CSI feedback method, transmitting end device and receiving end device

Abstract

Provided in an embodiment of the present application are a CSI feedback method, a transmitting device and a receiving device, a CSI sample set is clustered to obtain a clustering center, and the clustering center obtained by clustering is used as a CSI feedback codebook, and CSI feedback implemented based on the CSI feedback codebook can have the performance advantage of scene pertinence based on data driving, thereby improving the CSI feedback performance. The CSI feedback method comprises the following steps: a transmitting end device encodes target CSI information according to a CSI feedback codebook to obtain a target feedback bit stream; the sending end equipment sends the target feedback bit stream; wherein the CSI feedback codebook is K clustering centers obtained by clustering CSI samples in a first data set, the first data set comprises S CSI samples, both K and S are positive integers, K = 2B < S, and B is a CSI feedback bit number.

Description

Method for CSI feedback, transmitting device, and receiving device Technical Field

[0001] Embodiments of the present application relate to the field of communications, and more particularly, to a method for CSI feedback, a transmitting device, and a receiving device. Background Art

[0002] In a New Radio (NR) system, a terminal device needs to feedback Channel State Information (CSI) to a network device for the network device to determine a precoding matrix for downlink transmission. Specifically, CSI feedback mainly uses a CSI feedback codebook-based scheme to achieve channel feature extraction and feedback. That is, after channel estimation is performed at the transmitting end and the corresponding precoding matrix is obtained according to the channel estimation result, the precoding matrix that best matches the precoding matrix to be feedback is selected from a pre-set CSI feedback codebook according to a certain optimization criterion, and information such as the index of the matrix is feedback to the receiving end through the feedback link of the air interface for the receiving end to implement precoding.

[0003] However, the above CSI feedback method cannot meet the requirements for CSI feedback in the evolution of communication systems.

[0004] Summary of the Invention

[0005] Embodiments of the present application provide a method for CSI feedback, a transmitting device, and a receiving device, which cluster a CSI sample set to obtain cluster centers, and use the cluster centers obtained by clustering as a CSI feedback codebook. The CSI feedback implemented based on this CSI feedback codebook can have the performance advantage of data-driven scenario targeting, thereby improving the CSI feedback performance.

[0006] In a first aspect, a method for CSI feedback is provided, the method comprising:

[0007] The transmitting device encodes target CSI information according to a CSI feedback codebook to obtain a target feedback bitstream;

[0008] The transmitting device transmits the target feedback bitstream;

[0009] where the CSI feedback codebook is K cluster centers obtained by clustering CSI samples in a first data set, the first data set includes S CSI samples, both K and S are positive integers, K = 2 B <S, B is the number of CSI feedback bits.

[0010] In a second aspect, a method for CSI feedback is provided, the method comprising:

[0011] The receiving device receives the target feedback bitstream;

[0012] The receiving device decodes the target feedback bitstream according to the CSI feedback codebook to obtain target CSI samples;

[0013] Among them, the CSI feedback codebook is K clustering centers obtained by clustering CSI samples in the first dataset. The first dataset includes S CSI samples. Both K and S are positive integers, and K = 2 B <S, B is the number of CSI feedback bits.

[0014] In a third aspect, a transmitting device is provided for performing the method in the first aspect above.

[0015] Specifically, the transmitting device includes functional modules for performing the method in the first aspect above.

[0016] In a fourth aspect, a receiving device is provided for performing the method in the second aspect above.

[0017] Specifically, the receiving device includes functional modules for performing the method in the second aspect above.

[0018] In a fifth aspect, a transmitting device is provided, including a processor and a memory; the memory is used to store a computer program, and the processor is used to call and run the computer program stored in the memory, so that the transmitting device performs the method in the first aspect above.

[0019] In a sixth aspect, a receiving device is provided, including a processor and a memory; the memory is used to store a computer program, and the processor is used to call and run the computer program stored in the memory, so that the receiving device performs the method in the second aspect above.

[0020] In a seventh aspect, a device is provided for implementing the method in any one of the first aspect to the second aspect above.

[0021] Specifically, the device includes: a processor, which is used to call and run a computer program from the memory, so that the device equipped with the device performs the method in any one of the first aspect to the second aspect above.

[0022] In an eighth aspect, a computer-readable storage medium is provided for storing a computer program, and the computer program enables a computer to perform the method in any one of the first aspect to the second aspect above.

[0023] In a ninth aspect, a computer program product is provided, including computer program instructions, and the computer program instructions enable a computer to perform the method in any one of the first aspect to the second aspect above.

[0024] In a tenth aspect, a computer program is provided, which, when executed on a computer, enables the computer to execute the method in any one of the first to second aspects above.

[0025] Through the above technical solution, the CSI samples in the first data set are clustered to obtain K cluster centers, and the K cluster centers obtained by clustering are used as the CSI feedback codebook corresponding to the first data set. CSI feedback implemented based on this CSI feedback codebook can have the advantage of data-driven scenario-specific performance, thereby improving CSI feedback performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] FIG1 is a schematic diagram of a communication system architecture applied in an embodiment of the present application.

[0027] FIG2 is a schematic diagram of a neuron structure.

[0028] FIG3 is a schematic diagram of a neural network provided by the present application.

[0029] FIG4 is a schematic diagram of a convolutional neural network provided in this application.

[0030] FIG5 is a schematic diagram of an LSTM unit provided in this application.

[0031] FIG6 is a schematic diagram of a channel information feedback provided by the present application.

[0032] FIG7 is a schematic diagram of another channel information feedback provided in this application.

[0033] FIG8 is a schematic flowchart of a CSI feedback method provided according to an embodiment of the present application.

[0034] FIG9 is a schematic block diagram of a transmitting device provided according to an embodiment of the present application.

[0035] FIG10 is a schematic block diagram of a receiving device provided according to an embodiment of the present application.

[0036] FIG11 is a schematic block diagram of a communication device provided according to an embodiment of the present application.

[0037] FIG12 is a schematic block diagram of a device provided according to an embodiment of the present application.

[0038] FIG13 is a schematic block diagram of a communication system provided according to an embodiment of the present application. DETAILED DESCRIPTION

[0039] The following will describe the technical solutions in the embodiments of this application in conjunction with the drawings in the embodiments of this application. Obviously, the embodiments described are part of the embodiments of this application, not all of the embodiments. With respect to the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0040] The technical solutions of the embodiments of the present application can be applied to various communication systems, such as: Global System of Mobile communication (GSM) system, Code Division Multiple Access (CDMA) system, Wideband Code Division Multiple Access (WCDMA) system, General Packet Radio Service (GPRS), Long Term Evolution (LTE) system, Advanced Long Term Evolution (LTE-A) system, New Radio (NR) system, NR system evolution system, LTE-based access to unlicensed spectrum (LTE-U) system on unlicensed spectrum, NR-based access to unlicensed spectrum (NR-U) system on unlicensed spectrum, Non-Terrestrial Networks (NTN) system, Universal Mobile Telecommunication System (UMTS), Wireless Local Area Networks (WLAN), Internet of Things (IoT), Wireless Fidelity (WFI) system. Fidelity, WiFi), fifth-generation communication (5th-Generation, 5G) system or other communication systems, etc.

[0041] Generally speaking, traditional communication systems support a limited number of connections and are easy to implement. However, with the development of communication technology, mobile communication systems will not only support traditional communications, but will also support, for example, device-to-device (D2D) communication, machine-to-machine (M2M) communication, machine-type communication (MTC), vehicle-to-vehicle (V2V) communication, or vehicle-to-everything (V2X) communication, etc. The embodiments of the present application can also be applied to these communication systems.

[0042] In some embodiments, the communication system in the embodiments of the present application can be applied to a carrier aggregation (CA) scenario, a dual connectivity (DC) scenario, an independent (SA) networking scenario, or a non-standalone (NSA) networking scenario.

[0043] In some embodiments, the communication system in the embodiments of the present application can be applied to an unlicensed spectrum, where the unlicensed spectrum can also be considered as a shared spectrum; or, the communication system in the embodiments of the present application can also be applied to an authorized spectrum, where the authorized spectrum can also be considered as an unshared spectrum.

[0044] In some embodiments, the communication system in the embodiments of the present application can be applied to the FR1 frequency band (corresponding to the frequency band range of 410MHz to 7.125GHz), can also be applied to the FR2 frequency band (corresponding to the frequency band range of 24.25GHz to 52.6GHz), and can also be applied to new frequency bands such as high-frequency bands corresponding to the frequency band range of 52.6GHz to 71GHz or the frequency band range of 71GHz to 114.25GHz.

[0045] The embodiments of the present application describe various embodiments in conjunction with network devices and terminal devices, wherein the terminal device may also be referred to as user equipment (UE), access terminal, user unit, user station, mobile station, mobile station, remote station, remote terminal, mobile device, user terminal, terminal, wireless communication device, user agent or user device, etc.

[0046] The terminal device can be a station (ST) in a WLAN, a cellular phone, a cordless phone, a Session Initiation Protocol (SIP) phone, a Wireless Local Loop (WLL) station, a Personal Digital Assistant (PDA) device, a handheld device with wireless communication capabilities, a computing device or other processing device connected to a wireless modem, a vehicle-mounted device, a wearable device, a terminal device in a next-generation communication system such as an NR network, or a terminal device in a future evolved Public Land Mobile Network (PLMN) network, etc.

[0047] In an embodiment of the present application, the terminal device can be deployed on land, including indoors or outdoors, handheld, wearable or vehicle-mounted; it can also be deployed on the water surface (such as ships, etc.); it can also be deployed in the air (such as airplanes, balloons and satellites, etc.).

[0048] In an embodiment of the present application, the terminal device may be a mobile phone, a tablet computer, a computer with wireless transceiver function, a virtual reality (VR) terminal device, an augmented reality (AR) terminal device, a wireless terminal device in industrial control, a wireless terminal device in self-driving, a wireless terminal device in remote medical, a wireless terminal device in a smart grid, a wireless terminal device in transportation safety, a wireless terminal device in a smart city or a wireless terminal device in a smart home, an in-vehicle communication device, a wireless communication chip / application specific integrated circuit (ASIC) / system on chip (SoC), etc.

[0049] As an example and not a limitation, in the embodiment of the present application, the terminal device may also be a wearable device. Wearable devices may also be called wearable smart devices, which are a general term for wearable devices that are intelligently designed and developed using wearable technology for daily wear, such as glasses, gloves, watches, clothing, and shoes. A wearable device is a portable device that is worn directly on the body or integrated into the user's clothes or accessories. Wearable devices are not only hardware devices, but also achieve powerful functions through software support, data interaction, and cloud interaction. Broadly speaking, wearable smart devices include those that are fully functional, large in size, and can achieve complete or partial functions without relying on smartphones, such as smart watches or smart glasses, as well as those that only focus on a certain type of application function and need to be used in conjunction with other devices such as smartphones, such as various smart bracelets and smart jewelry for vital sign monitoring.

[0050] In an embodiment of the present application, the network device may be a device for communicating with a mobile device. The network device may be an access point (AP) in WLAN, a base station (BTS) in GSM or CDMA, a base station (NodeB, NB) in WCDMA, an evolved base station (eNB or eNodeB) in LTE, or a relay station or access point, or a vehicle-mounted device, a wearable device, and a network device or base station (gNB) in an NR network, or a network device in a future evolved PLMN network or a network device in an NTN network, etc.

[0051] As an example and not a limitation, in an embodiment of the present application, the network device may have a mobile feature, for example, the network device may be a mobile device. In some embodiments, the network device may be a satellite or a balloon station. For example, the satellite may be a low earth orbit (LEO) satellite, a medium earth orbit (MEO) satellite, a geostationary earth orbit (GEO) satellite, a high elliptical orbit (HEO) satellite, etc. In some embodiments, the network device may also be a base station set up in a location such as land or water.

[0052] In an embodiment of the present application, the network device can provide services for a cell, and the terminal device communicates with the network device through the transmission resources used by the cell (for example, frequency domain resources, or spectrum resources). The cell can be a cell corresponding to the network device (for example, a base station). The cell can belong to a macro base station or a base station corresponding to a small cell. The small cells here may include: metro cells, micro cells, pico cells, femto cells, etc. These small cells have the characteristics of small coverage and low transmission power, and are suitable for providing high-speed data transmission services.

[0053] For example, a communication system 100 used in an embodiment of the present application is shown in FIG1 . The communication system 100 may include a network device 110, which may be a device that communicates with a terminal device 120 (or a communication terminal or terminal). The network device 110 may provide communication coverage for a specific geographic area and may communicate with terminal devices within the coverage area.

[0054] FIG1 exemplarily shows a network device and two terminal devices. In some embodiments, the communication system 100 may include multiple network devices and each network device may include another number of terminal devices within its coverage area, which is not limited in the embodiments of the present application.

[0055] In some embodiments, the communication system 100 may further include other network entities such as a network controller and a mobility management entity, which is not limited in the embodiments of the present application.

[0056] It should be understood that in the embodiments of the present application, a device having a communication function in a network / system may be referred to as a communication device. Taking the communication system 100 shown in FIG1 as an example, the communication device may include a network device 110 and a terminal device 120 having a communication function. The network device 110 and the terminal device 120 may be the specific devices described above and will not be described in detail here. The communication device may also include other devices in the communication system 100, such as a network controller, a mobility management entity, and other network entities, which are not limited in the embodiments of the present application.

[0057] It should be understood that the terms "system" and "network" are often used interchangeably herein. The term "and / or" is simply a description of an association between related objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A exists alone, A and B exist simultaneously, or B exists alone. Furthermore, the character " / " generally indicates that the related objects are in an "or" relationship.

[0058] It should be understood that this document relates to a first communication device and a second communication device. The first communication device may be a terminal device, such as a mobile phone, a machine facility, a customer premises equipment (CPE), industrial equipment, a vehicle, etc.; the second communication device may be a peer communication device of the first communication device, such as a network device, a mobile phone, an industrial device, a vehicle, etc. This document describes a specific example in which the first communication device is a terminal device and the second communication device is a network device.

[0059] The terms used in the embodiments of this application are intended only to explain the specific embodiments of this application and are not intended to limit this application. The terms "first," "second," "third," and "fourth," etc. in the specification and claims of this application and the accompanying drawings are used to distinguish different objects, not to describe a specific order. In addition, the terms "including" and "having," and any variations thereof, are intended to cover non-exclusive inclusions.

[0060] It should be understood that the "indication" mentioned in the embodiments of this application can be a direct indication, an indirect indication, or an indication of an association. For example, "A indicates B" can mean that A directly indicates B, for example, B can be obtained through A; it can also mean that A indirectly indicates B, for example, A indicates C, and B can be obtained through C; it can also mean that there is an association between A and B.

[0061] In the description of the embodiments of the present application, the term "corresponding" may indicate a direct or indirect correspondence between the two, or an association relationship between the two, or a relationship between indication and being indicated, configuration and being configured, etc.

[0062] In the embodiments of the present application, "pre-definition" or "pre-configuration" may be implemented by pre-storing corresponding codes, tables, or other methods that can be used to indicate relevant information in a device (e.g., a terminal device and a network device). The present application does not limit the specific implementation method. For example, pre-definition may refer to information defined in a protocol.

[0063] In the embodiments of the present application, the “protocol” may refer to a standard protocol in the communications field, for example, it may include an LTE protocol, an NR protocol, and related protocols used in future communication systems, and the present application does not limit this.

[0064] To facilitate understanding of the technical solutions of the embodiments of the present application, the following describes the CSI feedback based on the traditional codebook related to the present application.

[0065] For 5G NR systems, current CSI feedback designs primarily utilize codebook-based solutions to extract and feedback channel features. Specifically, after performing channel estimation at the transmitter and obtaining the corresponding precoding matrix based on the channel estimation results, the transmitter selects the precoding matrix that best matches the precoding matrix to be fed back from a pre-set codebook according to a certain optimization criterion. Information such as the matrix index is then fed back to the receiver via an air interface feedback link for precoding. Specifically, codebook designs can be categorized as Type 1 (Type I), Type 2 (Type II), and enhanced Type II (eType II). Intuitively, codebook solutions consider using one of the many precoding matrices in the codebook to encode a class of similar precoding matrices to be fed back. Traditional codebook solutions are designed based on spatially uniformly distributed Discrete Fourier Transform (DFT) vectors. While they offer high generalization, they lack the high performance advantages required for specific scenarios.

[0066] To facilitate a better understanding of the embodiments of the present application, the neural network related to the present application is described.

[0067] A neural network is a computational model consisting of multiple interconnected neuron nodes, where the connections between nodes represent weighted values ​​from input signals to output signals, called weights. Each node performs a weighted summation (SUM) on different input signals and outputs them through a specific activation function (f). Figure 2 is a schematic diagram of a neuron structure, where a1, a2, …, an represent input signals, w1, w2, …, wn represent weights, f represents the activation function, and t represents the output.

[0068] A simple neural network, shown in Figure 3, consists of an input layer, hidden layers, and an output layer. By using different connections, weights, and activation functions among multiple neurons, different outputs can be generated, thereby fitting the mapping relationship from input to output. Each node in the previous level is connected to all nodes in the next level. This neural network is a fully connected neural network, also known as a deep neural network (DNN).

[0069] Deep learning utilizes deep neural networks with multiple hidden layers, significantly improving the network's ability to learn features and fitting complex, nonlinear mappings from input to output. Consequently, it has found widespread application in speech and image processing. In addition to deep neural networks, deep learning also includes other commonly used basic structures such as convolutional neural networks (CNNs) and recurrent neural networks (RNNs) for different tasks.

[0070] The basic structure of a convolutional neural network includes an input layer, multiple convolutional layers, multiple pooling layers, a fully connected layer, and an output layer, as shown in Figure 4. Each neuron in the convolution kernel of the convolutional layer is locally connected to its input, and the introduction of the pooling layer extracts the local maximum or average features of a certain layer, effectively reducing the network parameters and mining local features, enabling the convolutional neural network to converge quickly and achieve excellent performance.

[0071] RNNs are neural networks that model sequential data and have achieved remarkable success in natural language processing applications such as machine translation and speech recognition. Specifically, the network memorizes information from past moments and uses it in the calculation of current outputs. This means that nodes in the hidden layers are no longer disconnected but connected, and the input to a hidden layer includes not only the input layer but also the output of the previous hidden layer. Common RNN structures include long short-term memory (LSTM) and gated recurrent unit (GRU). Figure 5 shows a basic LSTM cell structure, which can include a tanh activation function. Unlike RNNs, which only consider the most recent state, the LSTM cell state determines which states should be retained and which should be forgotten, addressing the shortcomings of traditional RNNs in long-term memory.

[0072] To facilitate a better understanding of the embodiments of the present application, the CSI feedback method based on artificial intelligence (AI) related to the present application is described.

[0073] Given the tremendous success of AI technology in areas such as computer vision and natural language processing, the communications field has begun exploring the use of deep learning, such as deep learning, to address technical challenges that are difficult to address with traditional communications methods. The neural network architecture commonly used in deep learning is nonlinear and data-driven. It can extract features from actual channel matrix data and, at the base station, restore the channel matrix information compressed and fed back by the terminal as closely as possible. This ensures the restoration of channel information while also enabling the reduction of CSI feedback overhead on the terminal side.

[0074] Deep learning-based CSI feedback treats channel information as an image to be compressed, uses a deep learning autoencoder to compress and feed back the channel information, and reconstructs the compressed channel image at the transmitter, which can preserve the channel information to a greater extent, as shown in Figure 6.

[0075] Figure 7 shows a typical channel information feedback system. The entire feedback system consists of an encoder and a decoder, deployed at the transmitter and receiver, respectively. After obtaining channel information through channel estimation, the transmitter compresses and encodes the channel information matrix using the encoder's neural network. The compressed bitstream is then fed back to the receiver via an air interface feedback link. The receiver uses the decoder to recover the channel information from the feedback bitstream to obtain complete feedback channel information. The encoder shown in Figure 7 employs a stack of multiple fully connected layers, while the decoder employs convolutional layers and a residual architecture. While maintaining the same encoding and decoding framework, the network model structures within the encoder and decoder can be flexibly designed.

[0076] To facilitate understanding of the technical solutions of the embodiments of the present application, the problems solved by the present application are described below.

[0077] First, the traditional codebook scheme is designed based on DFT vectors uniformly distributed in the spatial domain, and therefore lacks the high-performance advantage of targeted scenarios (such as macrocell scenarios, indoor scenarios, high-speed mobile scenarios, etc.). Secondly, the current AI-based channel information feedback considers using the encoder of the AI ​​autoencoder to compress the channel information at the transmitting end and using the decoder of the AI ​​autoencoder to reconstruct the channel information at the receiving end. It uses the nonlinear fitting ability of the neural network to compress and feedback the channel information, which can greatly improve the compression efficiency and feedback accuracy. However, due to the "black box" (i.e., unexplainable) nature of the nonlinear neural network in the end-to-end AI solution, it cannot be deterministically formulated and written into the standard like Type I, Type II, and eType II in the CSI feedback enhancement standardization process. Therefore, the existing AI solution is likely to be classified as a type of implementation solution during the standardization discussion and will not have a standardization impact.

[0078] Based on the above technical problems, this application proposes a CSI feedback solution, clustering the CSI sample set to obtain cluster centers, and using the cluster centers obtained by clustering as the CSI feedback codebook. The CSI feedback implemented based on this CSI feedback codebook can have targeted performance advantages based on data-driven scenarios (such as macro cell scenarios, indoor scenarios, high-speed mobile scenarios, etc.), thereby improving the CSI feedback performance. In addition, the technical solution of this application is a data-driven solution, which not only takes into account the performance advantages of data-driven, but also makes up for the shortcoming of "unexplainability" of data-driven solutions, increases the feasibility of data-driven solutions, and thus makes it possible to standardize data-driven CSI feedback solutions.

[0079] To facilitate the understanding of the technical solutions of the embodiments of the present application, the technical solutions of the present application are described in detail below through specific embodiments. The following related technologies can be arbitrarily combined with the technical solutions of the embodiments of the present application as optional solutions, and they all fall within the protection scope of the embodiments of the present application. The embodiments of the present application include at least some of the following contents.

[0080] FIG. 8 is a schematic flowchart of a CSI feedback method 200 according to an embodiment of the present application. As shown in FIG. 8, the CSI feedback method 200 may include at least some of the following contents:

[0081] S210, the transmitting device encodes the target CSI information according to the CSI feedback codebook to obtain a target feedback bitstream; wherein, the CSI feedback codebook is K clustering centers obtained by clustering CSI samples in the first dataset, the first dataset includes S CSI samples, both K and S are positive integers, and K = 2 B <S, B is the number of CSI feedback bits;

[0082] S220, the transmitting device sends the target feedback bitstream;

[0083] S230, the receiving device receives the target feedback bitstream;

[0084] S240, the receiving device decodes the target feedback bitstream according to the CSI feedback codebook to obtain the target CSI sample.

[0085] In the embodiments of the present application, the CSI feedback codebook may be generated by a terminal device, or may be generated by a network device, or may be generated by a third-party device or server, or may be generated by a core network element. The present application does not limit this.

[0086] In the embodiments of the present application, the CSI samples in the first dataset can be clustered to obtain K clustering centers, and the K clustering centers obtained by clustering are used as the CSI feedback codebook corresponding to the first dataset. The CSI feedback implemented based on such a CSI feedback codebook can have performance advantages targeted at data-driven scenarios (such as macro cell scenarios, indoor scenarios, high-speed mobile scenarios, etc.), thereby improving the CSI feedback performance.

[0087] In some embodiments, the CSI samples in the first dataset may belong to a certain specific scenario, such as a macro cell scenario, an indoor scenario, a high-speed mobile scenario, etc. Of course, it may also be other scenarios, and the embodiments of the present application do not limit this.

[0088] In some embodiments, the transmitting device is a terminal device and the receiving device is a network device; or, the transmitting device is a network device and the receiving device is a terminal device.

[0089] In some embodiments, the CSI samples in the first data set are a full channel information matrix in the time domain, or the CSI samples in the first data set are a full channel information matrix in the frequency domain, or the CSI samples in the first data set are a matrix consisting of eigenvectors obtained by performing eigenvalue decomposition on multiple subbands of the full channel information matrix.

[0090] In some embodiments, as Example 1, the CSI samples in the first data set are the full channel information matrix in the time domain, or the CSI samples in the first data set are the full channel information matrix in the frequency domain. Optionally, the first data set T = {H1, ..., H S}, where H i (i=1,...,S) is the full channel information matrix in the time domain or the full channel information matrix in the frequency domain.

[0091] In some embodiments, as Example 2, the CSI samples in the first data set are matrices consisting of eigenvectors obtained by performing eigenvalue decomposition on multiple subbands of the full channel information matrix. Optionally, the first data set T = {W1, ..., W S}, where W i =[w i,1 ,...,w i,Nsb ]∈C Nsb×Nt (i=1,...,S) is the matrix composed of eigenvectors obtained by eigenvalue decomposition of multiple subbands of the full channel information matrix, C represents a complex number set, N sb Indicates the number of subbands, N t Indicates the number of transmit antenna ports.

[0092] In some embodiments, the K cluster centers are obtained by clustering through the following steps:

[0093] performing a normalization operation on each CSI sample in the first data set;

[0094] Randomly select K CSI samples from the first data set as initial cluster centers;

[0095] Repeat the following steps 1 and 2 Q times to obtain the K cluster centers, where Q is a positive integer:

[0096] Step 1: Determine the class to which each CSI sample in the first data set belongs;

[0097] Step 2: Determine the cluster center of each class.

[0098] In some embodiments, the value of Q may be determined based on implementation, or the value of Q may be an empirical value.

[0099] For example, the value of Q is 20, 30, 50, 80, 100, etc., which is not limited in the embodiments of the present application.

[0100] In some embodiments, the K cluster centers are obtained by clustering through the following steps:

[0101] performing a normalization operation on each CSI sample in the first data set;

[0102] Randomly select K CSI samples from the first data set as initial cluster centers;

[0103] The following steps 1 and 2 are performed cyclically, and when the quantization errors of all CSI samples in the first data set are less than a first preset threshold, the K cluster centers are obtained:

[0104] Step 1: Determine the class to which each CSI sample in the first data set belongs;

[0105] Step 2: Determine the cluster center of each class.

[0106] In some embodiments, the first preset threshold is agreed upon by a protocol, or the first preset threshold is configured by a network device, or the first preset threshold is determined based on implementation.

[0107] In some embodiments, the K cluster centers are obtained by clustering through the following steps:

[0108] performing a normalization operation on each CSI sample in the first data set;

[0109] Randomly select K CSI samples from the first data set as initial cluster centers;

[0110] The following steps 1 and 2 are performed cyclically, and when the similarity between all CSI samples in the first data set and the corresponding cluster centers is greater than a second preset threshold, the K cluster centers are obtained:

[0111] Step 1: Determine the class to which each CSI sample in the first data set belongs;

[0112] Step 2: Determine the cluster center of each class.

[0113] In some embodiments, the second preset threshold is agreed upon by a protocol, or the second preset threshold is configured by a network device, or the second preset threshold is determined based on implementation.

[0114] In some embodiments, in Example 1, the initial cluster centers may be U1, ..., U K .

[0115] In some embodiments, in Example 1, the CSI samples H in the first data set may be calculated according to the following formula 1: i (i=1,...,S) performs normalization operation:

[0116] ||H1|| F =||H2|| F =...=||H S || F =1 Formula 1

[0117] Among them, ||·|| F represents the Frobenius norm.

[0118] That is, in Example 1, a normalization operation may be performed on each CSI sample in the first data set based on Formula 1.

[0119] In some embodiments, in Example 1, the CSI samples H in the first data set may be determined according to the following formula 2: i The class to which (i=1,...,S) belongs:

[0120] c i =argmin j＝1,...,K ||H i -U j || F Formula 2

[0121] Among them, c i Denotes the CSI sample H i Class, U j represents the jth cluster center, argmin j＝1,...,K ||H i -U j || F Indicates ||H i -U j || F The value of j when taking the minimum value, ||·|| F represents the Frobenius norm.

[0122] That is, in Example 1, the class to which each CSI sample in the first data set belongs can be determined based on Formula 2.

[0123] In some embodiments, in Example 1, the class c can be determined according to the following formula 3: i The cluster center U j , where j = 1, ..., K:

[0124] U j =(∑ S i＝1f{c i = j}H i ) / (∑ S i＝1 f{c i = j}) Equation 3

[0125] where c i represents the class to which the CSI sample H i belongs. When c i = j holds, f{c i = j}= 1. When c i = j does not hold, f{c i = j}= 0. ∑ represents the summation operation.

[0126] That is, in Example 1, the cluster center of each class can be determined based on Equation 3.

[0127] In some embodiments, the CSI samples in the first dataset are the full-channel information matrices in the time domain, or the CSI samples in the first dataset are the full-channel information matrices in the frequency domain. That is, in Example 1, the quantization error of the CSI samples H i (i = 1,..., S) is less than the first preset threshold:

[0128] ∑ S i＝1 ||H i - U c i || F < X Equation 4

[0129] where c i represents the class to which the CSI sample H i belongs, U c i represents the cluster center corresponding to class c i , X represents the first preset threshold, and ||·|| F represents the Frobenius norm.

[0130] In some embodiments, in Example 1, the transmitting device encodes the target CSI information based on the following Equation 5 to generate the target feedback bitstream:

[0131] b1 = g(argmin j＝1,...,K ||H - U j || F ) Equation 5

[0132] where b1 represents the target feedback bitstream, H represents the target CSI information, and U jrepresents the jth cluster center among the K cluster centers, g(·) represents the mapping function between the K cluster centers and the B-bit feedback bit stream, g(·)∈{0,1} B , argmin j＝1,...,K ||HU j || F Indicates || HU j || F The value of j when taking the minimum value, ||·|| F represents the Frobenius norm.

[0133] That is, in Example 1, after obtaining the CSI feedback codebook corresponding to the first data set, the transmitting device can encode the target CSI information based on Formula 5 to generate a target feedback bit stream b1. The transmitting device then sends the target feedback bit stream b1 to the receiving device.

[0134] In some embodiments, in Example 1, the receiving device decodes the target feedback bit stream based on the following formula 6 to obtain target CSI information:

[0135] H=g -1 (b1) Formula 6

[0136] Where b1=g(argmin j＝1,...,K ||HU j || F ), b1 represents the target feedback bit stream, H represents the target CSI information, U j represents the jth cluster center among the K cluster centers, g(·) represents the mapping function between the K cluster centers and the B-bit feedback bit stream, g(·)∈{0,1} B , argmin j＝1,...,K ||HU j || F Indicates || HU j || F The value of j when taking the minimum value, ||·|| F represents the Frobenius norm.

[0137] That is, in Example 1, after obtaining the CSI feedback codebook corresponding to the first data set, the receiving device can decode the target feedback bit stream b1 based on Formula 6 to obtain the target CSI information H.

[0138] In some embodiments, in Example 2, the initial cluster center may be U j =[u j,1 ,...,u j,Nsb ]∈C Nsb×Nt, where j = 1,...,K.

[0139] In some embodiments, in Example 2, the CSI samples W in the first data set are calculated according to the following formula 7: i (i=1,...,S) performs normalization operation, where W i =[w i,1 ,...,w i,Nsb ] T ∈C Nsb×Nt , N sb Indicates the number of subbands, N t represents the number of transmit antenna ports, and C represents a complex set:

[0140] ||w i,1 || F =||w i,2 || F =...=||w i,Nsb || F Formula 7

[0141] Among them, ||·|| F represents the Frobenius norm.

[0142] It should be noted that the Frobenius norm in Formula 7 can also be a two-norm.

[0143] That is, in Example 2, a normalization operation may be performed on each CSI sample in the first data set based on Formula 7.

[0144] In some embodiments, in Example 2, the CSI samples W in the first data set may be determined according to the following formula 7: i The class to which (i=1,...,S) belongs:

[0145] c i =argmax j＝1,...,K ∑ Nsb v＝1 ||w i,v H u j,v || F Formula 8

[0146] Among them, c i represents the CSI sample W i The class to which it belongs, w i,v H represents the CSI sample W i The conjugate transpose of the corresponding vector, u j,v Represents the cluster center U j The vector corresponding to the neutron band v, U j=[u j,1 ,...,u j,Nsb ]∈C Nsb×Nt , argmax j＝1,...,K ∑ Nsb v＝1 ||w i,v H u j,v || F Represents ∑ Nsb v＝1 ||w i,v H u j,v || F The value of j when taking the maximum value, ||·|| F represents the Frobenius norm, and ∑ represents the summation operation.

[0147] That is, in Example 2, the class to which each CSI sample in the first data set belongs can be determined based on Formula 8.

[0148] It should be noted that the Frobenius norm in Formula 8 can also be a two-norm.

[0149] In some embodiments, in Example 2, the class c can be determined according to the following formula 9: i The cluster center U j (j=1,...,K):

[0150] U j =(∑ S i＝1 f{c i =j}W i ) / (∑ S i＝1 f{c i =j}) Formula 9

[0151] Among them, c i represents the CSI sample W i (i=1,...,S) belongs to the class, when c i =j holds true when f{c i =j}=1, when c i =j does not hold true when f{c i =j}=0, ∑ represents the summation operation.

[0152] That is, in Example 2, the cluster center of each class can be determined based on Formula 9.

[0153] In some embodiments, the CSI samples in the first data set are matrices consisting of eigenvectors obtained by performing eigenvalue decomposition on multiple subbands of the full channel information matrix. That is, in Example 2, the CSI samples W in the first data set can be determined according to the following formula 10: i The similarity with the corresponding cluster center is greater than the second preset threshold, where W i =[w i,1 ,...,w i,Nsb ] T ∈C Nsb×Nt , i=1,...,S,N sb Indicates the number of subbands, N t represents the number of transmit antenna ports, and C represents a complex set:

[0154]

[0155] Where v represents the subband, v = 1,...,N sb , w i,v H represents the CSI sample W i The conjugate transpose of the corresponding vector, c i represents the CSI sample W i The class to which it belongs, Represents class c i The corresponding cluster center U c i The vector corresponding to the neutron band v, Y represents the second preset threshold, ∑ represents a summation operation, ||·|| F represents the Frobenius norm.

[0156] In some embodiments, in Example 2, the transmitting device encodes the target CSI information based on the following formula 11 to generate a target feedback bit stream:

[0157] b2=g(argmax j＝1,...,K ∑ Nsb v＝1 ||w v H u j,v || F ) Formula 11

[0158] Where b2 represents the target feedback bit stream, W represents the target CSI information, W=[w1,...,w Nsb ] T ∈C Nsb×Nt , w v H Represents the conjugate transpose of the vector corresponding to the target CSI information W, u j,vRepresents the cluster center U among the K cluster centers j The vector corresponding to the neutron band v, U j =[u j,1 ,...,u j,Nsb ]∈C Nsb×Nt , g(·) represents the mapping function between K cluster centers and B-bit feedback bit stream, g(·)∈{0,1} B , argmax j＝1,...,K ∑ Nsb v＝1 ||w v H u j,v || F Represents ∑ Nsb v＝1 ||w v H u j,v || F The value of j when taking the maximum value, ||·|| F represents the Frobenius norm, and ∑ represents the summation operation.

[0159] That is, in Example 2, after obtaining the CSI feedback codebook corresponding to the first data set, the transmitting device can encode the target CSI information W based on Formula 11 to generate a target feedback bit stream b2. The transmitting device then sends the target feedback bit stream b2 to the receiving device.

[0160] In some embodiments, in Example 2, the receiving device decodes the target feedback bit stream based on the following formula 12 to obtain target CSI information:

[0161] W = g -1 (b2) Formula 12

[0162] Where b2=g(argmax j＝1,...,K ∑ Nsb v＝1 ||w v H u j,v || F ), W represents the target CSI information, W=[w1,...,w Nsb ] T ∈C Nsb× Nt , b2 represents the target feedback bit stream, w v H Represents the conjugate transpose of the vector corresponding to the target CSI information W, u j,v Represents the cluster center U among the K cluster centers j The vector corresponding to the neutron band v, U j=[u j,1 ,...,u j,Nsb ]∈C Nsb×Nt , g(·) represents the mapping function between K cluster centers and B-bit feedback bit stream, g(·)∈{0,1} B , argmax j＝1,...,K ∑ Nsb v＝1 ||w v H u j,v || F Represents ∑ Nsb v＝1 ||w v H u j,v || F The value of j when taking the maximum value, ||·|| F represents the Frobenius norm, and ∑ represents the summation operation.

[0163] That is, in Example 2, after obtaining the CSI feedback codebook corresponding to the first data set, the receiving device can decode the target feedback bit stream b2 based on Formula 12 to obtain the target CSI information W.

[0164] It should be noted that a variety of simple modifications can be made to the formulas involved in the embodiments of the present application, and the formulas after the simple modifications also fall within the scope of protection of the present application.

[0165] For example, the Frobenius norm in Formula 6 to Formula 10 can be replaced by the binary norm.

[0166] Therefore, in the embodiments of the present application, the algorithm flow of this solution is concise and clear. The most important protection point is that the data set is first clustered, and then the cluster centers obtained by clustering are regarded as codebooks for bitstream indication encoding. The beneficial effects of this process include feedback performance gains compared to end-to-end training solutions at low feedback overhead, and the standardization potential brought by the interpretability of the entire process. That is, it takes into account the performance advantages of data-driven solutions while also making up for the shortcomings of data-driven solutions that have always been "unexplainable."

[0167] It should be noted that the encoder and decoder of the AI-based CSI feedback method are both implemented using neural networks and are not interpretable. From a standardization perspective, only the input and output interfaces and the corresponding interaction processes can be protected and standardized. On the other hand, interpretable and formulatable codebook construction methods such as TypeI and eTypeII can be directly standardized, but lack the high-performance advantages of data-driven solutions. The embodiments of this application consider using the method of machine learning clustering to construct the CSI feedback codebook, which not only retains the high-performance advantages of data-driven, but also the construction process of the CSI feedback codebook can be analogous to that of TypeI and eTypeII codebooks, that is, it can be given interpretably and formulaically, and has certain potential for standardization.

[0168] The technical solution of this application is described in detail below through Embodiment 1 and Embodiment 2.

[0169] Embodiment 1 mainly relates to the construction process of the CSI feedback codebook for the full channel information, as well as the encoding and decoding processes.

[0170] Data preparation:

[0171] For a specific scenario (such as a macro cell scenario, an indoor scenario, a high-speed moving scenario, etc.), data-driven CSI feedback codebook design is carried out, and a first data set T is constructed in this scenario. The first data set T contains S CSI samples, and the first data set T = {H1,..., H S}, where H i (i = 1,..., S) is the full channel information matrix in the time domain or the full channel information matrix in the frequency domain. Each CSI sample in the first data set T has been normalized, that is, ||H1|| F = ||H2|| F =... = ||H S || F = 1, where ||·|| F represents the Frobenius norm. Among them, the CSI feedback bit number is B, and the CSI feedback codebook can be constructed through the following clustering algorithm.

[0172] The codebook construction algorithm process can include the following S11 to S15:

[0173] S11. Randomly select K (K = 2 B <S) CSI samples from the first data set T as the initial clustering centers U1,..., U K .

[0174] S12. For each CSI sample H i (i = 1,..., S) in the first data set T, calculate the class it should belong to, that is

[0175] c i =argmin j＝1,...,K ||H i -U j || F

[0176] Among them, c i Denotes the CSI sample H i Class, U j represents the jth cluster center, argmin j＝1,...,K ||H i -U j || F Indicates ||H i -U j || F The value of j when taking the minimum value, ||·|| F represents the Frobenius norm.

[0177] S13. For each class, calculate the cluster center of the class, that is,

[0178] U j =(∑ S i＝1 f{c i =j}H i ) / (∑ S i＝1 f{c i =j})

[0179] Among them, c i Denotes the CSI sample H i Belongs to the class, when c i =j holds true when f{c i =j}=1, when c i =j does not hold true when f{c i =j}=0, ∑ represents the summation operation.

[0180] S14. Loop through S12 and S13 until a termination condition is reached. The termination condition can be one of the following:

[0181] Termination condition 1: reaching the preset number of iterations Q;

[0182] Termination condition 2: the quantization error of all CSI samples in the first data set T is less than the first preset threshold X, that is,

[0183] ∑ S i＝1 ||H i -U c i || F <X

[0184] Among them, c i Denotes the CSI sample H i Class, U c i Represents class c i The corresponding cluster center, X represents the first preset threshold, ||·|| F represents the Frobenius norm.

[0185] S15. The cluster centers U1,...,U K As the CSI feedback codebook under the first data set T, and the CSI feedback codebook is indicated by B bits, that is, a one-to-one mapping b from the cluster center to the bit stream b is established. j =g(·)∈{0,1} B .

[0186] The CSI encoding and decoding process may include the following S16 to S17:

[0187] S16. Given the CSI sample H to be encoded (also called CSI information H), the transmitter encodes it according to the following formula based on the constructed CSI feedback codebook to obtain the bit stream b1 to be fed back, namely:

[0188] b1=g(argmin j＝1,...,K ||HU j || F )

[0189] Among them, U j represents the jth cluster center among the K cluster centers, g(·) represents the mapping function between the K cluster centers and the B-bit feedback bit stream, g(·)∈{0,1} B , argmin j＝1,...,K ||HU j || F Indicates || HU j || F The value of j when taking the minimum value, ||·|| F represents the Frobenius norm.

[0190] S17. After receiving the bit stream b1, the receiving end decodes it according to the mapping function g(·), that is, H = g -1 (b1), from now on, CSI feedback based on full channel information is completed.

[0191] Embodiment 2 mainly involves a CSI feedback codebook construction process, as well as encoding and decoding processes, of eigenvectors obtained by performing eigenvalue decomposition on multiple subbands of the full channel information matrix.

[0192] Data preparation:

[0193] Design a data-driven CSI feedback codebook for a specific scenario (such as a macro cell scenario, an indoor scenario, a high-speed mobile scenario, etc.). Construct a first data set T in this scenario. The first data set T contains S CSI samples, and the first data set T = {W1,..., W S}, where W i＝ [w i,1 ,..., w i,Nsb ∈ C Nsb×Nt (i = 1,..., S) is a matrix composed of eigenvectors obtained by performing eigenvalue decomposition on multiple subbands of the full channel information matrix. C represents the set of complex numbers, N sb represents the number of subbands, and N t represents the number of transmit antenna ports. Among them, the CSI feedback bit number is B, and the CSI feedback codebook can be constructed through the following clustering algorithm.

[0194] The codebook construction algorithm process can include the following S21 to S25:

[0195] S21. Randomly select K (K = 2 B < S) CSI samples from the first data set T as the initial clustering centers U j = [u j,1 ,..., u j,Nsb ∈ C Nsb×Nt , where j = 1,..., K.

[0196] B S22. For each CSI sample W i (i = 1,..., S) in the first data set T, calculate the class it should belong to, that is

[0197] c i = argmax j＝1,...,K ∑ Nsb v＝1 ||w i,v H u j,v || F

[0198] where c i represents the class to which the CSI sample W i belongs, w i,v H represents the conjugate transpose of the vector corresponding to the CSI sample W i , u j,v represents the vector corresponding to the subband v in the clustering center U j , and U j = [u j,1 ,..., u j,Nsb ∈ CNsb×Nt , argmax j＝1,...,K ∑ Nsb v＝1 ||w i,v H u j,v || F Represents ∑ Nsb v＝1 ||w i,v H u j,v || F The value of j when taking the maximum value, ||·|| F represents the Frobenius norm, and ∑ represents the summation operation.

[0199] S23. For each class, calculate the cluster center of the class, that is,

[0200] U j =(∑ S i＝1 f{c i =j}W i ) / (∑ S i＝1 f{c i =j})

[0201] Among them, c i represents the CSI sample W i (i=1,...,S) belongs to the class, when c i =j holds true when f{c i =j}=1, when c i =j does not hold true when f{c i =j}=0, ∑ represents the summation operation.

[0202] S24. Loop through S22 and S23 until a termination condition is reached. The termination condition can be one of the following:

[0203] Termination condition 1: reaching the preset number of iterations Q;

[0204] Termination condition 2: the similarity between all CSI samples in the first data set T and the corresponding cluster center is greater than the second preset threshold Y, that is,

[0205]

[0206] Where v represents the subband, v = 1,...,N sb , w i,v H represents the CSI sample W i The conjugate transpose of the corresponding vector, c i represents the CSI sample W iThe class to which it belongs, Represents class c i The corresponding cluster center U c i The vector corresponding to the neutron band v, Y represents the second preset threshold, ∑ represents a summation operation, ||·|| F represents the Frobenius norm.

[0207] S25. The cluster centers U1,...,U K As the CSI feedback codebook under the first data set T, and the CSI feedback codebook is indicated by B bits, that is, a one-to-one mapping b from the cluster center to the bit stream b is established. j =g(·)∈{0,1} B .

[0208] The CSI encoding and decoding process may include the following S26 to S27:

[0209] S26. Given the CSI sample W to be encoded (also called CSI information W), the transmitter encodes it according to the following formula based on the constructed CSI feedback codebook to obtain the bit stream b2 to be fed back, namely:

[0210] b2=g(argmax j＝1,...,K ∑ Nsb v＝1 ||w v H u j,v || F )

[0211] Where W=[w1,...,w Nsb ] T ∈C Nsb×Nt , w v H Represents the conjugate transpose of the vector corresponding to the CSI sample W, u j,v Represents the cluster center U among the K cluster centers j The vector corresponding to the neutron band v, U j =[u j,1 ,...,u j,Nsb ]∈C Nsb×Nt , g(·) represents the mapping function between K cluster centers and B-bit feedback bit stream, g(·)∈{0,1} B , argmax j＝1,...,K ∑ Nsb v＝1 ||w v H u j,v || F Represents ∑Nsb v＝1 ||w v H u j,v || F The value of j when taking the maximum value, ||·|| F represents the Frobenius norm, and ∑ represents the summation operation.

[0212] S27. After the receiving end receives the bitstream b2, it decodes according to the mapping g(·), that is, W = g -1 (b2), and thus completes the CSI feedback based on the eigenvector.

[0213] The method embodiments of the present application are described in detail above in conjunction with FIG. 8. The device embodiments of the present application are described in detail below in conjunction with FIGS. 9 to 11. It should be understood that the device embodiments and the method embodiments correspond to each other, and similar descriptions can refer to the method embodiments.

[0214] FIG. 9 shows a schematic block diagram of a transmitting device 300 according to an embodiment of the present application. As shown in FIG. 9, the transmitting device 300 includes:

[0215] A processing unit 310, configured to encode target CSI information according to a CSI feedback codebook to obtain a target feedback bitstream;

[0216] A communication unit 320, configured to send the target feedback bitstream;

[0217] Among them, the CSI feedback codebook is K clustering centers obtained by clustering CSI samples in a first data set. Among them, the first data set includes S CSI samples, and both K and S are positive integers, K = 2 B <S, and B is the number of CSI feedback bits.

[0218] In some embodiments, the CSI samples in the first data set are full-channel information matrices in the time domain, or the CSI samples in the first data set are full-channel information matrices in the frequency domain, or the CSI samples in the first data set are matrices composed of eigenvectors obtained by performing eigenvalue decomposition on multiple sub-bands of the full-channel information matrix.

[0219] In some embodiments, the K clustering centers are obtained by clustering through the following steps:

[0220] Perform a normalization operation on each CSI sample in the first data set;

[0221] Randomly select K CSI samples from the first data set as initial clustering centers;

[0222] Loop through the following steps 1 and step 2 Q times to obtain the K clustering centers, where Q is a positive integer:

[0223] Step 1: Determine the class to which each CSI sample in the first data set belongs;

[0224] Step 2: Determine the cluster center of each class.

[0225] In some embodiments, the K cluster centers are obtained by clustering through the following steps:

[0226] performing a normalization operation on each CSI sample in the first data set;

[0227] Randomly select K CSI samples from the first data set as initial cluster centers;

[0228] The following steps 1 and 2 are performed cyclically, and when the quantization errors of all CSI samples in the first data set are less than a first preset threshold, the K cluster centers are obtained:

[0229] Step 1: Determine the class to which each CSI sample in the first data set belongs;

[0230] Step 2: Determine the cluster center of each class.

[0231] In some embodiments, the CSI samples in the first data set are a full channel information matrix in the time domain, or the CSI samples in the first data set are a full channel information matrix in the frequency domain;

[0232] The CSI samples H in the first data set i The quantization error of is less than the first preset threshold value, which is determined based on the following formula, where i=1,...,S:

[0233] ∑ S i＝1 ||H i -U c i || F <X;

[0234] Among them, c i Denotes the CSI sample H i Class, U c i Represents class c i The corresponding cluster center, X represents the first preset threshold, ||·|| F represents the Frobenius norm.

[0235] In some embodiments, the K cluster centers are obtained by clustering through the following steps:

[0236] performing a normalization operation on each CSI sample in the first data set;

[0237] Randomly select K CSI samples from the first data set as initial cluster centers;

[0238] The following steps 1 and 2 are performed cyclically, and when the similarity between all CSI samples in the first data set and the corresponding cluster centers is greater than a second preset threshold, the K cluster centers are obtained:

[0239] Step 1: Determine the class to which each CSI sample in the first data set belongs;

[0240] Step 2: Determine the cluster center of each class.

[0241] In some embodiments, the CSI samples in the first data set are a matrix consisting of eigenvectors obtained by performing eigenvalue decomposition on multiple subbands of the full channel information matrix;

[0242] The CSI samples W in the first data set i The similarity with the corresponding cluster center is greater than the second preset threshold value based on the following formula, where W i =[w i,1 ,...,w i,Nsb ] T ∈C Nsb×Nt , i=1,...,S,N sb Indicates the number of subbands, N t represents the number of transmit antenna ports, and C represents a complex set:

[0243]

[0244] Where v represents the subband, v = 1,...,N sb , w i，v H represents the CSI sample W i The conjugate transpose of the corresponding vector, c i represents the CSI sample W i The class to which it belongs, Represents class c i The corresponding cluster center U c i The vector corresponding to the neutron band v, Y represents the second preset threshold, ∑ represents a summation operation, ||·|| F represents the Frobenius norm.

[0245] In some embodiments, the CSI samples in the first data set are a full channel information matrix in the time domain, or the CSI samples in the first data set are a full channel information matrix in the frequency domain;

[0246] The normalization operation is performed on each CSI sample in the first data set, including:

[0247] The CSI samples H in the first data set are calculated according to the following formula: i Perform normalization operation, where i = 1,...,S:

[0248] ||H1|| F =||H2|| F =...=||H S || F =1;

[0249] Among them, ||·|| F represents the Frobenius norm.

[0250] In some embodiments, determining the class to which each CSI sample in the first data set belongs includes:

[0251] The CSI samples H in the first data set are determined according to the following formula: i Belongs to the category:

[0252] c i =argmin j＝1,...,K ||H i -U j || F ;

[0253] Among them, c i Denotes the CSI sample H i Class, U j represents the jth cluster center, argmin j＝1,...,K ||H i -U j || F Indicates ||H i -U j || F The value of j when taking the minimum value, ||·|| F represents the Frobenius norm.

[0254] In some embodiments, determining the cluster center of each class includes:

[0255] Determine the class c according to the following formula i The cluster center U j , where j = 1, ..., K:

[0256] U j =(∑ S i＝1 f{c i =j}H i ) / (∑S i＝1 f{c i =j});

[0257] Among them, c i Denotes the CSI sample H i Belongs to the class, when c i =j holds true when f{c i =j}=1, when c i =j does not hold true when f{c i =j}=0, ∑ represents the summation operation.

[0258] In some embodiments, the processing unit 310 is specifically configured to:

[0259] The target CSI information is encoded based on the following formula to generate the target feedback bit stream:

[0260] b1=g(argmin j＝1,...,K ||HU j || F );

[0261] Where b1 represents the target feedback bit stream, H represents the target CSI information, and U j represents the jth cluster center among the K cluster centers, g(·) represents the mapping function between the K cluster centers and the B-bit feedback bit stream, g(·)∈{0,1} B , argmin j＝1,...,K ||HU j || F Indicates || HU j || F The value of j when taking the minimum value, ||·|| F represents the Frobenius norm.

[0262] In some embodiments, the CSI samples in the first data set are a matrix consisting of eigenvectors obtained by performing eigenvalue decomposition on multiple subbands of the full channel information matrix;

[0263] The normalization operation is performed on each CSI sample in the first data set, including:

[0264] The CSI samples W in the first data set are calculated according to the following formula: i Perform normalization operation, where W i =[w i,1 ,...,w i,Nsb ] T ∈C Nsb ×Nt , i=1,...,S,N sb Indicates the number of subbands, Nt represents the number of transmit antenna ports, and C represents a complex set:

[0265] ||w i,1 || F =||w i,2 || F =...=||w i,Nsb || F ;

[0266] Among them, ||·|| F represents the Frobenius norm.

[0267] In some embodiments, determining the class to which each CSI sample in the first data set belongs includes:

[0268] The CSI samples W in the first data set are determined according to the following formula: i Belongs to the category:

[0269] c i =argmax j＝1,...,K ∑ Nsb v＝1 ||w i,v H u j,v || F ;

[0270] Among them, c i represents the CSI sample W i The class to which it belongs, w i,v H represents the CSI sample W i The conjugate transpose of the corresponding vector, u j,v Represents the cluster center U j The vector corresponding to the neutron band v, U j =[u j,1 ,...,u j,Nsb ]∈C Nsb×Nt , argmax j＝1,...,K ∑ Nsb v＝1 ||w i,v H u j,v || F Represents ∑ Nsb v＝1 ||w i,v H u j,v || F The value of j when taking the maximum value, ||·|| F represents the Frobenius norm, and ∑ represents the summation operation.

[0271] In some embodiments, determining the cluster center of each class includes:

[0272] Determine the class c according to the following formula i The cluster center U j , where i = 1, ..., S and j = 1, ..., K:

[0273] U j =(∑ S i＝1 f{c i =j}W i ) / (∑ S i＝1 f{c i =j});

[0274] Among them, c i represents the CSI sample W i Belongs to the class, when c i =j holds true when f{c i =j}=1, when c i =j does not hold true when f{c i =j}=0, ∑ represents the summation operation.

[0275] In some embodiments, the processing unit 310 is specifically configured to:

[0276] The target CSI information is encoded based on the following formula to generate the target feedback bit stream:

[0277] b2=g(argmax j＝1,...,K ∑ Nsb v＝1 ||w v H u j,v || F );

[0278] Where b2 represents the target feedback bit stream, W represents the target CSI information, W=[w1,...,w Nsb ] T ∈C Nsb×Nt , w v H Represents the conjugate transpose of the vector corresponding to the target CSI information W, u j,v Represents the cluster center U among the K cluster centers j The vector corresponding to the neutron band v, U j =[u j,1 ,...,u j,Nsb ]∈C Nsb×Nt , g(·) represents the mapping function between the K cluster centers and the B-bit feedback bit stream, g(·)∈{0,1} B, argmax j＝1,...,K ∑ Nsb v＝1 ||w v H u j,v || F denotes ∑ Nsb v＝1 ||w v H u j,v || F the value of j when taking the maximum value, ||·|| F denotes the Frobenius norm, and ∑ denotes the summation operation.

[0279] In some embodiments, the above communication unit may be a communication interface or a transceiver, or an input / output interface of a communication chip or a system-on-chip. The above processing unit may be one or more processors.

[0280] It should be understood that the transmitting device 300 according to the embodiments of the present application may correspond to the transmitting device in the method embodiments of the present application, and the above and other operations and / or functions of each unit in the transmitting device 300 respectively implement the corresponding processes of the transmitting device in the method 200 shown in FIG. 8. For the sake of brevity, they will not be described herein again.

[0281] FIG. 10 shows a schematic block diagram of a receiving device 400 according to an embodiment of the present application. As shown in FIG. 10, the receiving device 400 includes:

[0282] A communication unit 410 for receiving a target feedback bitstream;

[0283] A processing unit 420 for decoding the target feedback bitstream according to a CSI feedback codebook to obtain a target CSI sample;

[0284] wherein, the CSI feedback codebook is K clustering centers obtained by clustering CSI samples in a first dataset, the first dataset includes S CSI samples, both K and S are positive integers, and K = 2 B <S, and B is the number of CSI feedback bits.

[0285] In some embodiments, the CSI samples in the first dataset are time-domain full channel information matrices, or the CSI samples in the first dataset are frequency-domain full channel information matrices, or the CSI samples in the first dataset are matrices composed of eigenvectors obtained by performing eigenvalue decomposition on multiple subbands of the full channel information matrix.

[0286] In some embodiments, the K clustering centers are obtained by clustering through the following steps:

[0287] performing a normalization operation on each CSI sample in the first data set;

[0288] Randomly select K CSI samples from the first data set as initial cluster centers;

[0289] Repeat the following steps 1 and 2 Q times to obtain the K cluster centers, where Q is a positive integer:

[0290] Step 1: Determine the class to which each CSI sample in the first data set belongs;

[0291] Step 2: Determine the cluster center of each class.

[0292] In some embodiments, the K cluster centers are obtained by clustering through the following steps:

[0293] performing a normalization operation on each CSI sample in the first data set;

[0294] Randomly select K CSI samples from the first data set as initial cluster centers;

[0295] The following steps 1 and 2 are performed cyclically, and when the quantization errors of all CSI samples in the first data set are less than a first preset threshold, the K cluster centers are obtained:

[0296] Step 1: Determine the class to which each CSI sample in the first data set belongs;

[0297] Step 2: Determine the cluster center of each class.

[0298] In some embodiments, the CSI samples in the first data set are a full channel information matrix in the time domain, or the CSI samples in the first data set are a full channel information matrix in the frequency domain;

[0299] The CSI samples H in the first data set i The quantization error of is less than the first preset threshold value, which is determined based on the following formula, where i=1,...,S:

[0300] ∑ S i＝1 ||H i -U c i || F <X;

[0301] Among them, c i Denotes the CSI sample H i Class, U c i Represents class c i The corresponding cluster center, X represents the first preset threshold, ||·||F represents the Frobenius norm.

[0302] In some embodiments, the K cluster centers are obtained by clustering through the following steps:

[0303] performing a normalization operation on each CSI sample in the first data set;

[0304] Randomly select K CSI samples from the first data set as initial cluster centers;

[0305] The following steps 1 and 2 are performed cyclically, and when the similarity between all CSI samples in the first data set and the corresponding cluster centers is greater than a second preset threshold, the K cluster centers are obtained:

[0306] Step 1: Determine the class to which each CSI sample in the first data set belongs;

[0307] Step 2: Determine the cluster center of each class.

[0308] In some embodiments, the CSI samples in the first data set are a matrix consisting of eigenvectors obtained by performing eigenvalue decomposition on multiple subbands of the full channel information matrix;

[0309] The CSI samples W in the first data set i The similarity with the corresponding cluster center is greater than the second preset threshold value based on the following formula, where W i =[w i,1 ,...,w i,Nsb ] T ∈C Nsb×Nt , i=1,...,S,N sb Indicates the number of subbands, N t represents the number of transmit antenna ports, and C represents a complex set:

[0310]

[0311] Where v represents the subband, v = 1,...,N sb , w i,v H represents the CSI sample W i The conjugate transpose of the corresponding vector, c i represents the CSI sample W i The class to which it belongs, Represents class c i The corresponding cluster center U c i The vector corresponding to the neutron band v, Y represents the second preset threshold, ∑ represents a summation operation, ||·|| Frepresents the Frobenius norm.

[0312] In some embodiments, the CSI samples in the first data set are a full channel information matrix in the time domain, or the CSI samples in the first data set are a full channel information matrix in the frequency domain;

[0313] The normalization operation is performed on each CSI sample in the first data set, including:

[0314] The CSI samples H in the first data set are calculated according to the following formula: i Perform normalization operation, where i = 1,...,S:

[0315] ||H1|| F =||H2|| F =...=||H S || F =1;

[0316] Among them, ||·|| F represents the Frobenius norm.

[0317] In some embodiments, determining the class to which each CSI sample in the first data set belongs includes:

[0318] The CSI samples H in the first data set are determined according to the following formula: i Belongs to the category:

[0319] c i =argmin j＝1,...,K ||H i -U j || F ;

[0320] Among them, c i Denotes the CSI sample H i Class, U j represents the jth cluster center, argmin j＝1,...,K ||H i -U j || F Indicates ||H i -U j || F The value of j when taking the minimum value, ||·|| F represents the Frobenius norm.

[0321] In some embodiments, determining the cluster center of each class includes:

[0322] Determine the class c according to the following formula i The cluster center U j, where j = 1, ..., K:

[0323] U j =(∑ S i＝1 f{c i =j}H i ) / (∑ S i＝1 f{c i =j});

[0324] Among them, c i Denotes the CSI sample H i Belongs to the class, when c i =j holds true when f{c i =j}=1, when c i =j does not hold true when f{c i =j}=0, ∑ represents the summation operation.

[0325] In some embodiments, the processing unit 420 is specifically configured to:

[0326] The target feedback bit stream is decoded based on the following formula to obtain the target CSI sample:

[0327] H=g -1 (b1);

[0328] Where b1=g(argmin j＝1,...,K ||HU j || F ), b1 represents the target feedback bit stream, H represents the target CSI information, U j represents the jth cluster center among the K cluster centers, g(·) represents the mapping function between the K cluster centers and the B-bit feedback bit stream, g(·)∈{0,1} B , argmin j＝1,...,K ||HU j || F Indicates || HU j || F The value of j when taking the minimum value, ||·|| F represents the Frobenius norm.

[0329] In some embodiments, the CSI samples in the first data set are a matrix consisting of eigenvectors obtained by performing eigenvalue decomposition on multiple subbands of the full channel information matrix;

[0330] The normalization operation is performed on each CSI sample in the first data set, including:

[0331] The CSI samples W in the first data set are calculated according to the following formula:i Perform normalization operation, where W i =[w i,1 ,...,w i,Nsb ] T ∈C Nsb ×Nt , i=1,...,S,N sb Indicates the number of subbands, N t represents the number of transmit antenna ports, and C represents a complex set:

[0332] ||w i,1 || F =||w i,2 || F =...=||w i,Nsb || F ;

[0333] Among them, ||·|| F represents the Frobenius norm.

[0334] In some embodiments, determining the class to which each CSI sample in the first data set belongs includes:

[0335] The CSI samples W in the first data set are determined according to the following formula: i Belongs to the category:

[0336] c i =argmax j＝1,...,K ∑ Nsb v＝1 ||w i,v H u j,v || F ;

[0337] Among them, c i represents the CSI sample W i The class to which it belongs, w i,v H represents the CSI sample W i The conjugate transpose of the corresponding vector, u j,v Represents the cluster center U j The vector corresponding to the neutron band v, U j =[u j,1 ,...,u j,Nsb ]∈C Nsb×Nt , argmax j＝1,...,K ∑ Nsb v＝1 ||w i,v H u j,v || F Represents ∑ Nsbv＝1 ||w i,v H u j,v || F The value of j when taking the maximum value, ||·|| F represents the Frobenius norm, and ∑ represents the summation operation.

[0338] In some embodiments, determining the cluster center of each class includes:

[0339] Determine the class c according to the following formula i The cluster center U j , where i = 1, ..., S and j = 1, ..., K:

[0340] U j =(∑ S i＝1 f{c i =j}W i ) / (∑ S i＝1 f{c i =j});

[0341] Among them, c i represents the CSI sample W i Belongs to the class, when c i =j holds true when f{c i =j}=1, when c i =j does not hold true when f{c i =j}=0, ∑ represents the summation operation.

[0342] In some embodiments, the processing unit 420 is specifically configured to:

[0343] The target feedback bit stream is decoded based on the following formula to obtain the target CSI sample:

[0344] W = g -1 (b2);

[0345] Where b2=g(argmax j＝1,...,K ∑ Nsb v＝1 ||w v H u j,v || F ), W represents the target CSI information, W=[w1,...,w Nsb ] T ∈C Nsb× Nt , b2 represents the target feedback bit stream, w v Hrepresents the conjugate transpose of the vector corresponding to the target CSI sample W, u j,v Represents the cluster center U among the K cluster centers j The vector corresponding to the neutron band v, U j =[u j,1 ,...,u j,Nsb ]∈C Nsb×Nt , g(·) represents the mapping function between the K cluster centers and the B-bit feedback bit stream, g(·)∈{0,1} B , argmax j＝1,...,K ∑ Nsb v＝1 ||w v H u j,v || F Represents ∑ Nsb v＝1 ||w v H u j,v || F The value of j when taking the maximum value, ||·|| F represents the Frobenius norm, and ∑ represents the summation operation.

[0346] In some embodiments, the communication unit may be a communication interface or a transceiver, or an input / output interface of a communication chip or a system on chip. The processing unit may be one or more processors.

[0347] It should be understood that the receiving device 400 according to the embodiment of the present application may correspond to the receiving device in the embodiment of the method of the present application, and the above-mentioned and other operations and / or functions of each unit in the receiving device 400 are respectively for realizing the corresponding processes of the receiving device in the method 200 shown in Figure 8. For the sake of brevity, they will not be repeated here.

[0348] Figure 11 is a schematic structural diagram of a communication device 500 provided in an embodiment of the present application. The communication device 500 shown in Figure 11 includes a processor 510, which can call and run a computer program from a memory to implement the method in the embodiment of the present application.

[0349] In some embodiments, as shown in FIG11 , the communication device 500 may further include a memory 520. The processor 510 may call and execute a computer program from the memory 520 to implement the method in the embodiment of the present application.

[0350] The memory 520 may be a separate device independent of the processor 510 , or may be integrated into the processor 510 .

[0351] In some embodiments, as shown in FIG11 , the communication device 500 may further include a transceiver 530 , and the processor 510 may control the transceiver 530 to communicate with other devices. Specifically, the transceiver 530 may send information or data to other devices, or receive information or data sent by other devices.

[0352] The transceiver 530 may include a transmitter and a receiver. The transceiver 530 may further include an antenna, and the number of antennas may be one or more.

[0353] In some embodiments, the processor 510 may implement the functions of a processing unit in a transmitting device, or the processor 510 may implement the functions of a processing unit in a receiving device, which will not be described in detail here for the sake of brevity.

[0354] In some embodiments, the transceiver 530 may implement the functions of a communication unit in a transmitting device, which will not be described in detail here for the sake of brevity.

[0355] In some embodiments, the transceiver 530 may implement the functions of a communication unit in a receiving device, which will not be described in detail here for the sake of brevity.

[0356] In some embodiments, the communication device 500 may specifically be the transmitting device of the embodiment of the present application, and the communication device 500 may implement the corresponding processes implemented by the transmitting device in each method of the embodiment of the present application. For the sake of brevity, they will not be repeated here.

[0357] In some embodiments, the communication device 500 may specifically be a receiving device of an embodiment of the present application, and the communication device 500 may implement the corresponding processes implemented by the receiving device in each method of the embodiment of the present application. For the sake of brevity, they will not be repeated here.

[0358] Figure 12 is a schematic structural diagram of an apparatus according to an embodiment of the present application. The apparatus 600 shown in Figure 12 includes a processor 610, which can call and execute a computer program from a memory to implement the method according to the embodiment of the present application.

[0359] In some embodiments, as shown in FIG12 , the apparatus 600 may further include a memory 620. The processor 610 may call and execute a computer program from the memory 620 to implement the method in the embodiment of the present application.

[0360] The memory 620 may be a separate device independent of the processor 610 , or may be integrated into the processor 610 .

[0361] In some embodiments, the processor 610 may implement the functions of a processing unit in a transmitting device, or the processor 610 may implement the functions of a processing unit in a receiving device, which will not be described in detail here for the sake of brevity.

[0362] In some embodiments, the apparatus 600 may further include an input interface 630. The processor 610 may control the input interface 630 to communicate with other devices or chips, and specifically, may obtain information or data sent by other devices or chips. Optionally, the processor 610 may be located inside or outside the chip.

[0363] In some embodiments, the input interface 630 may implement the function of a communication unit in a transmitting device, or the input interface 630 may implement the function of a communication unit in a receiving device.

[0364] In some embodiments, the apparatus 600 may further include an output interface 640. The processor 610 may control the output interface 640 to communicate with other devices or chips, and specifically, may output information or data to other devices or chips. Optionally, the processor 610 may be located inside or outside the chip.

[0365] In some embodiments, the output interface 640 may implement the function of a communication unit in a transmitting device, or the output interface 640 may implement the function of a communication unit in a receiving device.

[0366] In some embodiments, the apparatus can be applied to the originating device in the embodiments of the present application, and the apparatus can implement the corresponding processes implemented by the originating device in the various methods of the embodiments of the present application. For the sake of brevity, they will not be repeated here.

[0367] In some embodiments, the apparatus can be applied to the receiving device in the embodiments of the present application, and the apparatus can implement the corresponding processes implemented by the receiving device in each method of the embodiments of the present application. For the sake of brevity, they will not be repeated here.

[0368] In some embodiments, the device mentioned in the embodiments of the present application may also be a chip, such as a system-on-chip, a system-on-chip, a chip system, or a system-on-chip chip.

[0369] FIG13 is a schematic block diagram of a communication system 700 provided in an embodiment of the present application. As shown in FIG13 , the communication system 700 includes a transmitting device 710 and a receiving device 720 .

[0370] Among them, the transmitting device 710 can be used to implement the corresponding functions implemented by the transmitting device in the above method, and the receiving device 720 can be used to implement the corresponding functions implemented by the receiving device in the above method. For the sake of brevity, they will not be repeated here.

[0371] It should be understood that the processor of the embodiments of the present application may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above method embodiment can be completed by hardware integrated logic circuits in the processor or software instructions. The above processor can be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. The various methods, steps, and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in the embodiments of the present application can be directly embodied as being executed by a hardware decoding processor, or can be executed by a combination of hardware and software modules in the decoding processor. The software module can be located in a storage medium mature in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, etc. The storage medium is located in the memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware.

[0372] It is understood that the memory in the embodiments of the present application may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and 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 link dynamic random access memory (SLDRAM), and direct RAM bus random access memory (DR RAM). It should be noted that the memory of the systems and methods described herein is intended to include, but is not limited to, these and any other suitable types of memory.

[0373] It should be understood that the above-mentioned memories are exemplary but not restrictive. For example, the memories in the embodiments of the present application may also be 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 link dynamic random access memory (SLDRAM), and direct RAM RAM (DR RAM), etc. In other words, the memories in the embodiments of the present application are intended to include, but are not limited to, these and any other suitable types of memories.

[0374] An embodiment of the present application also provides a computer-readable storage medium for storing a computer program.

[0375] In some embodiments, the computer-readable storage medium can be applied to the originating device in the embodiments of the present application, and the computer program enables the computer to execute the corresponding processes implemented by the originating device in the various methods of the embodiments of the present application. For the sake of brevity, they will not be repeated here.

[0376] In some embodiments, the computer-readable storage medium can be applied to the receiving device in the embodiments of the present application, and the computer program enables the computer to execute the corresponding processes implemented by the receiving device in the various methods of the embodiments of the present application. For the sake of brevity, they will not be repeated here.

[0377] An embodiment of the present application also provides a computer program product, including computer program instructions.

[0378] In some embodiments, the computer program product can be applied to the originating device in the embodiments of the present application, and the computer program instructions enable the computer to execute the corresponding processes implemented by the originating device in the various methods of the embodiments of the present application. For the sake of brevity, they will not be repeated here.

[0379] In some embodiments, the computer program product can be applied to the receiving device in the embodiments of the present application, and the computer program instructions enable the computer to execute the corresponding processes implemented by the receiving device in the various methods of the embodiments of the present application. For the sake of brevity, they will not be repeated here.

[0380] The embodiment of the present application also provides a computer program.

[0381] In some embodiments, the computer program can be applied to the originating device in the embodiments of the present application. When the computer program runs on a computer, the computer executes the corresponding processes implemented by the originating device in the various methods of the embodiments of the present application. For the sake of brevity, they will not be repeated here.

[0382] In some embodiments, the computer program can be applied to the receiving device in the embodiments of the present application. When the computer program runs on the computer, the computer executes the corresponding processes implemented by the receiving device in the various methods of the embodiments of the present application. For the sake of brevity, they will not be repeated here.

[0383] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0384] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0385] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0386] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0387] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0388] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. In view of this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0389] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present application should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. A method for channel state information (CSI) feedback, characterized in that: include: The transmitting device encodes the target CSI information according to the CSI feedback codebook to obtain the target feedback bit stream; The transmitting device sends the target feedback bit stream; Among them, the CSI feedback codebook is K clustering centers obtained by clustering CSI samples in the first dataset. The first dataset includes S CSI samples, where both K and S are positive integers, and K = 2 B <S, B is the number of CSI feedback bits.

2. The method according to claim 1, wherein The CSI samples in the first data set are a full channel information matrix in the time domain, or the CSI samples in the first data set are a full channel information matrix in the frequency domain, or the CSI samples in the first data set are a matrix consisting of eigenvectors obtained by performing eigenvalue decomposition on multiple subbands of the full channel information matrix.

3. The method according to claim 1 or 2, wherein: The K cluster centers are obtained by clustering through the following steps: performing a normalization operation on each CSI sample in the first data set; Randomly select K CSI samples from the first data set as initial cluster centers; The following steps 1 and 2 are looped and executed Q times to obtain the K cluster centers, where Q is a positive integer: Step 1: Determine the class to which each CSI sample in the first data set belongs; Step 2: Determine the cluster center of each class.

4. The method according to claim 1 or 2, wherein: The K cluster centers are obtained by clustering through the following steps: performing a normalization operation on each CSI sample in the first data set; Randomly select K CSI samples from the first data set as initial cluster centers; The following steps 1 and 2 are performed cyclically, and when the quantization errors of all CSI samples in the first data set are less than a first preset threshold, the K cluster centers are obtained: Step 1: Determine the class to which each CSI sample in the first data set belongs; Step 2: Determine the cluster center of each class.

5. The method according to claim 4, wherein The CSI samples in the first data set are a full channel information matrix in the time domain, or the CSI samples in the first data set are a full channel information matrix in the frequency domain; The CSI samples H in the first data set i The quantization error of is less than the first preset threshold value, which is determined based on the following formula, where i=1,...,S: ∑ S i＝1 ||H i -U c i || F <X; Among them, c i Denotes the CSI sample H i Class, U c i Represents class c i The corresponding cluster center, X represents the first preset threshold, ||·|| F represents the Frobenius norm.

6. The method according to claim 1 or 2, wherein: The K cluster centers are obtained by clustering through the following steps: performing a normalization operation on each CSI sample in the first data set; Randomly select K CSI samples from the first data set as initial cluster centers; The following steps 1 and 2 are performed cyclically, and when the similarity between all CSI samples in the first data set and the corresponding cluster centers is greater than a second preset threshold, the K cluster centers are obtained: Step 1: Determine the class to which each CSI sample in the first data set belongs; Step 2: Determine the cluster center of each class.

7. The method according to claim 6, wherein The CSI samples in the first data set are a matrix consisting of eigenvectors obtained by performing eigenvalue decomposition on multiple subbands of the full channel information matrix; The CSI samples W in the first data set i The similarity with the corresponding cluster center is greater than the second preset threshold value based on the following formula, where W i =[w i,1 ,...,w i,Nsb ] T ∈C Nsb×Nt , i=1,...,S,N sb Indicates the number of subbands, N t represents the number of transmit antenna ports, and C represents a complex set: Where v represents the subband, v = 1,...,N sb , w i,v H represents the CSI sample W i The conjugate transpose of the corresponding vector, c i represents the CSI sample W i The class to which it belongs, Represents class c i The corresponding cluster center U c i The vector corresponding to the neutron band v, Y represents the second preset threshold, ∑ represents a summation operation, ||·|| F represents the Frobenius norm.

8. The method according to any one of claims 3 to 5, characterized in that The CSI samples in the first data set are a full channel information matrix in the time domain, or the CSI samples in the first data set are a full channel information matrix in the frequency domain; The performing a normalization operation on each CSI sample in the first data set includes: The CSI samples H in the first data set are calculated according to the following formula: i Perform normalization operation, where i = 1,...,S: ||H1|| F =||H2|| F =...=||H S || F =1; Among them, ||·|| F represents the Frobenius norm.

9. The method according to claim 8, wherein The determining the class to which each CSI sample in the first data set belongs includes: The CSI samples H in the first data set are determined according to the following formula: i Belongs to the category: c i =argmin j＝1,...,K ||H i -U j || F 4 Among them, c i Denotes the CSI sample H i Class, U j represents the jth cluster center, argmin j＝1,...,K ||H i -U j || F Indicates ||H i -U j || F The value of j when taking the minimum value, ||·|| F represents the Frobenius norm.

10. The method according to claim 8 or 9, characterized in that Determining the cluster center of each class includes: Determine the class c according to the following formula i The cluster center U j , where j = 1, ..., K: U j =(∑ S i＝1 f{c i =j}H i ) / (∑ S i＝1 f{c i =j}); Among them, c i Denotes the CSI sample H i Belongs to the class, when c i =j holds true when f{c i =j}=1, when c i =j does not hold true when f{c i =j}=0, ∑ represents the summation operation.

11. The method according to any one of claims 8 to 10, characterized in that The transmitting device encodes target CSI information according to the CSI feedback codebook to obtain a target feedback bit stream, including: The transmitting device encodes the target CSI information based on the following formula to generate the target feedback bit stream: b1=g(argmin j＝1,...,K ||H-U j || F ); Wherein, b1 represents the target feedback bit stream, H represents the target CSI information, and U j represents the jth cluster center among the K cluster centers, g(·) represents the mapping function between the K cluster centers and the B-bit feedback bit stream, g(·)∈{0,1} B , argmin j＝1,...,K ||HU j || F Indicates || HU j || F The value of j when taking the minimum value, ||·|| F represents the Frobenius norm.

12. The method according to claim 3, 6 or 7, wherein: The CSI samples in the first data set are a matrix consisting of eigenvectors obtained by performing eigenvalue decomposition on multiple subbands of the full channel information matrix; The performing a normalization operation on each CSI sample in the first data set includes: The CSI samples W in the first data set are calculated according to the following formula: i Perform normalization operation, where W i =[w i,1 ,...,w i,Nsb ] T ∈C Nsb×Nt , i=1,...,S,N sb Indicates the number of subbands, N t represents the number of transmit antenna ports, and C represents a complex set: ||in i,1 || F =||in i,2 || F =...=||in i,Nsb || F ; Among them, ||·|| F represents the Frobenius norm.

13. The method according to claim 12, wherein: The determining the class to which each CSI sample in the first data set belongs includes: The CSI samples W in the first data set are determined according to the following formula: i Belongs to the category: c i =argmax j＝1,...,K ∑ Nsb v＝1 ||w i,v H u j,v || F ; Among them, c i represents the CSI sample W i The class to which it belongs, w i,v H represents the CSI sample W i The conjugate transpose of the corresponding vector, u j,v Represents the cluster center U j The vector corresponding to the neutron band v, U j =[u j,1 ,...,u j,Nsb ]∈C Nsb×Nt , argmax j＝1,...,K ∑ Nsb v＝1 ||w i,v H u j,v || F Represents ∑ Nsb v＝1 ||w i,v H u j,v || F The value of j when taking the maximum value, ||·|| F represents the Frobenius norm, and ∑ represents the summation operation.

14. The method according to claim 12 or 13, wherein: Determining the cluster center of each class includes: Determine the class c according to the following formula i The cluster center U j , where i = 1, ..., S and j = 1, ..., K: U j =(∑ S i＝1 f{c i =j}W i ) / (∑ S i＝1 f{c i =j}); Among them, c i represents the CSI sample W i Belongs to the class, when c i =j holds true when f{c i =j}=1, when c i =j does not hold true when f{c i =j}=0, ∑ represents the summation operation.

15. The method according to any one of claims 12 to 14, characterized in that The transmitting device encodes target CSI information according to the CSI feedback codebook to obtain a target feedback bit stream, including: The transmitting device encodes the target CSI information based on the following formula to generate the target feedback bit stream: b2=g(argmax j＝1,...,K ∑ Nsb v＝1 ||w v H u j,v || F ); Wherein, b2 represents the target feedback bit stream, W represents the target CSI information, W=[w1,...,w Nsb ] T ∈C Nsb×Nt , w v H represents the conjugate transpose of the vector corresponding to the target CSI information W, u j,v Represents the cluster center U among the K cluster centers j The vector corresponding to the neutron band v, U j =[u j,1 ,...,u j,Nsb ]∈C Nsb×Nt , g(·) represents the mapping function between the K cluster centers and the B-bit feedback bit stream, g(·)∈{0,1} B , argmax j＝1,...,K ∑ Nsb v＝1 ||w v H u j,v || F Represents ∑ Nsb v＝1 ||w v H u j,v || F The value of j when taking the maximum value, ||·|| F represents the Frobenius norm, and ∑ represents the summation operation.

16. A method for channel state information (CSI) feedback, characterized in that: include: The receiving device receives the target feedback bit stream; The receiving device decodes the target feedback bit stream according to the CSI feedback codebook to obtain a target CSI sample; Among them, the CSI feedback codebook is the K clustering centers obtained by clustering CSI samples in the first data set. The first data set includes S CSI samples, and both K and S are positive integers, where K = 2 B <S, B are the CSI feedback bit numbers.

17. The method according to claim 16, wherein The CSI samples in the first data set are a full channel information matrix in the time domain, or the CSI samples in the first data set are a full channel information matrix in the frequency domain, or the CSI samples in the first data set are a matrix consisting of eigenvectors obtained by performing eigenvalue decomposition on multiple subbands of the full channel information matrix.

18. The method according to claim 16 or 17, wherein: The K cluster centers are obtained by clustering through the following steps: performing a normalization operation on each CSI sample in the first data set; Randomly select K CSI samples from the first data set as initial cluster centers; The following steps 1 and 2 are looped and executed Q times to obtain the K cluster centers, where Q is a positive integer: Step 1: Determine the class to which each CSI sample in the first data set belongs; Step 2: Determine the cluster center of each class.

19. The method according to claim 16 or 17, wherein: The K cluster centers are obtained by clustering through the following steps: performing a normalization operation on each CSI sample in the first data set; Randomly select K CSI samples from the first data set as initial cluster centers; The following steps 1 and 2 are performed cyclically, and when the quantization errors of all CSI samples in the first data set are less than a first preset threshold, the K cluster centers are obtained: Step 1: Determine the class to which each CSI sample in the first data set belongs; Step 2: Determine the cluster center of each class.

20. The method according to claim 19, wherein The CSI samples in the first data set are a full channel information matrix in the time domain, or the CSI samples in the first data set are a full channel information matrix in the frequency domain; The CSI samples H in the first data set i The quantization error of is less than the first preset threshold value, which is determined based on the following formula, where i=1,...,S: ∑ S i＝1 ||H i -U c i || F <X; Among them, c i Denotes the CSI sample H i Class, U c i Represents class c i The corresponding cluster center, X represents the first preset threshold, ||·|| F represents the Frobenius norm.

21. The method according to claim 16 or 17, wherein: The K cluster centers are obtained by clustering through the following steps: performing a normalization operation on each CSI sample in the first data set; Randomly select K CSI samples from the first data set as initial cluster centers; The following steps 1 and 2 are performed cyclically, and when the similarity between all CSI samples in the first data set and the corresponding cluster centers is greater than a second preset threshold, the K cluster centers are obtained: Step 1: Determine the class to which each CSI sample in the first data set belongs; Step 2: Determine the cluster center of each class.

22. The method according to claim 21, wherein The CSI samples in the first data set are a matrix consisting of eigenvectors obtained by performing eigenvalue decomposition on multiple subbands of the full channel information matrix; The CSI samples W in the first data set i The similarity with the corresponding cluster center is greater than the second preset threshold value based on the following formula, where W i =[w i,1 ,...,w i,Nsb ] T ∈C Nsb×Nt , i=1,...,S,N sb Indicates the number of subbands, N t represents the number of transmit antenna ports, and C represents a complex set: Where v represents the subband, v = 1,...,N sb , w i,v H represents the CSI sample W i The conjugate transpose of the corresponding vector, c i represents the CSI sample W i The class to which it belongs, Represents class c i The corresponding cluster center U c i The vector corresponding to the neutron band v, Y represents the second preset threshold, ∑ represents a summation operation, ||·|| F represents the Frobenius norm.

23. The method according to any one of claims 18 to 20, characterized in that The CSI samples in the first data set are a full channel information matrix in the time domain, or the CSI samples in the first data set are a full channel information matrix in the frequency domain; The performing a normalization operation on each CSI sample in the first data set includes: The CSI samples H in the first data set are calculated according to the following formula: i Perform normalization operation, where i = 1,...,S: ||H1|| F =||H2|| F =...=||H S || F =1; Among them, ||·|| F represents the Frobenius norm.

24. The method according to claim 23, wherein The determining the class to which each CSI sample in the first data set belongs includes: The CSI samples H in the first data set are determined according to the following formula: i Belongs to the category: c i =argmin j＝1,...,K ||H i -U j || F 4 Among them, c i Denotes the CSI sample H i Class, U j represents the jth cluster center, argmin j＝1,...,K ||H i -U j || F Indicates ||H i -U j || F The value of j when taking the minimum value, ||·|| F represents the Frobenius norm.

25. The method according to claim 23 or 24, wherein: Determining the cluster center of each class includes: Determine the class c according to the following formula i The cluster center U j , where j = 1, ..., K: U j =(∑ S i＝1 f{c i =j}H i ) / (∑ S i＝1 f{c i =j}); Among them, c i Denotes the CSI sample H i Belongs to the class, when c i =j holds true when f{c i =j}=1, when c i =j does not hold true when f{c i =j}=0, ∑ represents the summation operation.

26. The method according to any one of claims 23 to 25, characterized in that The receiving device decodes the target feedback bit stream according to the CSI feedback codebook to obtain a target CSI sample, including: The receiving device decodes the target feedback bit stream based on the following formula to obtain the target CSI sample: H=g -1 (b1); Where b1=g(argmin j＝1,...,K ||HU j || F ), b1 represents the target feedback bit stream, H represents the target CSI information, U j represents the jth cluster center among the K cluster centers, g(·) represents the mapping function between the K cluster centers and the B-bit feedback bit stream, g(·)∈{0,1} B , argmin j＝1,...,K ||HU j || F Indicates || HU j || F The value of j when taking the minimum value, ||·|| F represents the Frobenius norm.

27. The method of claim 18, 20 or 21, wherein: The CSI samples in the first data set are a matrix consisting of eigenvectors obtained by performing eigenvalue decomposition on multiple subbands of the full channel information matrix; The performing a normalization operation on each CSI sample in the first data set includes: The CSI samples W in the first data set are calculated according to the following formula: i Perform normalization operation, where W i =[w i,1 ,...,w i,Nsb ] T ∈C Nsb×Nt , i=1,...,S,N sb Indicates the number of subbands, N t represents the number of transmit antenna ports, and C represents a complex set: ||in i,1 || F =||in i,2 || F =...=||in i,Nsb || F ; Among them, ||·|| F represents the Frobenius norm.

28. The method of claim 27, wherein: The determining the class to which each CSI sample in the first data set belongs includes: The CSI samples W in the first data set are determined according to the following formula: i Belongs to the category: c i =argmax j＝1,...,K ∑ Nsb v＝1 ||w i,v H u j,v || F ; Among them, c i represents the CSI sample W i The class to which it belongs, w i,v H represents the CSI sample W i The conjugate transpose of the corresponding vector, u j,v Represents the cluster center U j The vector corresponding to the neutron band v, U j =[u j,1 ,...,u j,Nsb ]∈C Nsb×Nt , argmax j＝1,...,K ∑ Nsb v＝1 ||w i,v H u j,v || F Represents ∑ Nsb v＝1 ||w i,v H u j,v || F The value of j when taking the maximum value, ||·|| F represents the Frobenius norm, and ∑ represents the summation operation.

29. The method according to claim 27 or 28, wherein Determining the cluster center of each class includes: Determine the class c according to the following formula i The cluster center U j , where i = 1, ..., S and j = 1, ..., K: U j =(∑ S i＝1 f{c i =j}W i ) / (∑ S i＝1 f{c i =j}); Among them, c i represents the CSI sample W i Belongs to the class, when c i =j holds true when f{c i =j}=1, when c i =j does not hold true when f{c i =j}=0, ∑ represents the summation operation.

30. The method according to any one of claims 27 to 29, wherein The receiving device decodes the target feedback bit stream according to the CSI feedback codebook to obtain a target CSI sample, including: The receiving device decodes the target feedback bit stream based on the following formula to obtain the target CSI sample: W=g -1 (b2); Where b2=g(argmax j＝1,...,K ∑ Nsb v＝1 ||w v H u j,v || F ), W represents the target CSI information, W=[w1,...,w Nsb ] T ∈C Nsb ×Nt , b2 represents the target feedback bit stream, w v H represents the conjugate transpose of the vector corresponding to the target CSI sample W, u j,v Represents the cluster center U among the K cluster centers j The vector corresponding to the neutron band v, U j =[u j,1 ,...,u j,Nsb ]∈C Nsb×Nt , g(·) represents the mapping function between the K cluster centers and the B-bit feedback bit stream, g(·)∈{0,1} B , argmax j＝1,...,K ∑ Nsb v＝1 ||w v H u j,v || F Represents ∑ Nsb v＝1 ||w v H u j,v || F The value of j when taking the maximum value, ||·|| F represents the Frobenius norm, and ∑ represents the summation operation.

31. A terminating device, characterized in that: include: a processing unit, configured to encode target CSI information according to a CSI feedback codebook to obtain a target feedback bit stream; a communication unit, configured to send the target feedback bit stream; Among them, the CSI feedback codebook is K clustering centers obtained by clustering CSI samples in the first data set. The first data set includes S CSI samples. Both K and S are positive integers, and K = 2 B <S, B are the number of CSI feedback bits.

32. A receiving device, characterized in that: include: A communication unit, configured to receive a target feedback bit stream; a processing unit, configured to decode the target feedback bit stream according to a CSI feedback codebook to obtain a target CSI sample; Among them, the CSI feedback codebook is K clustering centers obtained by clustering CSI samples in the first data set. The first data set includes S CSI samples. Both K and S are positive integers, and K = 2 B <S, B is the number of CSI feedback bits.

33. A terminating device, characterized in that: include: A processor and a memory, the memory being used to store a computer program, the processor being used to call and run the computer program stored in the memory, so that the originating device executes the method according to any one of claims 1 to 15.

34. A receiving device, characterized in that: include: A processor and a memory, the memory being used to store a computer program, the processor being used to call and run the computer program stored in the memory, so that the receiving device executes the method as described in any one of claims 16 to 30.

35. A chip, characterized in that: include: A processor, configured to call and run a computer program from a memory, so that a device equipped with the chip executes a method as described in any one of claims 1 to 15, or so that a device equipped with the chip executes a method as described in any one of claims 16 to 30.

36. A computer-readable storage medium, characterized in that Used to store a computer program, when the computer program is executed, the method according to any one of claims 1 to 15 is implemented, or the method according to any one of claims 16 to 30 is implemented.

37. A computer program product, characterized in that The method comprises computer program instructions, and when the computer program instructions are executed, the method according to any one of claims 1 to 15 is implemented, or the method according to any one of claims 16 to 30 is implemented.

38. A computer program, characterized in that When the computer program is executed, the method according to any one of claims 1 to 15 is implemented, or the method according to any one of claims 16 to 30 is implemented.

Images