the coding feedback sequence is a joint coding sequence corresponding to the channel information of the multiple bandwidth units, or... The coding model is used to encode the channel information of each of the plurality of bandwidth units, and the coding feedback sequence includes the coding sequence corresponding to the channel information of each of the plurality of bandwidth units.

14. The apparatus according to claim 13, characterized in that, The channel feedback sequence includes the coded feedback sequence, or, The coding feedback sequence includes a sub-coding sequence corresponding to the channel information of each of the plurality of bandwidth units, and the channel feedback sequence includes a sub-coding sequence corresponding to the channel information of a second bandwidth unit, wherein the second bandwidth unit is a bandwidth unit other than the first bandwidth unit among the plurality of bandwidth units.

15. The apparatus according to claim 11, characterized in that, The verification model includes the decoder model in the autoencoder neural network model.

16. The apparatus according to claim 11, characterized in that, The processing unit inputs the encoded feedback sequence into the verification model before, and, The transceiver unit is further configured to receive third information from the first device, the third information being used to indicate the configuration parameter information of the decoding model of the first device, the decoding model being used to decode the channel feedback information; The processing unit is also configured to configure the verification model based on the configuration parameter information.

17. A communication device, characterized in that, include: The transceiver unit is used to receive first information from the second device. The first information includes channel feedback information and second information. The channel feedback information includes a channel feedback sequence corresponding to the channel information of multiple bandwidth units. The second information is used to indicate that the encoding and decoding result of the channel information of the first bandwidth unit among the multiple bandwidth units is inaccurate. The processing unit is configured to obtain the recovered channel information of the plurality of bandwidth units based on the first information, wherein the channel feedback information further includes the codebook sequence corresponding to the channel information of the first bandwidth unit, and input the codebook sequence into the decoding model to obtain the codebook recovered channel information of the first bandwidth unit; the decoding model includes a decoder model in an autoencoder neural network model, and the decoder model is located on the first device side; The second device is configured to input the channel information of the plurality of bandwidth units into an encoding model to obtain an encoding feedback sequence output by the encoding model; input the encoding feedback sequence into a verification model to obtain verification channel information of the plurality of bandwidth units output by the verification model; determine that the encoding and decoding result of the channel information of the first bandwidth unit is inaccurate based on the difference between the verification channel information of the first bandwidth unit and the channel information of the first bandwidth unit in the channel verification information of the plurality of bandwidth units; the encoding model includes an encoder model in an autoencoder neural network model, the encoder model is located on the second device side, and the verification model is located on the second device side; the second device is further configured to determine the codebook sequence corresponding to the channel information of the first bandwidth unit in the codebook set.

18. The apparatus according to claim 17, characterized in that, The channel feedback sequence is a joint coding sequence corresponding to the channel information of the plurality of bandwidth units, or the channel feedback sequence includes a sub-coding sequence corresponding to the channel information of each of the plurality of bandwidth units; Furthermore, the processing unit is specifically used to input the channel feedback sequence into the decoding model to obtain the decoding recovery channel information of each bandwidth unit in the plurality of bandwidth units.

19. The apparatus according to claim 17, characterized in that, The channel feedback sequence includes a sub-coded sequence corresponding to the channel information of the second bandwidth unit, wherein the second bandwidth unit is a bandwidth unit other than the first bandwidth unit among the plurality of bandwidth units, and the channel feedback sequence also includes a codebook sequence corresponding to the channel information of the first bandwidth unit. Furthermore, the processing unit is specifically used to input the channel feedback sequence into the decoding model to obtain the decoding recovery channel information of the second bandwidth unit and the codebook recovery channel information of the first bandwidth unit.

20. The apparatus according to claim 17 or 19, characterized in that, The processing unit is also used for: Based on the codebook of the first bandwidth unit, channel information is recovered, and data is transmitted in the first bandwidth unit; and / or, Data is transmitted in the second bandwidth unit according to the channel recovery information of the second bandwidth unit, wherein the second bandwidth unit is a bandwidth unit other than the first bandwidth unit among the plurality of bandwidth units.

21. A communication device, characterized in that, include: Processor, memory, and interfaces for communicating with terminal devices; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory, causing the processor to perform the communication method as described in any one of claims 1 to 10.

22. A computer-readable storage medium, characterized in that, The method includes a computer program that, when executed by one or more processors, causes a device including the processor to perform the method as described in any one of claims 1 to 10.

23. A computer program product, characterized in that, The computer program product includes: a computer program that, when run, causes a computer to perform the method as described in any one of claims 1 to 10.

24. A chip, characterized in that, Includes at least one processor and a communication interface; The communication interface is used to receive signals input to the chip or signals output from the chip, and the processor communicates with the communication interface and implements the method as described in any one of claims 1 to 10 through logic circuits or execution code instructions.

Images