Communication method and communication apparatus

Abstract

The application discloses a communication method and a communication device. The method comprises the following steps: a terminal device is provided with an encoder and a first decoder, a network device is provided with a second decoder, and the first decoder and the second decoder are a student network model and a teacher network model in knowledge distillation technology, respectively; the terminal device obtains first feedback information by compressing CSI based on the encoder, inputs the first feedback information into the first decoder to obtain first reconstruction information, and determines first feedback accuracy based on the first reconstruction information; and the terminal device sends one of the first feedback information, second feedback information obtained by compressing CSI based on a codebook, or third feedback information obtained by compressing CSI based on the encoder after adjusting the compression rate to the network device based on the first feedback accuracy. Through the application, the terminal device can determine the feedback accuracy of the CSI feedback information based on the first decoder, and flexibly select the CSI feedback scheme according to the feedback accuracy, which is beneficial to improving the stability of the CSI feedback scheme performance.

Description

Technical Field

[0001] This application relates to the field of communications, and more particularly to communication methods and communication devices. Background Technology

[0002] In wireless communication systems, channel state information (CSI) is a key factor in optimizing transmission performance, and its accuracy and reliability directly determine the downlink transmission efficiency and quality. Among CSI feedback schemes, codebook-based and AI-based schemes are two typical examples. Codebook-based schemes have limitations in terms of computational complexity and feedback overhead. AI-based schemes excel in reducing feedback overhead and improving channel information compression efficiency, but their performance stability is poor in certain situations, such as when facing rapid changes in the channel environment or noise interference, which may lead to fluctuations in feedback accuracy.

[0003] How to provide a stable CSI feedback solution is an urgent problem to be solved. Summary of the Invention

[0004] This application provides a communication method and a communication device. Based on the method described in this application, it is beneficial to improve the performance stability of the CSI feedback scheme.

[0005] Firstly, a communication method is provided. This method can be executed by a terminal device, or by a component (such as a circuit, chip, or chip system) configured in the terminal device, or by a logic module or software capable of implementing all or part of the functions of the terminal device. This application does not limit this. The following description uses a terminal device as an example. The method includes:

[0006] The terminal device is equipped with a first decoder, and the network device is equipped with a second decoder. The first decoder and the second decoder represent the student network model and teacher network model in the knowledge distillation technique, respectively. The terminal device acquires first feedback information, which is obtained by compressing Codebook Information (CSI) based on the encoder. Further, the terminal device sends one of the following feedback information—first feedback information, second feedback information, or third feedback information—to the network device based on a first feedback accuracy: the second feedback information is obtained by compressing Codebook Information (CSI), the third feedback information is obtained by compressing CSI based on the encoder with an adjusted compression rate, and the first feedback accuracy is obtained by the terminal device based on first reconstruction information, which is information reconstructed from the first feedback information by the terminal device using the first decoder.

[0007] In the method provided in the first aspect, the terminal device deploys a first decoder obtained through knowledge distillation technology. This first decoder can simulate the prediction / data processing capabilities of a second decoder on the network device side. The first feedback information is feedback information obtained by the terminal device based on an AI-based CSI feedback scheme. Before the terminal device sends this first feedback information to the network device, the terminal device can reconstruct the first feedback information through the first decoder to obtain first reconstructed information, and monitor the feedback accuracy (i.e., the first feedback accuracy) of the first feedback information based on the first reconstructed information. Furthermore, the terminal device can dynamically select either an AI-based CSI feedback scheme or a codebook-based CSI feedback scheme based on the first feedback accuracy. Alternatively, in the AI-based CSI feedback scheme, the terminal device can dynamically select whether to adjust the encoder's compression rate based on the first feedback accuracy, which is beneficial for improving the reliability and stability of CSI feedback. In particular, compared to the CSI feedback method based on knowledge distillation technology (i.e., the specific network structure design and training method for reconstructing CSI through knowledge distillation technology), this application can determine the first feedback accuracy based on the first decoder and dynamically select the feedback information based on the first feedback accuracy, which is beneficial for improving the reliability and stability of CSI feedback.

[0008] In one possible implementation, if the first feedback accuracy is greater than or equal to a first accuracy threshold, the terminal device sends first feedback information to the network device; if the first feedback accuracy is less than the first accuracy threshold, the terminal device sends second or third feedback information to the network device. Through this possible implementation, when the first feedback accuracy meets system requirements (e.g., the first feedback accuracy is greater than or equal to the first accuracy threshold), the terminal device selects a feedback scheme that reduces CSI feedback overhead (i.e., an AI-based feedback scheme) and sends the first feedback information obtained by encoder compression of CSI to the network device; when the first feedback accuracy does not meet system requirements, the terminal device selects a feedback scheme with more stable CSI feedback performance (i.e., a codebook-based feedback scheme, or a scheme that adjusts the encoder's compression rate in an AI-based feedback scheme) and sends the second or third feedback information to the network device. This approach helps to reduce CSI feedback overhead while also improving the performance stability of the CSI feedback scheme.

[0009] In one possible implementation, before sending the second or third feedback information to the network device, the terminal device sends the first feedback information to the network device. Further, the terminal device receives a first or second instruction from the network device. The first instruction instructs the terminal device to provide the second feedback information, and the second instruction instructs the terminal device to provide the third information or adjust the encoder's compression rate. Through this possible implementation, even if the terminal device detects that the accuracy of the first feedback information does not meet system requirements, it will still provide the first feedback information to the network device. The network device then verifies whether the accuracy of the first feedback information does not meet system requirements (i.e., determines whether the terminal device needs to provide the second or third feedback information). This helps avoid the situation where the terminal device directly provides the second or third feedback information when the first decoder has an error, thus increasing the CSI feedback overhead.

[0010] In one possible implementation, the terminal device also sends a feedback information quality identifier to the network device, which is used to indicate whether the first feedback accuracy is less than a first accuracy threshold.

[0011] In one possible implementation, the terminal device also sends a CSI quality report to the network device. This CSI quality report includes information indicating the accuracy of the first feedback and / or channel state parameters. Through this possible implementation, the terminal device can send a CSI quality report to the network device, allowing the network device to understand the current channel conditions or verify the performance of the first decoder, which helps improve the stability of the first decoder's performance.

[0012] Secondly, a communication method is provided, which can be executed by a network device, or by a component (such as a circuit, chip, or chip system) configured in the network device, or by a logic module or software capable of implementing all or part of the functions of the network device. This application does not limit this. The following description uses a network device as an example. The method includes: the network device receiving first feedback information from a terminal device, the first feedback information being obtained based on encoder-compressed CSI; if the first feedback accuracy of the first feedback information is less than a first accuracy threshold, the network device receiving second or third feedback information from the terminal device, the second feedback information being obtained based on codebook-compressed CSI, the third feedback information being obtained based on encoder-compressed CSI with adjusted compression ratio, the first feedback accuracy being obtained based on first reconstructed information, the first reconstructed information being information reconstructed from the first feedback information by a first decoder; wherein the first decoder is a decoder deployed on the terminal device, the second decoder is a decoder deployed on the network device, the first decoder is a student network model in knowledge distillation technology, and the second decoder is a teacher network model in knowledge distillation technology.

[0013] The beneficial effects of the method described in the second aspect can be found in the beneficial effects of possible implementations of the first aspect, which will not be repeated here.

[0014] In one possible implementation, before the network device receives the second or third feedback information, the network device sends a first or second instruction to the terminal device. The first instruction is used to instruct the terminal device to provide the second feedback information, and the second instruction is used to instruct the terminal device to adjust the compression rate of the encoder. Alternatively, the second instruction is used to instruct the terminal device to provide the third feedback information.

[0015] In one possible implementation, the network device receives a feedback information quality identifier from the terminal device, which is used to indicate whether the first feedback accuracy is less than a first accuracy threshold.

[0016] In one possible implementation, the network device receives a CSI quality report from the terminal device, the CSI quality report including information and / or channel state parameters indicating the accuracy of the first feedback.

[0017] Thirdly, embodiments of this application provide a communication device for executing the methods of the first aspect to the second aspect, or any possible implementation of the first aspect to the second aspect. The communication device includes modules having the ability to execute the methods of the first aspect to the second aspect, or any possible implementation of the first aspect to the second aspect.

[0018] Fourthly, embodiments of this application provide a communication device including a processing circuit for executing methods from the first aspect to the second aspect, or any possible implementation of the first aspect to the second aspect. The processing circuit executes a program stored in a memory, and when the program is executed, the methods shown in any of the first aspect to the second aspect or any possible implementation thereof are executed.

[0019] In one possible implementation, the memory is located outside the aforementioned communication device.

[0020] In one possible implementation, the memory is located within the aforementioned communication device.

[0021] In this embodiment, the processing circuitry and memory can be integrated into a single device; that is, the processing circuitry and memory can be integrated together. For example, the communication device can be a chip.

[0022] In one possible implementation, the communication device further includes a transceiver circuit for receiving information (or inputting information) or sending information (or outputting information).

[0023] Fifthly, embodiments of this application provide a communication device, which includes a processing circuit and a transceiver circuit. The processing circuit can be a logic circuit, and the transceiver circuit can be an interface circuit. The logic circuit and the interface circuit are coupled. The interface circuit is used to input and / or output information, and the logic circuit is used to execute the method in the first aspect to the second aspect, or any possible implementation of the first aspect to the second aspect.

[0024] In a sixth aspect, embodiments of this application provide a chip including a processing circuit and an interface circuit, the processing circuit and the interface circuit being coupled; the interface circuit is used for inputting and / or outputting information, and the processing circuit is used for executing code instructions to cause the method shown in any of the first to second aspects or any possible implementation thereof to be executed.

[0025] In a seventh aspect, embodiments of this application provide a computer-readable storage medium for storing a computer program that, when run on a computer, causes the methods shown in any of the first to third aspects or any possible implementation thereof to be executed.

[0026] Eighthly, embodiments of this application provide a computer program product that, when run on a computer, causes the methods shown in any of the first to second aspects or any possible implementations described above to be executed.

[0027] Ninthly, this application provides a communication system, which includes a terminal device and a network device. The terminal device is used to execute the method shown in any possible implementation of any of the first to second aspects above, and the network device is used to execute the method shown in any possible implementation of any of the first to second aspects above. Attached Figure Description

[0028] Figure 1 This is a schematic diagram of the architecture of a communication system provided in an embodiment of this application;

[0029] Figure 2 This is a schematic diagram of a neural network provided in an embodiment of this application;

[0030] Figure 3 This is a schematic diagram illustrating the collaborative operation of an encoder and decoder provided in an embodiment of this application;

[0031] Figure 4 This is a schematic diagram of a communication architecture provided in an embodiment of this application;

[0032] Figure 5 This is a data flow diagram of the AI-based CSI feedback scheme provided in the embodiments of this application;

[0033] Figure 6 This is a flowchart illustrating a communication method provided in an embodiment of this application;

[0034] Figure 7 This is a flowchart illustrating another communication method provided in an embodiment of this application;

[0035] Figure 8 This is a flowchart illustrating another communication method provided in an embodiment of this application;

[0036] Figure 9 This is a schematic diagram of the structure of a communication device provided in an embodiment of this application;

[0037] Figure 10 This is a schematic diagram of another communication device provided in an embodiment of this application;

[0038] Figure 11 This is a schematic diagram of a hardware system of a communication device provided in an embodiment of this application. Detailed Implementation

[0039] The technical solutions in the embodiments of this application will now be described with reference to the accompanying drawings.

[0040] The technical solutions provided in this application can be applied to various communication systems, such as: Global System for Mobile Communications (GSM) systems, General Packet Radio Service (GPRS), Wireless Local Area Network (WLAN), Long Term Evolution (LTE) systems, LTE Frequency Division Duplex (FDD) systems, LTE Time Division Duplex (TDD) systems, sidelink communication systems, Universal Mobile Telecommunication System (UMTS), Worldwide Interoperability for Microwave Access (WiMAX) communication systems, non-terrestrial network (NTN) communication systems, 5th generation (5G) mobile communication systems or new radio access technology (NR), 6th generation mobile communication systems, or future mobile communication systems. Among these, 5G mobile communication systems can include non-standalone (NSA) and / or standalone (SA) networking. The technical solutions provided in this application can also be applied to future communication systems. This application does not impose any limitations on this.

[0041] Figure 1 The diagram illustrates the architecture of a communication system applicable to embodiments of this application. The communication system 10 includes a radio access network (RAN) 100, a core network (CN) 200, and a data network (DN) 300. The RAN 100 includes at least one RAN node (e.g., ...). Figure 1 110a and 110b (collectively referred to as 110) and at least one terminal (such as Figure 1 RAN 100, denoted as RAN 120a-120j, is collectively referred to as RAN 120. RAN 100 may also include other RAN nodes, such as wireless relay equipment and / or wireless backhaul equipment. Figure 1(Not shown in the image). Terminal 120 is connected to RAN node 110 wirelessly. RAN node 110 is connected to core network 200 wirelessly or via wired connection. The core network equipment in core network 200 and RAN node 110 in RAN 100 can be different physical devices, or they can be the same physical device integrating core network logical functions and radio access network logical functions. Core network 200 can also be connected to data network 300 wirelessly or via wired connection.

[0042] RAN 100 can be a cellular system related to the 3rd Generation Partnership Project (3GPP), such as 4G, 5G mobile communication systems, or future-oriented evolution systems. RAN 100 can also be an open access network (O-RAN or ORAN), a cloud radio access network (CRAN), or a wireless fidelity (WiFi) system. RAN 100 can also be a communication system that integrates two or more of the above systems.

[0043] Specifically, the following sections will discuss... Figure 1 The terminal equipment, RAN nodes (also known as access network equipment), core network equipment, and data network involved in the communication system are described in detail.

[0044] I. Terminal equipment (also called terminal)

[0045] Terminal equipment can be a wireless terminal device capable of receiving network device scheduling and instruction information. A wireless terminal device can be a device providing voice and / or data connectivity to a user, a handheld device with wireless connectivity, or other processing devices connected to a wireless modem. For example, a terminal device can communicate with one or more core networks or the Internet via a radio access network (RAN). Terminal equipment can also be called a terminal, user equipment (UE), mobile station, mobile terminal, etc. Terminal equipment can be widely used in various scenarios, such as device-to-device (D2D), vehicle-to-everything (V2X) communication, machine-type communication (MTC), Internet of Things (IoT), ultra-reliable low-latency communication (URLLC), virtual reality, augmented reality, industrial control, autonomous driving, telemedicine, smart grids, smart furniture, smart offices, smart wearables, smart transportation, smart cities, or satellite communication, etc. The terminal can be a mobile phone, tablet computer, computer with wireless transceiver capabilities, wearable device, vehicle, aircraft (such as drone, helicopter, airplane), hot air balloon, ship, robot, robotic arm, or smart home device, etc. The embodiments of this application do not limit the form of the terminal device.

[0046] II. RAN Nodes

[0047] RAN node 110, also known as access network equipment, RAN entity, or access node, constitutes part of the communication system and is used to help terminals achieve wireless access. Multiple RAN nodes 110 in communication system 10 can be of the same type or different types. In some scenarios, the roles of RAN node 110 and terminal 120 are relative, for example... Figure 1 Network element 120i can be a helicopter or a drone, and it can be configured as a mobile base station. For terminals 120j that access RAN 100 through network element 120i, network element 120i is a base station; however, for base station 110a, network element 120i is a terminal. RAN node 110 and terminal 120 are sometimes referred to as communication devices, for example... Figure 1 Network elements 110a and 110b can be understood as communication devices with base station functions, while network elements 120a-120j can be understood as communication devices with terminal functions.

[0048] In one possible scenario, a RAN node can be a base station, an evolved NodeB (eNodeB), an access point (AP), a transmission reception point (TRP), a next-generation NodeB (gNB), a base station in a future mobile communication system, or an access node in a WiFi system, etc. Figure 1 110a), micro base stations or indoor stations (such as Figure 1 The RAN node can be a relay node or donor node (as described in section 110b), or a wireless controller in a CRAN scenario. Optionally, the RAN node can also be a server, wearable device, vehicle, or in-vehicle equipment. For example, the access network equipment in vehicle-to-everything (V2X) technology can be a roadside unit (RSU). All or part of the functions of the RAN node in this application can also be implemented through software functions running on hardware, or through virtualization functions instantiated on a platform (e.g., a cloud platform). The RAN node can also be equipped with communication modules, circuits, or chips that perform corresponding communication functions. The RAN node can also be configured with program instructions for performing corresponding communication functions and corresponding program instructions. The RAN node in this application can also be a logical node, logical module, or software capable of implementing all or part of the RAN node functions.

[0049] In another possible scenario, multiple RAN nodes collaborate to assist the terminal in achieving wireless access, with different RAN nodes each implementing a portion of the base station's functions. For example, RAN nodes can be central units (CUs), distributed units (DUs), CU-control plane (CPs), CU-user plane (UPs), or radio units (RUs), etc. CUs and DUs can be set up separately or included in the same network element, such as a baseband unit (BBU). RUs can be included in radio frequency equipment or radio frequency units, such as remote radio units (RRUs), active antenna units (AAUs), or remote radio heads (RRHs).

[0050] In different systems, CU (or CU-CP and CU-UP), DU, or RU may have different names, but those skilled in the art will understand their meaning. For example, in an ORAN system, CU can also be called O-CU (open CU), DU can also be called O-DU, CU-CP can also be called O-CU-CP, CU-UP can also be called O-CU-UP, and RU can also be called O-RU. For ease of description, this application uses CU, CU-CP, CU-UP, DU, and RU as examples. Any of the units among CU (or CU-CP, CU-UP), DU, and RU in this application can be implemented through software modules, hardware modules, or a combination of software and hardware modules.

[0051] III. Core Network Equipment

[0052] Core network equipment refers to the equipment in the core network that provides service support for terminal devices. It is primarily responsible for registration, call setup, billing, mobility management, providing user connectivity, managing users, and carrying out service delivery, data processing, and routing. Core network equipment can correspond to different devices in different communication systems. For example, in a 4G communication system, it may correspond to one or more of the following: Mobility Management Entity (MME), Serving Gateway (S-GW), etc. Similarly, in a 5G communication system, it may correspond to one or more of the following: Access and Mobility Management Function (AMF) network elements, Session Management Function (SMF) network elements, User Plane Function (UPF) network elements, etc. In next-generation or future communication systems, it may correspond to one or more network elements, devices, or entities that provide service support for terminal devices.

[0053] IV. Data Network (DN)

[0054] A data network (DN) is a network located outside the mobile communication system that provides services to users. For example, a DN can be a packet data network (PDN), such as the Internet, Internet Protocol Multimedia Service (IMS) networks, dedicated data networks for certain applications, Ethernet, or Internet Protocol (IP) local area networks. The IP protocol can be IPv4 or IPv6. A DN can deploy various services, providing data and / or voice services to terminal devices. A DN can contain multiple application servers (AS), each providing at least one service.

[0055] It should be noted that, Figure 1 The communication system shown is not limited to the terminal equipment, access network equipment, core network equipment and data network shown in the figure, but may also include other equipment not shown in the figure. These will not be listed here.

[0056] In this application, the apparatus for implementing the functions of a terminal device can be the terminal device itself, or any apparatus capable of supporting the terminal device in implementing those functions, such as a processor, circuit, chip, or chip system. This apparatus can be installed in or connected to the terminal device. In the technical solutions provided in this application, the example of a terminal device being used to implement the functions of a terminal device is used to describe the technical solutions provided in this application.

[0057] The network devices in this application may include network-side devices such as access network devices and core network devices. For example, the data / signaling interaction between the network devices and terminal devices mentioned in this application may be data / signaling interaction between access network devices and terminal devices, or data / signaling interaction between core network devices and terminal devices through access network devices. In this application, the means for implementing the functions of the network devices may be the network devices themselves, or it may be any means that enables the network devices to implement those functions, such as processors, circuits, chips, or chip systems. These means may be installed in or connected to the network devices. In the technical solutions provided in this application, the example of a network device being used to implement the functions of the network devices is used to describe the technical solutions provided in this application.

[0058] To facilitate understanding of the embodiments of this application, the terminology used in this application will be briefly explained first. Optionally, the explanation of some terms may also refer to the explanations in the 3rd Generation Partnership Project (3GPP) standard protocol.

[0059] 1. CSI

[0060] In wireless communication systems, Channel Information System (CSI) is information reported by the receiver (e.g., a terminal device) to the transmitter (e.g., a network device) describing the channel attributes of the wireless communication link between the transmitter and receiver. CSI may include, but is not limited to, precoding matrix indicator (PMI), rank indicator (RI), channel quality indicator (CQI), channel state information reference signal (CSI-RS resource indicator (CRI), and layer indicator (LI).

[0061] 2. Precoding matrix

[0062] The precoding matrix can be, for example, a precoding matrix determined by the terminal device based on the channel matrix of each frequency domain unit (e.g., the frequency domain length of a frequency domain unit can be a subband, a resource block (RB), or R times the subband, where R <= 1, and R can take the value of 1 or 1 / 2). This channel matrix can be determined by the terminal device through methods such as channel estimation or based on channel reciprocity. However, it should be understood that the specific methods used by the terminal device to determine the precoding matrix are not limited to those described above, and will not be listed here.

[0063] For example, the precoding matrix can be obtained by performing singular value decomposition (SVD) on the channel matrix or its covariance matrix, or by performing eigenvalue decomposition (EVD) on the covariance matrix of the channel matrix. It should be understood that the methods for determining the precoding matrix listed above are merely examples and should not constitute any limitation on this application; they will not be listed individually here.

[0064] It should be noted that the precoding matrix can be directly used for downlink data transmission; alternatively, it can be processed using beamforming methods, such as zero forcing (ZF), regularized zero-forcing (RZF), minimum mean-squared error (MMSE), and signal-to-leakage-and-noise ratio (SLNR), to obtain the final precoding matrix for downlink data transmission. This application does not limit this. Unless otherwise specified, the precoding matrix mentioned below refers to the precoding matrix determined based on the method provided in this application.

[0065] It is understood that the precoding matrix determined by the terminal device can be interpreted as the precoding matrix to be fed back. Optionally, the terminal device can indicate the precoding matrix to be fed back via the PMI, so that the network device can recover the precoding matrix based on the PMI. It is understood that the precoding matrix recovered by the network device based on the PMI can be the same as or similar to the aforementioned precoding matrix to be fed back.

[0066] In downlink channel measurement, the higher the approximation between the precoding matrix determined by the network device based on the PMI and the precoding matrix determined by the terminal device, the better the precoding matrix determined by the network device for data transmission can be adapted to the channel state, thus improving the signal reception quality.

[0067] Optionally, in the CSI feedback scheme, the entire CSI can be fed back to the network device, for example, by compressing and quantizing the channel matrix (also known as the CSI matrix) before feeding it back to the network device. Alternatively, the precoding matrix can be fed back to the network device, for example, by compressing and quantizing the precoding matrix before feeding it back to the network device.

[0068] 3. AI, AI models, and neural networks

[0069] AI can refer to the intelligence exhibited by machines created by humans. Generally, artificial intelligence refers to the technology of using ordinary computer programs to represent human intelligence. Artificial intelligence can be defined as a machine or computer that imitates humans and possesses cognitive functions related to human thinking, such as learning and problem-solving. Artificial intelligence can learn from past experiences, make rational decisions, and respond quickly. The goal of artificial intelligence is to understand intelligence by constructing computer programs that demonstrate symbolic reasoning or logical reasoning.

[0070] An AI model is an algorithm or computer program that enables AI functionality. An AI model represents the mapping relationship between the model's inputs and outputs. An AI model can be a neural network or other machine learning model.

[0071] Neural networks (NNs) are a specific implementation of machine learning, a machine learning technique that simulates the neural network of the human brain in an attempt to achieve artificial intelligence-like results. According to the general approximation theorem, neural networks can theoretically approximate any continuous function, thus enabling them to learn arbitrary mappings. Therefore, neural networks can accurately abstract and model complex, high-dimensional problems.

[0072] Neural networks typically consist of multiple layers, each containing one or more neurons. Increasing the depth and / or width of a neural network enhances its expressive power, providing more robust information extraction and abstract modeling capabilities for complex systems. The depth of a neural network can be understood as the number of layers, while the number of neurons in each layer can be called the width of that layer. See also Figure 2 , Figure 2 This is a schematic diagram of a neural network provided in an embodiment of this application. In one implementation, the neural network includes an input layer and an output layer. The input layer of the neural network processes the received input through neurons and then passes the result to the output layer, which obtains the output result of the neural network. In another implementation, the neural network includes an input layer, hidden layers, and an output layer. The input layer of the neural network processes the received input through neurons and then passes the result to an intermediate hidden layer. The hidden layer then passes the calculation result to the output layer or an adjacent hidden layer, and finally, the output layer obtains the output result of the neural network. A neural network may include one or more hidden layers connected in sequence; this application does not limit the number of hidden layers in a neural network. Figure 2 The neural network shown includes one input layer, one hidden layer, and one output layer. The input layer has three neurons, the hidden layer has four neurons, and the output layer has two neurons. This is understandable. Figure 2 The number of layers and the number of neural network elements in each layer shown are merely examples.

[0073] Each connection between neurons corresponds to a weight (its value is called a weight), and these weights can be updated through training. Each neuron can also correspond to a bias value, which can be updated through training. Updating a neural network means updating these weights and bias values. Knowing the structure of a neural network—that is, how the output of one neuron is input into another—as well as the weights and biases, we know all the information about the neural network.

[0074] 4. Training dataset

[0075] AI is a data-driven technology. Before using an AI model, it needs to be trained. Model training refers to using a training dataset, selecting an appropriate loss function, and using optimization algorithms to train the model parameters so that the value of the loss function is minimized on the training dataset.

[0076] Data collection is the process of building a training dataset. In the field of wireless communication, data collection can be simulation data collected through a simulation platform, experimental data collected in experimental scenarios, or measured data collected in actual wireless communication networks.

[0077] 5. Knowledge distillation (KD) technology

[0078] Knowledge distillation can be understood as a neural network compression technique that can transfer the learning parameters or knowledge of a complex neural network model to a simpler neural network model, enabling the simpler neural network model to make the same or similar inferences (i.e., predictions) as the complex neural network model with less computational resource cost and time.

[0079] In knowledge distillation, complex neural network models can be called teacher neural network models (T) or simply teacher network models, while simpler neural network models can be called student neural network models (S) or simply student network models. Optionally, the student and teacher network models can be isomorphic or heterogeneous neural network models. The teacher network model requires a larger training dataset than the student network model.

[0080] 6. AI-based CSI feedback scheme

[0081] The AI-based CSI feedback scheme can be broadly described as follows: the terminal device compresses and quantizes the CSI using an encoder, and the network device reconstructs the CSI using a decoder. In this AI-based CSI feedback scheme, the encoder and decoder work collaboratively. Optionally, the encoder and decoder can be jointly trained end-to-end; alternatively, they can be trained separately, with the network device training the decoder independently and the terminal device training the encoder independently. During encoder and decoder training, the parameters are optimized to ensure that the CSI reconstructed by the decoder is as close as possible to the CSI before compression and quantization.

[0082] For example, see Figure 3As shown, the encoder input is the CSI (Current Sequence Index) obtained by measuring the reference signal from the terminal device. Figure 3 (represented by V in the code), the encoder output is feedback information ( Figure 3 In the diagram, S represents the input of the decoder, which is the feedback information. Figure 3 In the diagram, S represents the CSI (or reconstructed information) obtained from the reconstruction. Figure 3 Chinese express). Figure 3 The training method for the encoder and decoder can be to minimize the CSI of the decoder output (i.e., Figure 3 In ) and encoder input CSI (i.e. Figure 3 The difference between V in the text.

[0083] The reference signal used for measuring CSI can be a channel state information reference signal (CSI-RS), a synchronizing signal / physical broadcast channel block (SS / PBCH block, SSB), or a demodulation reference signal (DMRS); this application does not limit its use. The channel measured by CSI can be a physical downlink shared channel (PDSCH) or a physical downlink control channel (PDCCH).

[0084] 7. Codebook-based CSI feedback scheme

[0085] A codebook is a predefined set of vectors or matrices used in wireless communication to quantize CSI (Channel Signal Indication) vectors or matrices. Each entry in the codebook represents a possible channel state (such as beam direction, amplitude / phase combination, etc.). The terminal device selects the entry from the codebook that is closest to the actual channel and feeds it back to the network device. Codebooks typically include Type I and Type II codebooks.

[0086] Optionally, the CSI feedback scheme based on the Type I codebook is suitable for low-to-medium resolution requirements (such as single-user MIMO or low-rank computation) scenarios, while the CSI feedback scheme based on the Type II codebook is suitable for high resolution requirements (such as multi-antenna, multiple-input multiple-output (MIMO) scenarios and more complex channel quantization scenarios).

[0087] The following explanation uses a Type II codebook as an example to illustrate the codebook-based CSI feedback scheme. The Type II codebook provides channel state feedback through linear combination and amplitude / phase quantization. The terminal device can select a set of basis vectors from a preset Discrete Fourier Transform (DFT) codebook to represent the channel spatial characteristics, and then perform amplitude and phase quantization on the linear combination coefficients of these basis vectors. Further, the terminal device sends the index of the basis vectors and the quantized linear combination coefficients to the network device.

[0088] In the CSI feedback process, terminal devices typically perform CSI feedback according to a pre-defined CSI feedback scheme. Before feedback, the terminal device cannot know the performance of the current CSI feedback, i.e., it cannot know the deviation between the CSI reconstructed or recovered by the network device based on the feedback information and the original CSI measured by the terminal device. However, codebook-based feedback schemes have limitations in terms of computational complexity and feedback overhead. While AI-based feedback schemes excel in reducing feedback overhead and improving channel information compression efficiency, their performance stability is poor in certain situations (such as rapid changes in the channel environment or noise interference). Therefore, how to provide a stable CSI feedback scheme while reducing feedback overhead is a pressing problem to be solved.

[0089] In one possible approach, a decoder with the same structure as the decoder in the network device (hereinafter referred to as the second decoder) is deployed on the terminal device. This allows the terminal device to input the first feedback information into the decoder to obtain the reconstructed CSI before performing CSI feedback based on AI (i.e., the terminal device sending the first feedback information obtained by the encoder-compressed CSI to the network device), and then determine the feedback accuracy of the first feedback information based on the reconstructed CSI. However, terminal devices typically have limited computing resources, and the structure of the second decoder is relatively complex, making it difficult for terminal devices to support the deployment of a decoder with the exact same structure as the second decoder. Furthermore, even when the terminal device supports the deployment of a decoder with the same structure as the second decoder, the complexity of the decoder means that the process of obtaining reconstructed information through it usually takes a long time, potentially introducing significant latency into the CSI feedback process.

[0090] The communication method and apparatus provided in this application deploy a first decoder obtained through knowledge distillation technology. This first decoder can simulate the prediction / data processing capabilities of a second decoder on the network device side, and its structural complexity is lower than that of the network device's second decoder. Before feeding back the first feedback information, the terminal device can use the first decoder to determine the feedback performance of the first feedback information and adjust the CSI feedback scheme according to the feedback performance. This helps to reduce the additional CSI feedback latency, reduce CSI feedback overhead, and improve the performance stability of the CSI feedback scheme.

[0091] The solution provided in this application will be described in detail below with reference to the corresponding flowcharts. It is understood that the illustrative flowcharts provided in this application primarily use different devices (e.g., terminal devices, network devices) as examples of the execution subjects of this interactive illustration to illustrate the method, but this application does not limit the execution subjects of the interactive illustrations. For example, the devices (e.g., terminal devices, network devices) in the illustrative flowcharts can also be chips, chip systems, or processors that support the implementation of this method on the device, or logic modules or software that can implement all or part of the functions of the device.

[0092] As a general statement, the message or signaling interactions involved in the interaction process of this application embodiment can be standard messages or signaling or newly introduced messages or signaling. This application embodiment does not make specific limitations on this.

[0093] To facilitate understanding of the communication method provided in this application, the following is an example of a communication architecture applicable to the communication method of this application.

[0094] Figure 4 This is a schematic diagram of a communication architecture provided in an embodiment of this application. The communication architecture includes at least one network device (such as...). Figure 4 Network device 410) and at least one terminal device (such as Figure 4 The terminal device 420 is equipped with an encoder and a first decoder, while the network device 410 is equipped with a second decoder. In this application, the decoder deployed on the terminal device is referred to as the first decoder, but this application does not specifically limit the name; for example, it could also be called a proxy decoder.

[0095] In the AI-based CSI feedback scheme, the encoder and the second decoder work collaboratively. Optionally, the encoder and the second decoder can be jointly trained end-to-end; alternatively, they can be trained separately, with the network device training the second decoder independently and the terminal device training the encoder independently. During the training of the encoder and the second decoder, the parameters of the encoder and the second decoder are optimized so that the CSI reconstructed by the second decoder is as close as possible to the CSI before compression and quantization by the encoder.

[0096] The first decoder is obtained through knowledge distillation and the second decoder. Specifically, the first decoder is the student network model in the knowledge distillation technique, and the second decoder is the teacher network model. The inputs to the first and second decoders are the same, denoted as S. The training objective of the first decoder can be described as: under the same input, minimizing the difference between the outputs of the first and second decoders, and reducing the discrepancy between the outputs of the second and first decoders. This training objective can be described as shown in Equation (1).

[0097]

[0098] in, Indicates the first decoder, This represents the parameters of the first decoder, where S is the input to the first decoder. This is the output of the first decoder. This is the output of the second decoder.

[0099] Alternatively, the first decoder can be understood as follows: the first decoder is obtained by training an initial decoder (which can be understood as a neural network model that is isomorphic or heterogeneous with the second decoder) on a training dataset. This training dataset includes M training data points, where M is a positive integer. Each training data point consists of sample data and sample labels, which are obtained by inputting the sample data into the second decoder.

[0100] Optionally, the sample data in the M training data points correspond to one or more channel scenarios (e.g., channel scenarios with different fading rates or different interference intensities). This method can improve the diversity of the training data, which is beneficial to improving the performance of the first decoder.

[0101] For example, a device used to train the first decoder (such as a terminal device, a server with a communication connection to the terminal device, a network device, or a server with a communication connection to the network device) collects M CSI data under at least one channel scenario; further, the M CSI data are input into the encoder to obtain M sample data, and the M sample data are input into the second decoder to obtain M sample labels.

[0102] Optionally, before deploying the first decoder, the device used to train the first decoder may also determine a performance metric for the first decoder and update or retrain the network model of the first decoder based on the performance metric. This performance metric may include at least one of the following: raw prediction error, average prediction error, recall, or consistency parameter.

[0103] For example, in a MIMO system employing N subbands of orthogonal frequency division multiplexing (OFDM), each subband corresponds to a CSI matrix. In this scenario, the device used to train the first decoder acquires, for example... Figure 5 The CSI test data shown (denoted as V) 测 V 测 =[v1,…,v i ,…,v N ], the v i Let V be the CSI matrix corresponding to the i-th subband out of N subbands, where i is a positive integer less than or equal to N. The device used to train the first decoder will use V... 测 The input encoder yields the test sample (denoted as S). 测 ), will S 测 Inputting the first decoder yields the reconstructed information (denoted as...). ), Should The reconstruction information corresponding to the i-th subband reconstructed by the first decoder; S 测 Inputting the second decoder yields the reconstructed information (denoted as...). ), Should This refers to the reconstruction information corresponding to the i-th subband reconstructed by the second decoder. Further, the device used to train the first decoder uses the information reconstructed by the first decoder... Determine S 测 Feedback accuracy Should According to The feedback accuracy corresponding to the determined i-th sub-band; and based on the reconstruction from the second decoder Determine S 测 Actual feedback accuracy ρ 2 , Should According to The feedback accuracy corresponding to the determined i-th sub-band. The device used to train the first decoder is based on the degree... or At least one of the following is used to calculate the performance metric of the first decoder: raw prediction error, average prediction error, recall, or consistency parameter.

[0104] Raw prediction error: Characterized by the performance difference between the first decoder and the second decoder due to differences in structural complexity and generalization ability, denoted as . Should The original prediction error is used to indicate With ρ 2 The average of the root mean squared errors.

[0105] Average prediction error: used to indicate the error after compensating for the original prediction error. With ρ 2 The average of the root mean squared errors, denoted as G. ρ2 ,Should

[0106] Recall rate: Let the recall rate be R, which represents the percentage of a given number of days in a given year. Where TP represents a test dataset, Greater than or equal to a certain precision threshold (such as the first precision threshold mentioned later), and ρ 2 The number of test data points that are also greater than or equal to the precision threshold; FN represents the number of test data points in a set of test data. Less than this precision threshold, and ρ 2 The number of test data is also less than the accuracy threshold.

[0107] Consistency parameter: used to represent and The SGCS between them is denoted as Should

[0108] In one possible implementation, the average prediction error G ρ2 If the error threshold is greater than or equal to the error threshold, the device used to train the first decoder updates the model of the first decoder. The error threshold can be an empirical value or a preset value of the network device, and its specific value can be adjusted according to the actual application scenario; this application does not impose specific limitations on it.

[0109] In one possible implementation, the consistency parameter If the consistency threshold is less than or equal to the threshold value, the device used to train the first decoder retrains the decoder deployed on the terminal device side based on a new training dataset (different from the training dataset of the first decoder). The consistency threshold can be an empirical value or a preset value for the network device, and its specific value can be adjusted according to the actual application scenario; this application does not impose specific limitations on it.

[0110] Figure 6 This is a schematic diagram of a communication method 60 according to an embodiment of this application. It can be understood that... Figure 6 The terminal device in the middle can be Figure 1 or Figure 4 Any terminal device in the context of network equipment can also refer to a component within that terminal device (such as a processor, chip, or chip system). Network equipment can be... Figure 1 or Figure 4 Any access network device, or a component within an access network device (such as a processor, chip, or chip system). Figure 6 As shown, the communication method 60 includes the following steps:

[0111] S601. The terminal device obtains first feedback information, which is obtained based on encoder compression CSI.

[0112] Network devices send reference signals (such as CSI-RS, SSB, DMRS, etc.) to terminal devices. After receiving the reference signals, the terminal devices perform channel estimation to obtain the CSI of the downlink channel. The terminal devices then use an encoder (e.g., Figure 4 The encoder shown compresses and quantizes the CSI to obtain the first feedback information. This first feedback information can be in the form of a feature vector, a matrix, or a bit sequence, and this application does not specifically limit it.

[0113] After determining the first feedback information, the terminal device determines the first feedback accuracy of that first feedback information. For example, the terminal device uses a first decoder (e.g., Figure 4 The first decoder (shown) reconstructs the first feedback information to obtain the reconstructed CSI, which is denoted as the first reconstructed information. Further, the terminal device calculates the first feedback accuracy of this first feedback information based on the first reconstructed information and the CSI before encoder compression (i.e., the encoder input).

[0114] The feedback precision (e.g., the first feedback precision) mentioned in this application can be implemented using generalized cosine similarity (GCS) or square generalized cosine similarity (SGCS), etc., and this application does not impose any restrictions. For example, taking SGCS as the feedback precision, the first feedback precision is denoted as... Should Where N is the number of subbands in the system, and N is an integer greater than or equal to 1; i is one of the N subbands, and i is a positive integer less than or equal to N; v i The CSI measured for the terminal device corresponding to the i-th sub-band. This is the reconstruction information output by the first decoder corresponding to the i-th subband.

[0115] S602. The terminal device sends at least one of the following: first feedback information, second feedback information, or third feedback information to the network device based on the first feedback accuracy of the first feedback information. The second feedback information is obtained based on codebook compression CSI, and the third feedback information is obtained based on encoder compression CSI with adjusted compression rate.

[0116] The terminal device obtains CSI feedback information in the following ways: compressing the CSI using an encoder as described in S601 (considered as an encoder with an unadjusted compression rate) to obtain first feedback information; compressing the CSI based on the codebook to obtain second feedback information; adjusting the compression rate of the encoder described in S601, and compressing the CSI using the adjusted encoder to obtain third feedback information. Based on the accuracy of the first feedback, the terminal device selects to send the first feedback information, the second feedback information, or the third feedback information to the network device.

[0117] In this application, adjusting the encoder's compression ratio can be understood as increasing or decreasing the encoder's compression ratio based on an offset value (denoted as Δ compression ratio), or as using a compression ratio different from the current compression ratio, or as increasing or decreasing the number of bits output by the encoder based on an offset value (denoted as Δ bits), or as using a number of bits different from the encoder's current output bit number. The specific values ​​of Δ compression ratio and Δ bits can be adjusted according to the actual application scenario, and this application does not impose specific limitations on them.

[0118] In one possible implementation of S602, after determining the first feedback accuracy of the first feedback information, the terminal device can choose to send the first feedback information to the network device via an AI-based CSI feedback scheme based on the first feedback accuracy; or, it can choose to send the second feedback information to the network device via a codebook-based CSI feedback scheme. Alternatively, the terminal device can choose not to adjust the encoder's compression rate and send the first feedback information to the network device based on the first feedback accuracy; or, it can choose to adjust the encoder's compression rate and send the third feedback information to the network device.

[0119] For example, the terminal device determines whether the first feedback accuracy meets the system requirements, or in other words, whether the first feedback accuracy meets the system's requirement for CSI feedback accuracy. If the first feedback accuracy meets the system requirements, the terminal device executes the following branch one: sending the first feedback information to the network device. If the first feedback accuracy does not meet the system requirements, it executes the following branch two: sending the second feedback information to the network device, or executes the following branch three: sending the third feedback information to the network device.

[0120] Branch 1: If the system requirements are met, the terminal device sends the first feedback information to the network device.

[0121] For example, the system's requirement for CSI feedback accuracy is a first accuracy threshold. This first accuracy threshold can be predefined or it can be sent from the network device to the terminal device. In this application, the specific value of the first accuracy threshold can be set according to the actual communication application scenario, and this application does not impose specific limitations on it. If the first feedback accuracy is greater than or equal to the first accuracy threshold, it is considered that the first feedback accuracy meets the system requirements; if the first feedback accuracy is less than the first accuracy threshold, it is considered that the first feedback accuracy does not meet the system requirements.

[0122] If the first feedback precision is greater than or equal to a first precision threshold, the network device receives the first feedback information and reconstructs it using a second decoder to obtain second reconstructed information (i.e., the CSI reconstructed / decoded by the network device). Further, the network device performs scheduling (such as resource allocation) based on the second reconstructed information.

[0123] Optionally, the terminal device sends a feedback information quality identifier to the network device. The feedback information quality identifier is used to indicate whether the first feedback accuracy meets the system requirements (e.g., whether the first feedback accuracy is greater than or equal to a first accuracy threshold), or the feedback information quality identifier is used to indicate whether the first feedback accuracy does not meet the system requirements (e.g., whether the first feedback accuracy is less than a first accuracy threshold).

[0124] In this optional approach, the terminal device determines whether the first feedback accuracy meets the system requirements and generates a feedback information quality identifier based on the determination result. Further, the terminal device sends the first feedback information and the feedback information quality identifier to the network device. Alternatively, this can be understood as the feedback bitstream from the terminal device to the network device containing the first feedback information, in addition to the bits corresponding to the first feedback information, also containing at least one bit (or status bit) corresponding to the feedback information quality identifier. The feedback bitstream used in this application to feed back the first feedback information can be a physical uplink shared channel (PUSCH) or a physical uplink control channel (PUCCH).

[0125] For example, taking the feedback accuracy as SGCS, let the first feedback accuracy of the first feedback information be ρ. 2 The first precision threshold is t SGCS The PUCCH feedback bitstream used to provide the first feedback information contains not only the bits corresponding to the first feedback information but also the status bits corresponding to the feedback information quality identifier. In ρ 2 Greater than or equal to t SGCS In this case, the status bit of the PUCCH feedback code stream is 0, indicating the accuracy requirement of the first feedback information system. In ρ2 Less than t SGCS In this case, the status bit of the PUCCH feedback code stream is set to 1, indicating that the accuracy of the first feedback does not meet the system requirements.

[0126] Optionally, the terminal device sends a CSI quality report to the network device. This CSI quality report includes information indicating the accuracy of the first feedback and / or channel state parameters. The channel state parameters are parameters indicating channel characteristics, including but not limited to channel capacity, bit error rate (BER) / block error rate (BLER), channel attenuation parameters (e.g., path loss), delay spread parameters, multipath effects (e.g., Doppler correlation parameters), and interference level parameters (e.g., signal-to-noise ratio (SNR), signal-to-interference-plus-noise ratio (SINR)).

[0127] In this optional approach, the terminal device generates a CSI quality report based on the first feedback accuracy and channel state parameters. Further, if the first feedback accuracy does not meet system requirements, the terminal device sends the CSI quality report to the network device; or, regardless of whether the first feedback accuracy meets system requirements, the terminal device sends the CSI quality report to the network device.

[0128] Optionally, the CSI quality report can be transmitted via radio resource control (RRC) signaling. For example, if the initial feedback accuracy does not meet system requirements, the terminal device sends a CSI quality report to the network device via the RRC signaling "AI-PerformanceReport".

[0129] Branch 2: If the system requirements are not met, the terminal device sends a second feedback message to the network device.

[0130] For example, if the accuracy of the first feedback is less than the first accuracy threshold, the terminal device compresses the CSI based on the codebook to obtain the second feedback information and sends this second feedback information to the network device. Further, the network device reconstructs / decodes the second feedback information based on the codebook and performs scheduling (such as resource allocation) based on the reconstructed / restored information.

[0131] In one possible implementation of branch 2, after the terminal device determines that the accuracy of the first feedback does not meet the system requirements, the terminal device actively sends the second feedback information obtained based on codebook compression CSI to the network device.

[0132] In this possible implementation 1, the terminal device can also send identification information to the network device. This identification information indicates that the feedback information (i.e., the second feedback information) is obtained based on codebook compression CSI. Upon receiving this identification information, the network device knows that the second feedback information should be decoded / reconstructed based on the codebook.

[0133] In another possible implementation of branch two, after the terminal device determines that the first feedback accuracy does not meet the system requirements, the terminal device still sends the first feedback information to the network device. After receiving the first feedback information, the network device determines that the first feedback accuracy does not meet the system requirements, and then sends a first indication information to the terminal device, which instructs the terminal device to provide second feedback information.

[0134] Optionally, the first indication information can be transmitted via downlink control information (DCI). This application does not limit the specific format of the DCI; for example, the DCI could be DCI format1_0. Further, the terminal device sends second feedback information obtained based on codebook compressed CSI to the network device based on the first indication information.

[0135] In this possible implementation 2, the network device determines that the first feedback accuracy does not meet the system requirements in ways including but not limited to any of the following:

[0136] Method 1: If the terminal device also sends a feedback information quality identifier to the network device, the network device determines whether the first feedback accuracy meets the system requirements based on this feedback information quality identifier. If the feedback information quality identifier indicates that the first feedback accuracy does not meet the system requirements, the network device determines that the first feedback accuracy does not meet the system requirements. If the feedback information quality identifier indicates that the first feedback accuracy meets the system requirements, the network device determines that the first feedback accuracy meets the system requirements.

[0137] Method 2: When the terminal device also sends a CSI quality report to the network device, the network device determines whether the first feedback accuracy meets the system requirements based on the information in the CSI quality report indicating the first feedback accuracy and / or channel state parameters. For example, the network device determines the first feedback accuracy based on the information in the CSI quality report indicating the first feedback accuracy; if the first feedback accuracy is less than a first accuracy threshold, the network device determines that the first feedback accuracy does not meet the system requirements; if the first feedback accuracy is greater than or equal to the first accuracy threshold, the network device determines that the first feedback accuracy meets the system requirements. Another example is that the network device predicts / calculates the CSI before compression by the terminal device encoder (denoted as the predicted CSI) based on the channel state parameters in the CSI quality report, and inputs the first feedback information into the second decoder to obtain the second reconstruction information; further, it calculates the second feedback accuracy based on the second reconstruction information and the predicted CSI. If the second feedback accuracy meets the system requirements (e.g., the second feedback accuracy is greater than or equal to the first accuracy threshold), the network device determines that the first feedback accuracy meets the system requirements; if the second feedback accuracy does not meet the system requirements, the network device determines that the first feedback accuracy does not meet the system requirements.

[0138] Branch 3: When system requirements are not met, the terminal device sends a third feedback message to the network device.

[0139] For example, if the first feedback accuracy is less than the first accuracy threshold, the terminal device adjusts the encoder's compression rate and compresses the CSI based on the adjusted encoder to obtain third feedback information. The terminal device then sends this third feedback information to the network device. Further, the network device reconstructs this third feedback information using a second decoder to obtain third reconstructed information, and performs scheduling (such as resource allocation) based on the third reconstructed information.

[0140] In one optional implementation of branch three, after the terminal device determines that the first feedback accuracy does not meet the system requirements, the terminal device actively adjusts the compression ratio of the encoder and uses the encoder with the adjusted compression ratio to re-compress the CSI to obtain the third feedback information, and sends the third feedback information to the network device.

[0141] For example, if the first feedback accuracy does not meet system requirements, the terminal device lowers the encoder's compression rate based on the Δ compression rate, or increases the number of bits output by the encoder based on the Δ bit count. This Δ compression rate and Δ bit count are related to the difference between the first feedback accuracy and the first accuracy threshold. The larger the difference between the first feedback accuracy and the first accuracy threshold, the larger the absolute value of the Δ compression rate and the larger the absolute value of the Δ bit count. The terminal device then recompresses the CSI using the adjusted encoder to obtain third feedback information, which has a greater number of bits than the first feedback information.

[0142] In another possible implementation 2 of branch three, after the terminal device determines that the first feedback accuracy does not meet the system requirements, the terminal device still sends the first feedback information to the network device. After receiving the first feedback information, the network device determines that the first feedback accuracy does not meet the system requirements, and then sends a second indication information to the terminal device. This second indication information is used to instruct the terminal device to provide third feedback information, or to adjust the encoder's compression rate, or to adjust the encoder's output bit count. Optionally, this second indication information can be transmitted via DCI. This application does not limit the specific format of the DCI used; for example, the DCI can be DCI format 1_0. Optionally, the second indication information includes information for indicating Δ compression rate or Δ bit count. Further, the terminal device adjusts the encoder's compression rate based on the second indication information, and obtains the third feedback information by compressing the CSI based on the encoder after adjusting the compression rate, and sends the third feedback information to the network device.

[0143] In this possible implementation 2, the method by which the network device determines that the first feedback accuracy does not meet the system requirements can be found in the relevant description in implementation 2 of branch two, and will not be repeated here.

[0144] In another possible implementation of S602, after determining the first feedback accuracy of the first feedback information, the terminal device can dynamically select a CSI feedback scheme based on the first feedback accuracy, i.e., select an AI-based CSI feedback scheme or a codebook-based CSI feedback scheme. If the codebook-based CSI feedback scheme is selected, the terminal device sends the second feedback information to the network device. If the AI-based CSI feedback scheme is selected, the terminal device can also choose whether to adjust the encoder's compression ratio. If the encoder's compression ratio is adjusted, the terminal device sends the third feedback information to the network device. If the encoder's compression ratio is not adjusted, the terminal device sends the first feedback information obtained in S601 to the network device.

[0145] For example, the second precision threshold is less than the second precision threshold. The terminal device dynamically selects either the AI-based CSI feedback scheme or the codebook-based CSI feedback scheme based on the relationship between the first feedback precision and the second precision threshold. When the first feedback precision is less than the second precision threshold, the terminal device performs CSI feedback based on the codebook-based CSI feedback scheme, i.e., sends the second feedback information to the network device according to the execution process of the aforementioned branch two. When the first feedback precision is greater than or equal to the second precision threshold, the terminal device performs CSI feedback based on the AI-based CSI feedback scheme, and dynamically selects whether to adjust the encoder's compression ratio based on the relationship between the first feedback precision and the first precision threshold. When the first feedback precision is greater than or equal to the second precision threshold but less than the first precision threshold, the terminal device adjusts the encoder's compression ratio, i.e., sends the third feedback information to the network device according to the execution process of the aforementioned branch three; when the first feedback precision is greater than or equal to the first precision threshold, the terminal device does not adjust the encoder's compression, i.e., sends the first feedback information to the network device according to the execution process of the aforementioned branch one.

[0146] In conclusion, Figure 6 The provided method trains a first decoder that can simulate the second decoder on the network device side using knowledge distillation technology. This first decoder is then deployed on the terminal device side. After obtaining the first feedback information based on encoder-compressed CSI, and before sending this first feedback information to the network device, the terminal device can reconstruct the first feedback information using the first decoder and monitor the feedback accuracy (i.e., the first feedback accuracy) based on this first reconstructed information. Furthermore, the terminal device dynamically selects a CSI feedback scheme based on the first feedback accuracy, which helps improve the reliability and stability of CSI feedback. For example, the terminal device can prioritize a feedback scheme that reduces CSI feedback overhead (i.e., an AI-based feedback scheme) to send the first feedback information obtained from encoder-compressed CSI to the network device. If the feedback accuracy of the first feedback information does not meet system requirements, the terminal device selects a feedback scheme with more stable CSI feedback performance (i.e., a codebook-based feedback scheme or a feedback scheme that adjusts the encoder compression rate) to send the second or third feedback information to the network device. This helps reduce CSI feedback overhead while also improving the performance stability of the CSI feedback scheme.

[0147] It should be understood that, unless otherwise specified or logically contradictory, in Figure 6 The various steps and possible implementations of the provided embodiments can be combined to form a new embodiment. For example, please refer to [link to relevant documentation]. Figure 7 As shown:

[0148] S701. The terminal device obtains first feedback information, which is obtained based on encoder compression CSI.

[0149] For details regarding the first feedback information and the specific implementation of S701, please refer to the description of the specific implementation of S601 mentioned above, which will not be repeated here.

[0150] S702, The terminal device obtains the first feedback accuracy.

[0151] For details regarding the first feedback accuracy and the specific implementation of S702, please refer to the relevant description of the terminal device obtaining the first feedback accuracy in the aforementioned S601, which will not be repeated here.

[0152] S703, The terminal device generates a feedback information quality identifier, which is used to indicate whether the first feedback accuracy meets the system requirements.

[0153] For details regarding the feedback information quality identifier and the specific implementation of S703, please refer to the relevant description of the terminal device generating the feedback information quality identifier in the aforementioned S601, which will not be repeated here.

[0154] Furthermore, if the first feedback accuracy does not meet the system requirements, the terminal device sends a CSI quality report to the network device. This CSI quality report includes information indicating the first feedback accuracy and channel state parameters. A detailed description of this CSI quality report can be found in the relevant description in the aforementioned S601, and will not be repeated here.

[0155] S704 (Branch 1): When the first feedback accuracy is greater than or equal to the first accuracy threshold, the terminal device sends the first feedback information and the feedback information quality identifier to the network device.

[0156] The terminal device sends a PUCCH feedback stream to the network device. The PUCCH feedback stream includes the bits corresponding to the first feedback information and the status bits corresponding to the feedback information quality identifier.

[0157] For details on the specific implementation of S704, please refer to the description of the specific implementation of branch one of S602 above, which will not be repeated here.

[0158] S705 (Branch 2): If the first feedback accuracy is less than the first accuracy threshold, the terminal device sends the first feedback information, the feedback information quality identifier, and the CSI quality report to the network device.

[0159] The terminal device sends a PUCCH feedback stream to the network device. This PUCCH feedback stream includes the bits corresponding to the first feedback information and the status bits corresponding to the feedback information quality identifier. Furthermore, the RRC signaling "AI-PerformanceReport" sends a CSI quality report to the network device.

[0160] S706 (Branch 2): The network device sends a first instruction message to the terminal device, which is used to instruct the terminal device to provide a second feedback message.

[0161] After receiving the first feedback information, the network device determines that the accuracy of the first feedback does not meet the system requirements. The network device then sends DCI format 1_0 to the terminal device. This DCI format 1_0 includes indication information for instructing the terminal device to provide the second feedback information.

[0162] S707 (Branch 2): The terminal device sends a second feedback message to the network device, which is obtained based on codebook compression CSI.

[0163] The terminal device obtains the second feedback information by compressing the CSI based on the codebook according to the first instruction information, and sends the second feedback information to the network device.

[0164] The specific implementation methods of S705 to S707 can be found in the description of the specific implementation method of implementation method 2 in the aforementioned branch S602, and will not be repeated here.

[0165] For example, see also Figure 8 As shown:

[0166] S801, The terminal device obtains first feedback information, which is obtained based on encoder compression CSI.

[0167] S802, The terminal device obtains the first feedback accuracy.

[0168] S803, The terminal device generates a feedback information quality identifier, which is used to indicate whether the first feedback accuracy meets the system requirements.

[0169] S804 (Branch 1): When the first feedback accuracy is greater than or equal to the first accuracy threshold, the terminal device sends the first feedback information and the feedback information quality identifier to the network device.

[0170] S805 (Branch 3): If the first feedback accuracy is less than the first accuracy threshold, the terminal device sends the first feedback information, the feedback information quality identifier, and the CSI quality report to the network device.

[0171] The specific implementations of S801 to S805 can be referred to the description of the specific implementations of S701 to S705, and will not be repeated here.

[0172] S806 (Branch 3): The network device sends a second instruction message to the terminal device, which is used to instruct the terminal device to adjust the compression rate of the encoder.

[0173] After receiving the first feedback information, the network device determines that the accuracy of the first feedback does not meet the system requirements. The network device then sends DCI format 1_0 to the terminal device. The DCI format 1_0 includes indication information for instructing the terminal device to adjust the compression rate of the encoder, such as at least one of the following: an identifier for indicating the feedback of the third feedback information, Δ compression rate, or Δ number of bits, etc.

[0174] S807 (Branch 3): The terminal device sends a third feedback message to the network device, which is obtained based on the encoder compression CSI after adjusting the compression ratio.

[0175] The terminal device adjusts the encoder's compression rate based on the second instruction information, and compresses the CSI based on the adjusted encoder to obtain third feedback information. The terminal device then sends this third feedback information to the network device.

[0176] The specific implementation methods of S806 to S807 can be found in the description of the specific implementation method of implementation method 2 in the aforementioned branch 3 of S602, and will not be repeated here.

[0177] It should be understood that this application is only used as an example of the CSI feedback scenario for illustration. The solution provided in this application can also be applied to other scenarios where complex models need to be simulated on devices with limited computing power, such as image processing or speech recognition in edge computing.

[0178] It should be understood that Figures 6-8 The flowcharts or scene diagrams shown are for illustrative purposes only and are not intended to limit the embodiments of this application to the examples illustrated. In fact, those skilled in the art can interpret the embodiments based on... Figures 6-8 The possible examples can be freely combined as long as there is no logical conflict, resulting in more implementation methods.

[0179] The above text combined Figures 6-8 This document describes in detail the communication method provided in the embodiments of this application. The following will combine... Figures 9 to 10 The device embodiments of this application are described in detail below. It should be understood that the communication device of this application embodiment can execute the various communication methods of the foregoing embodiments of this application, that is, the specific working processes of the various products below can be referred to the corresponding processes in the foregoing method embodiments.

[0180] In the embodiments described above, the terminal device may execute some or all of the steps in each embodiment; the network device may execute some or all of the steps in each embodiment. These steps or operations are merely examples, and the embodiments of this application may also perform other operations or variations thereof. Furthermore, the steps may be executed in different orders as presented in the embodiments, and it is not necessary to execute all the operations in the embodiments of this application. Moreover, the sequence number of each step does not imply the order of execution; the execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0181] Figure 9 This is a schematic diagram of the communication device provided in an embodiment of this application. Figure 9 As shown, the communication device 90 may include a communication module 920. The communication module 920 can implement corresponding communication functions, which can be internal communication functions of the communication device 90 or communication functions between the communication device 90 and other devices. Optionally, the communication module 920 may also be referred to as a communication interface or transceiver module. Optionally, the communication device 90 also includes a processing module 910. The processing module 910 can implement corresponding processing functions.

[0182] Optionally, the communication device 90 further includes a storage module, which can be used to store instructions and / or data; the processing module 910 can read the instructions and / or data in the storage module so that the communication device 90 can implement the aforementioned method embodiments.

[0183] In one possible design, the communication device 90 may correspond to the terminal device in the above method embodiments, or a component (such as a circuit, chip, or chip system) configured in the terminal device. The communication device 90 can be used to perform the steps or processes performed by the terminal device in any of the above method embodiments.

[0184] For example, the processing module 910 is used to acquire first feedback information, which is obtained by compressing CSI based on an encoder; the communication module 920 is used to send one of the first feedback information, a second feedback information, or a third feedback information to the network device based on a first feedback accuracy of the first feedback information; wherein, the second feedback information is obtained by compressing the CSI based on a codebook, the third feedback information is obtained by compressing the CSI based on an encoder with an adjusted compression rate, the first feedback accuracy is obtained based on first reconstruction information, the first reconstruction information is information reconstructed by a first decoder of the first feedback information, the first decoder is a decoder deployed on a terminal device, the second decoder is a decoder deployed on the network device, the first decoder is a student network model in knowledge distillation technology, and the second decoder is a teacher network model in knowledge distillation technology.

[0185] In one possible implementation, if the first feedback accuracy of the first feedback information is greater than or equal to a first accuracy threshold, the communication module 920 is used to send the first feedback information to the network device; if the first feedback accuracy is less than the first accuracy threshold, the communication module 920 is used to send a second feedback information or a third feedback information to the network device.

[0186] In one possible implementation, the communication module 920 is used to send the first feedback information to the network device; the communication module 920 is used to receive instruction information from the network device, the instruction information being used to instruct the terminal device to provide the second feedback information.

[0187] In one possible implementation, the communication module 920 is used to send a feedback information quality identifier to the network device, the feedback information quality identifier being used to indicate whether the first feedback accuracy is less than the first accuracy threshold.

[0188] In one possible implementation, the communication module 920 is used to send a CSI quality report to the network device, the CSI quality report including information and / or channel state parameters indicating the accuracy of the first feedback.

[0189] The above are merely examples; for detailed steps or procedures, please refer to the descriptions in the foregoing embodiments.

[0190] In one possible design, the communication device 90 may correspond to the network device in the above method embodiments, or to a component (such as a circuit, chip, or chip system) configured in the network device. The communication device 90 can be used to perform the steps or processes performed by the network device in any of the above method embodiments.

[0191] For example, the processing module 910 is used to process data received by the transceiver module or to process data to be sent by the transceiver module;

[0192] The communication module 920 is used to receive first feedback information from the terminal device, which is obtained based on the encoder compressing the channel state information (CSI). If the first feedback accuracy of the first feedback information is less than a first accuracy threshold, the communication module 920 is used to receive second or third feedback information from the terminal device. The second feedback information is obtained based on the codebook compressing the CSI, and the third feedback information is obtained based on the encoder compressing the CSI with an adjusted compression rate. The first feedback accuracy is obtained based on first reconstruction information, which is information reconstructed by a first decoder. The first decoder is a decoder deployed on the terminal device, and the second decoder is a decoder deployed on the network device. The first decoder is a student network model in the knowledge distillation technique, and the second decoder is a teacher network model in the knowledge distillation technique.

[0193] In one possible implementation, the communication module 920 is used to send a first instruction information to the terminal device, the first instruction information being used to instruct the terminal device to provide the second feedback information, the second instruction information being used to instruct the terminal device to adjust the compression rate of the encoder, or the second instruction information being used to instruct the terminal device to provide the third feedback information.

[0194] In one possible implementation, the communication module 920 is further configured to receive a feedback information quality identifier from the terminal device, the feedback information quality identifier being used to indicate whether the first feedback accuracy is less than the first accuracy threshold.

[0195] In one possible implementation, the communication module 920 is used to receive a CSI quality report from the terminal device, the CSI quality report including information and / or channel state parameters indicating the first feedback accuracy.

[0196] The above are merely examples; for detailed steps or procedures, please refer to the descriptions in the foregoing embodiments.

[0197] Figure 10 This is a schematic diagram of another communication device structure provided in the embodiments of this application. The communication device 100 may be a chip, chip system, or processor, etc., used by a terminal device or network device to implement the above methods. The communication device 100 can be used to implement the methods described in the above method embodiments; for details, please refer to the descriptions in the above method embodiments.

[0198] like Figure 10 As shown, the communication device 100 may include one or more processors 1010, which may also be referred to as processing units or processing modules, and can implement certain control functions. The processor 1010 may be a general-purpose processor or a dedicated processor, such as a baseband processor or a central processing unit. The baseband processor can be used to process communication protocols and communication data, while the central processing unit can be used to control the communication device 100 (e.g., a base station, baseband chip, user, user chip), execute software programs, and process data from the software programs.

[0199] In an alternative design, the processor 1010 may also store instructions and / or data that can be executed by the processor 1010 to cause the communication device 100 to perform the methods described in the above method embodiments.

[0200] In another alternative design, the communication device 100 may include a communication interface 1020 for implementing receiving and transmitting functions. For example, the communication interface 1020 may be a transceiver circuit, interface, interface circuit, or transceiver. The transceiver circuit, interface, interface circuit, or transceiver for implementing receiving and transmitting functions may be separate or integrated. The aforementioned transceiver circuit, interface, interface circuit, or transceiver may be used for reading and writing code / data, or it may be used for transmitting or relaying signals.

[0201] Optionally, the communication device 100 may include one or more memories 1030, which may store instructions that can be executed on the processor 1010, causing the communication device 100 to perform the methods described in the above method embodiments. Optionally, the memories 1030 may also store data. Optionally, the processor 1010 may also store instructions and / or data. The processor 1010 and the memories 1030 may be provided separately or integrated together.

[0202] It should be understood that, in one possible design, the steps in the method embodiments provided in this application can be implemented by integrated logic circuits in the processor's hardware or by instructions in software form. The steps of the methods disclosed in the embodiments of this application can be directly implemented by a hardware processor, or implemented 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 not provided here.

[0203] In one implementation, the communication device 100 may correspond to the terminal device in the above method embodiments and may be used to execute the various steps and / or processes executed by the terminal device in the above method embodiments. The processor 1010 may be used to execute instructions stored in the memory 1030, and when the processor 1010 executes the instructions stored in the memory, the processor 1010 is used to execute the various steps and / or processes of the above method embodiments corresponding to the terminal device.

[0204] In another implementation, the communication device 100 may correspond to the network device in the above method embodiments and may be used to execute the various steps and / or processes executed by the network device in the above method embodiments. The processor 1010 may be used to execute instructions stored in the memory 1030, and when the processor 1010 executes the instructions stored in the memory, the processor 1010 is used to execute the various steps and / or processes of the above method embodiments corresponding to the network device.

[0205] Figure 11 This is a hardware system for a communication device applicable to this application, provided in an embodiment of this application. Figure 11 The communication device 1100 can be a mobile phone, smart screen, tablet computer, wearable communication device, vehicle communication device, augmented reality (AR) device, virtual reality (VR) device, laptop computer, ultra-mobile personal computer (UMPC), netbook, personal digital assistant (PDA), etc. This application embodiment does not limit the specific type of the communication device 1100.

[0206] The communication device 1100 may include a processor 110, an external memory interface 120, an internal memory 121, a universal serial bus (USB) interface 130, a charging management module 140, a power management module 141, a battery 142, an antenna 1, an antenna 2, a mobile communication module 150, a wireless communication module 160, an audio module 170, a speaker 170A, a receiver 170B, a microphone 170C, a headphone jack 170D, a sensor module 180, buttons 190, a motor 191, an indicator 192, a camera 193, a display screen 194, and a subscriber identification module (SIM) card interface 195, etc. The sensor module 180 may include a pressure sensor 180A, a gyroscope sensor 180B, a barometric pressure sensor 180C, a magnetic sensor 180D, an accelerometer sensor 180E, a distance sensor 180F, a proximity sensor 180G, a fingerprint sensor 180H, a temperature sensor 180J, a touch sensor 180K, an ambient light sensor 180L, a bone conduction sensor 180M, etc.

[0207] Processor 110 may include one or more processing units, such as application processors (APs), modems, graphics processing units (GPUs), image signal processors (ISPs), controllers, video codecs, digital signal processors (DSPs), baseband processors, and / or neural network processing units (NPUs). These different processing units may be independent devices or integrated into one or more processors.

[0208] In this embodiment, the modem can receive active commands sent by the SIM card and send response information for the active commands to the SIM card. The modem can also intercept active commands from the SIM card when it receives unexpected active commands triggered by the application layer (i.e., it will not send active commands from the SIM card to the STK application).

[0209] The controller can serve as the central nervous system and command center of the communication device 1100. The controller can generate operation control signals based on the instruction opcode and timing signals to control the fetching and execution of instructions.

[0210] The processor 110 may also include a memory for storing instructions and data. In some embodiments, the memory in the processor 110 is a cache memory. This memory can store instructions or data that the processor 110 has just used or that are used repeatedly. If the processor 110 needs to use the instruction or data again, it can directly retrieve it from the memory. This avoids repeated accesses, reduces the waiting time of the processor 110, and thus improves the efficiency of the system. For example, when the STK application receives a browser launch command sent by the SIM card, it stores the URL information contained in the browser launch command. After receiving a confirmation command to launch, it launches the browser and displays the corresponding webpage based on the stored URL information.

[0211] The modem processor may include a modulator and a demodulator. The modulator modulates the low-frequency baseband signal to be transmitted into a mid-to-high frequency signal. The demodulator demodulates the received electromagnetic wave signal into a low-frequency baseband signal. The demodulator then transmits the demodulated low-frequency baseband signal to the baseband processor for processing. After processing by the baseband processor, the low-frequency baseband signal is transmitted to the application processor. The application processor outputs sound signals through audio devices (not limited to speaker 170A, receiver 170B, etc.) or displays images or videos through the display screen 194.

[0212] In some embodiments, the modem processor (modem) is responsible for converting digital signals into analog signals (modulation) for transmission over a wireless channel, and also for converting received analog signals back into digital signals (demodulation). During the interaction between the STK application and the SIM card, digital commands and data are directly transmitted, but they must be processed by the modem before they can be transmitted over the wireless channel. The modem supports multiple network standards (such as 2G, 3G, 4G, 5G, etc.); during the interaction between the STK application and the SIM card, the modem can handle commands and data transmission under different network standards to ensure the compatibility and stability of the STK application.

[0213] In some embodiments, the modem processor may be a separate device. In other embodiments, the modem processor may be independent of the processor 110 and may be housed in the same device as the mobile communication module 150 or other functional modules.

[0214] In some embodiments, the processor 110 may include one or more interfaces. Interfaces may include a mobile industry processor interface (MIPI), a general-purpose input / output (GPIO) interface, a subscriber identity module (SIM) interface, and / or a universal serial bus (USB) interface, etc.

[0215] The external memory interface 120 can be used to connect an external memory card, such as a Micro SD card, to expand the storage capacity of the communication device 1100. The external memory card communicates with the processor 110 through the external memory interface 120.

[0216] The internal memory 121 can be used to store computer executable program code, which includes instructions. The processor 110 executes various functional applications and data processing of the communication device 1100 by running the instructions stored in the internal memory 121.

[0217] In addition, the internal memory 121 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, universal flash storage (UFS), etc.

[0218] It is understood that the interface connection relationships between the modules illustrated in the embodiments of this application are merely illustrative and do not constitute a structural limitation on the communication device 1100. In other embodiments of this application, the communication device 1100 may also employ different interface connection methods or combinations of multiple interface connection methods as described in the above embodiments.

[0219] Touch sensor 180K, also known as a "touch panel," can be located on display screen 194. The touch sensor 180K and display screen 194 together form a touchscreen, also known as a "touchscreen." Touch sensor 180K detects touch operations applied to or near it. The touch sensor can transmit the detected touch operation to the application processor to determine the type of touch event. Visual output related to the touch operation can be provided through display screen 194. In some embodiments, touch sensor 180K may also be located on the surface of communication device 1100, in a different position than display screen 194. The aforementioned pressure sensor 180A and / or touch sensor 180K can be used to trigger the SIM card in the communication device to report active commands, requesting the communication device to display a menu, send a text message, make a phone call, or perform any of these events.

[0220] The display screen 194 is used to display images, videos, etc. In some embodiments, the communication device 1100 may include one or N display screens 194, where N is a positive integer greater than 1. The display screen 194 can be used to display menus, SMS sending and receiving interfaces, telephone dialing interfaces, and browser interfaces, etc., based on the response information of the above-mentioned active commands.

[0221] The SIM card interface 195 is used to connect a SIM card. The physical SIM card can be inserted into or removed from the SIM card interface 195 to establish contact with and disconnect the communication device 1100. The communication device 1100 can support one or N SIM card interfaces, where N is a positive integer greater than 1.

[0222] In some embodiments, the communication device 1100 may employ an eSIM card. For example, if the software interface of one card slot in the communication device 1100 supports both physical SIM cards and eSIM cards, the SIM card interface 195 corresponding to that card slot is connected to the processor 110. The communication device 1100 may support one or more eSIM cards. In this embodiment, the SIM card can be either a physical SIM card or an eSIM card. In this embodiment, the SIM card can request the communication device 1100 to execute corresponding events by reporting active commands.

[0223] The SIM card interface 195 supports Nano SIM cards, Micro SIM cards, and other SIM cards. Multiple cards can be inserted into the same SIM card interface 195 simultaneously. These multiple cards can be of the same or different types. The SIM card interface 195 is also compatible with different types of SIM cards. The SIM card interface 195 is also compatible with external memory cards. The communication device 1100 interacts with the network through the SIM card to realize functions such as voice calls and data communication. In some embodiments, the communication device 1100 uses an eSIM, i.e., an embedded SIM card. The eSIM card can be embedded in the communication device 1100 and cannot be separated from the communication device 1100.

[0224] In other embodiments, the SIM card is also configured with various application services. The communication device interacts with the SIM card by installing an STK application (SIM card toolkit) to realize different business services, such as querying SIM card related information (such as message records).

[0225] It should be noted that, Figure 11 The structure shown does not constitute a specific limitation on the communication device 1100. In other embodiments of this application, the communication device 1100 may include a... Figure 11 The components shown may include more or fewer components, or the communication device 1100 may include... Figure 11 The components shown may be a combination of certain components, or the communication device 1100 may include... Figure 11 Sub-components of some of the components shown. Figure 11 The components shown can be implemented in hardware, software, or a combination of both.

[0226] The communication methods provided in the above embodiments can all be implemented in the communication device 1100 having the above hardware structure.

[0227] 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.

[0228] It is understood that the memory in the embodiments of this application can be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous linked dynamic random access memory (SLDRAM), and direct rambus RAM (DR RAM). It should be noted that the memory used in the systems and methods described herein is intended to include, but is not limited to, these and any other suitable types of memory.

[0229] According to the method provided in the embodiments of this application, this application also provides a chip system, which includes one or more processors for calling and executing instructions stored in memory, thereby causing the method described in the embodiments of this application to be executed. The chip system may be composed of chips or may include chips and other discrete devices.

[0230] The chip system may include input circuits or interfaces for transmitting information or data, and output circuits or interfaces for receiving information or data.

[0231] According to the method provided in the embodiments of this application, this application also provides a communication system, which includes the aforementioned network device and terminal device.

[0232] According to the method provided in the embodiments of this application, this application also provides a computer program product, which includes: computer program code, which, when run on a computer, causes the computer to execute the various steps or processes executed by the network device or terminal device in any of the foregoing method embodiments.

[0233] According to the method provided in the embodiments of this application, this application also provides a computer-readable storage medium storing program code, which, when run on a computer, causes the computer to execute the various steps or processes executed by the network device or terminal device in any of the foregoing method embodiments.

[0234] The computer-readable storage medium may be the aforementioned volatile memory or non-volatile memory, or it may include both volatile memory and non-volatile memory.

[0235] In the embodiments of this application, the terms and English abbreviations are exemplary examples given for ease of description and should not be construed as limiting the application in any way. This application does not preclude the possibility of defining other terms that can achieve the same or similar functions in existing or future agreements.

[0236] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions. When these computer instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated.

[0237] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0238] It should be understood that in the various embodiments of this application, the sequence number of each process does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0239] In summary, the above description is merely a preferred embodiment of the technical solution of this application and is not intended to limit the scope of protection of this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A communication method, characterized in that, The method includes: Obtain first feedback information, which is obtained based on encoder compressed channel state information (CSI). Based on the first feedback accuracy of the first feedback information, send one of the first feedback information, the second feedback information, or the third feedback information to the network device; Wherein, the second feedback information is obtained by compressing the CSI based on the codebook, the third feedback information is obtained by compressing the CSI based on the encoder after adjusting the compression ratio, the first feedback accuracy is obtained based on the first reconstruction information, the first reconstruction information is the information of the first decoder reconstructing the first feedback information, the first decoder is a decoder deployed on the terminal device, the second decoder is a decoder deployed on the network device, the first decoder is the student network model in the knowledge distillation technology, and the second decoder is the teacher network model in the knowledge distillation technology.

2. The method according to claim 1, characterized in that, Sending one of the first feedback information, the second feedback information, or the third feedback information to the network device based on the first feedback accuracy of the first feedback information includes: If the first feedback accuracy is greater than or equal to the first accuracy threshold, the first feedback information is sent to the network device. If the accuracy of the first feedback is less than the first accuracy threshold, the second feedback information or the third feedback information is sent to the network device.

3. The method according to claim 2, characterized in that, Before sending the second feedback information or the third feedback information to the network device, the following steps are included: Send the first feedback information to the network device; The terminal device receives a first instruction or a second instruction from the network device; the first instruction is used to instruct the terminal device to provide the second feedback information; the second instruction is used to instruct the terminal device to adjust the compression rate of the encoder, or the second instruction is used to instruct the terminal device to provide the third feedback information.

4. The method according to claim 2 or 3, characterized in that, The method further includes: A feedback information quality identifier is sent to the network device, the feedback information quality identifier being used to indicate whether the first feedback accuracy is less than the first accuracy threshold.

5. The method according to any one of claims 1-4, characterized in that, The method further includes: Send a CSI quality report to the network device, the CSI quality report including information and / or channel state parameters indicating the accuracy of the first feedback.

6. A communication method, characterized in that, The method includes: Receive first feedback information from the terminal device, the first feedback information being obtained based on encoder compression channel state information (CSI); If the first feedback accuracy of the first feedback information is less than the first accuracy threshold, the second or third feedback information from the terminal device is received. Wherein, the second feedback information is obtained by compressing the CSI based on the codebook, the third feedback information is obtained by compressing the CSI based on the encoder after adjusting the compression ratio, the first feedback accuracy is obtained based on the first reconstruction information, the first reconstruction information is the information of the first decoder reconstructing the first feedback information, the first decoder is a decoder deployed on the terminal device, the second decoder is a decoder deployed on the network device, the first decoder is the student network model in the knowledge distillation technology, and the second decoder is the teacher network model in the knowledge distillation technology.

7. The method according to claim 6, characterized in that, Before receiving the second feedback information or the third feedback information from the terminal device, the method further includes: Send a first instruction message or a second instruction message to the terminal device; the first instruction message is used to instruct the terminal device to provide the second feedback message; the second instruction message is used to instruct the terminal device to adjust the compression rate of the encoder, or the second instruction message is used to instruct the terminal device to provide the third feedback message.

8. The method according to claim 6 or 7, characterized in that, The method further includes: The system receives a feedback information quality identifier from the terminal device, which indicates whether the first feedback accuracy is less than the first accuracy threshold.

9. The method according to any one of claims 6-8, characterized in that, The method further includes: Receive a CSI quality report from the terminal device, the CSI quality report including information and / or channel state parameters for indicating the accuracy of the first feedback.

10. A communication device, characterized in that, It includes a module for performing the method as described in any one of claims 1-5, or includes a module for performing the method as described in any one of claims 6-9.

11. A communication device, characterized in that, It includes a processor and a storage medium, the storage medium storing instructions that, when executed by the processor, cause the communication device to perform the method according to any one of claims 1-5, or to perform the method according to any one of claims 6-9.

12. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store a computer program, which, when executed, performs the method as described in any one of claims 1-5, or the method as described in any one of claims 6-9.

13. A computer program product, characterized in that, When the computer program product is executed, the method described in any one of claims 1-5 is executed, or the method described in any one of claims 6-9 is executed.

Images