Quantization information receiving method, quantization information sending method, apparatus, device, medium, and program product
By quantizing the information transmission and reception method, the encoder and decoder parameters are made consistent, which solves the decoding error problem caused by parameter misalignment during the joint coding of source and channel, and improves system performance.
Patent Information
- Application Number
- PCT/CN2025/093732
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2025-05-09
- Publication Date
- 2026-11-12
AI Technical Summary
In the joint encoding and decoding process of source and channel, misalignment of encoder and decoder parameters leads to decoding errors and affects decoding performance.
By using a quantitative information transmission and reception method, the quantization and dequantization processes between the first and second devices are ensured to be consistent, and the same encoder and decoder parameters are used to achieve information unification.
It improves decoding accuracy and system performance, and avoids decoding errors caused by parameter misalignment.
Smart Images

Figure CN2025093732_12112026_PF_FP_ABST
Abstract
Description
Quantitative information transmission and reception methods, devices, equipment, media and program products Technical Field
[0001] This application relates to the field of communication technology, and in particular to a method, apparatus, device, medium and program product for quantitative information transmission and reception. Background Technology
[0002] During the joint encoding and decoding of the source and channel, the parameters of the encoder (quantization module) and decoder (dequantization module) need to be aligned, that is, the same parameters need to be used; otherwise, decoding errors may occur.
[0003] Related technologies rely on fixed parameters for encoding and decoding. However, if the parameters change and become misaligned, it will affect the decoding performance or even cause decoding errors. Summary of the Invention
[0004] This application provides a method, apparatus, device, medium, and program product for quantitative information transmission and reception, the technical solution of which includes at least:
[0005] According to one aspect of the embodiments of this application, a method for receiving quantized information is provided, the method being executed by a first device, the method comprising:
[0006] The system receives first information sent by the second device. The first information is related to quantization and / or dequantization between the first device and the second device. Quantization is related to the encoder, and dequantization is related to the decoder.
[0007] According to another aspect of the embodiments of this application, a method for quantizing information transmission is provided, the method being executed by a second device, the method comprising:
[0008] Send first information to the first device. The first information is related to quantization and / or dequantization between the first device and the second device. Quantization is related to the encoder, and dequantization is related to the decoder.
[0009] According to another aspect of the embodiments of this application, a first device is provided, the first device comprising:
[0010] The receiving module is used to receive first information sent by the second device. The first information is related to quantization and / or dequantization between the first device and the second device. Quantization is related to the encoder, and dequantization is related to the decoder.
[0011] According to another aspect of the embodiments of this application, a second apparatus is provided, the second apparatus comprising:
[0012] The transmitting module is used to transmit first information to the first device. The first information is related to quantization and / or dequantization between the first device and the second device. Quantization is related to the encoder, and dequantization is related to the decoder.
[0013] According to another aspect of the embodiments of this application, a first device is provided, the first device comprising:
[0014] A processor; a transceiver connected to the processor; a memory for storing executable instructions of the processor; wherein the processor is configured to load and execute the executable instructions to implement the quantized information receiving method as described above.
[0015] According to another aspect of the embodiments of this application, a second device is provided, the second device comprising:
[0016] A processor; a transceiver connected to the processor; a memory for storing executable instructions of the processor; wherein the processor is configured to load and execute the executable instructions to implement the quantitative information transmission method as described above.
[0017] According to another aspect of the embodiments of this application, a computer-readable storage medium is provided, which stores at least one program that is loaded and executed by a processor to implement the quantization information receiving method and / or quantization information transmitting method as described in the above aspects.
[0018] According to another aspect of the embodiments of this application, a chip is provided, the chip including programmable logic circuits and / or program instructions, which, when running on a first device, is used to implement the quantization information receiving method of the above aspects; and when running on a second device, is used to implement the quantization information transmitting method of the above aspects.
[0019] According to another aspect of the embodiments of this application, a computer program product or computer program is provided, which includes computer instructions stored in a computer-readable storage medium, a processor retrieving the computer instructions from the computer-readable storage medium, and the processor executing the computer instructions to implement the quantized information receiving method and / or quantized information transmitting method as described in the above aspects.
[0020] The technical solutions provided in this application embodiment may include the following beneficial effects:
[0021] This method receives first information sent by a second device. The first information is related to quantization and / or dequantization between the first and second devices. Quantization is related to the encoder, and dequantization is related to the decoder, so that the first and second devices can achieve information consistency for quantization and / or dequantization. Attached Figure Description
[0022] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0023] Figure 1 shows a schematic diagram of a communication system provided in an exemplary embodiment of this application;
[0024] Figure 2 shows a schematic diagram of the self-encoder framework provided by the related technology;
[0025] Figure 3 illustrates a schematic diagram of compressed feedback of joint channel state information (CSI) in the spatial, frequency, and time domains based on artificial intelligence (AI) provided by related technologies;
[0026] Figure 4 illustrates a schematic diagram of AI-based joint CSI prediction and compression provided by related technologies;
[0027] Figure 5 shows a schematic diagram of the CSI feedback method based on joint source-channel coding provided by related technologies;
[0028] Figure 6 shows a schematic diagram of the CSI feedback method based on joint source-channel coding provided by related technologies;
[0029] Figure 7 shows a schematic diagram of a quantitative information transmission and reception method provided in an exemplary embodiment of this application;
[0030] Figure 8 shows a flowchart of a quantization information receiving method provided in an exemplary embodiment of this application;
[0031] Figure 9 shows a flowchart of a quantitative information transmission method provided in an exemplary embodiment of this application;
[0032] Figure 10 shows a flowchart of a quantitative information transmission and reception method provided in an exemplary embodiment of this application;
[0033] Figure 11 shows a block diagram of a first apparatus provided in an exemplary embodiment of this application;
[0034] Figure 12 shows a block diagram of a second apparatus provided in an exemplary embodiment of this application;
[0035] Figure 13 shows a schematic diagram of the structure of a first device provided in an exemplary embodiment of this application;
[0036] Figure 14 shows a schematic diagram of the structure of a second device provided in an exemplary embodiment of this application. Detailed Implementation
[0037] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be further described in detail below with reference to the accompanying drawings. Exemplary embodiments will be described in detail here, examples of which are illustrated in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.
[0038] The terminology used in this disclosure is for the purpose of describing particular embodiments only and is not intended to be limiting of the disclosure. The singular forms “a,” “the,” and “the” as used in this disclosure and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any and all possible combinations of one or more of the associated listed items.
[0039] It should be understood that although the terms first, second, third, etc., may be used in this disclosure to describe various information, such information should not be limited to these terms. These terms are used only to distinguish information of the same type from one another. For example, without departing from the scope of this disclosure, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to determination."
[0040] The technical solutions described in some embodiments of this application can be applied to various communication systems, such as: Long Term Evolution (LTE) systems, Advanced Long Term Evolution (LTE-A) systems, New Radio (NR) systems, evolution systems of NR systems, LTE-based access to unlicensed spectrum (LTE-U) systems, NR-based access to unlicensed spectrum (NR-U) systems, Non-Terrestrial Networks (NTN) systems, Universal Mobile Telecommunication System (UMTS), Wireless Local Area Networks (WLAN), Wireless Fidelity (WiFi), 5th-Generation (5G) systems, cellular IoT systems, cellular passive IoT systems, and can also be applied to subsequent evolution systems of 5G NR systems, as well as 6G and subsequent evolution systems.
[0041] It should be understood that in some embodiments of this application, "5G" may also be referred to as "5G NR" or "NR".
[0042] It should be understood that in the description of the embodiments of this application, the term "correspondence" may indicate that there is a direct or indirect correspondence between the two, or that there is an association between the two, or that there is a relationship of instruction and being instructed, configuration and being configured, etc.
[0043] In this application embodiment, "predefined" can be implemented by pre-storing corresponding codes, tables, or other means that can be used to indicate relevant information in the device (e.g., including terminal devices and network devices). This application does not limit the specific implementation method. For example, predefined can refer to what is defined in the protocol.
[0044] In this application embodiment, "protocol" may refer to standard protocols in the field of communication, such as IEEE 802.11ax protocol, IEEE 802.11be protocol, IEEE 802.11bn protocol, and related protocols applied in future communication systems. This application does not limit this.
[0045] Figure 1 shows a schematic diagram of a communication system 10 provided in an exemplary embodiment of this application. The communication system 10 includes a first device 110 and a second device 120. The two devices can be a first terminal device and a second terminal device, or a terminal device and a network device, or a station (STA) and an access point (AP), and this application does not limit this to any particular type. This application uses the example of the communication system 10 including a first device 110 (terminal device) and a second device 120 (network device) for illustration.
[0046] In some embodiments, communication in the communication system can be communication between the first device 110 and the second device 120, or communication between the first device 110 and other terminal devices. For example, there are two communication scenarios between the second device 120 and the first device 110: an uplink communication scenario and a downlink communication scenario. Uplink communication involves the first device 110 sending data to the second device 120; downlink communication involves the second device 120 sending data to the first device 110.
[0047] The first device 110 in this embodiment can be a device with wireless transceiver capabilities, such as a device supporting the 802.11 series of protocols, or a device capable of communicating with the second device 120 or other terminal devices. For example, the first device 110 is any user communication device that allows users to communicate with the second device 120 and thus with a WLAN. The first device 110 can be, for example, a user equipment (UE), a mobile station (MS), a mobile terminal (MT), an access terminal, a user unit, a user station, a mobile station, a mobile station, a remote station, a remote terminal, a mobile device, a user terminal, a terminal, a wireless communication device, a user agent, or a user device, etc.
[0048] The first device 110 in this embodiment can also be a device that provides voice / data / image connectivity to a user. For example, it can be a handheld device, in-vehicle device, home appliance, gaming device, etc., that has wireless connectivity or is equipped with a wireless communication module. Examples include: mobile phones, tablets, laptops, PDAs, mobile internet devices (MIDs), wearable devices, virtual reality (VR) devices, augmented reality (AR) devices, wireless terminals in industrial control, wireless terminals in self-driving vehicles, drones or aerial photography equipment, wireless terminals in remote medical surgery, wireless terminals in smart grids, wireless terminals in transportation safety, wireless terminals in smart cities, wireless terminals in smart homes, cellular phones, cordless phones, Session Initiation Protocol (SIP) phones, Wireless Local Loop (WLL) stations, personal digital assistants (PDAs), handheld devices with wireless communication capabilities, computing devices or other processing devices connected to a wireless modem, in-vehicle devices, wearable devices, terminal devices in 5G networks, or future evolved Public Land Mobile networks. Terminal devices in a network (PLMN) can also be televisions, refrigerators, washing machines, kitchen appliances, door locks, fish tanks, robot vacuum cleaners, game consoles, cameras / camcorders, etc. with wireless connectivity, but this application embodiment is not limited to these.
[0049] By way of example and not limitation, in this embodiment, the first device 110 can also be a wearable device. Wearable devices, also known as wearable smart devices, are a general term for devices that utilize wearable technology to intelligently design and develop everyday wearables, such as glasses, gloves, watches, clothing, and shoes. Examples include smartwatches or smart glasses, as well as devices that focus on a specific type of application function and require cooperation with other devices such as smartphones, such as various smart bracelets and smart jewelry for vital sign monitoring.
[0050] Furthermore, in some embodiments, the first device 110 can also be a terminal device in an Internet of Things (IoT) system. IoT is an important component of future information technology development, and its main technical feature is connecting objects to networks through communication technologies, thereby realizing an intelligent network of human-machine interconnection and object-to-object interconnection. In the embodiments of this application, IoT technology can achieve massive connectivity, deep coverage, and terminal power saving through technologies such as narrowband (NB).
[0051] Furthermore, in some embodiments, the first device 110 may also be an in-vehicle communication device in the vehicle-to-everything (V2X) system or the vehicle itself. The communication methods in the V2X system are collectively referred to as V2X (where X represents anything). For example, V2X communication includes: vehicle-to-vehicle (V2V) communication, vehicle-to-infrastructure (V2I) communication, vehicle-to-pedestrian (V2P) communication, or vehicle-to-network (V2N) communication, etc.
[0052] In some embodiments, the second device 120 may be a device that supports the 802.11be standard. The second device 120 may also be a device that supports various current and future 802.11 family WLAN standards such as 802.11bn, 802.11ax, 802.11ac, 802.11n, 802.11g, 802.11b, and 802.11a.
[0053] In this embodiment, the first device 110 may be a mobile phone, tablet computer, computer, virtual reality device, augmented reality device, communication device in industrial control, set-top box, communication device in autonomous driving, vehicle communication device, communication device in telemedicine, communication device in smart grid, communication device in transportation safety, communication device in smart city, or communication device in smart home, wireless communication chip, etc., that supports WLAN / Wi-Fi technology. WLAN technology may support frequency bands including but not limited to: low frequency bands (2.4GHz, 5GHz, 6GHz) and high frequency bands (45GHz, 60GHz).
[0054] The following section describes the relevant technologies involved in the embodiments of this application:
[0055] Artificial intelligence and semantic communication technologies;
[0056] Artificial intelligence (AI) technology relies on the development of different types of neural networks and deep learning algorithms, and has already achieved widespread application in various fields such as image, speech, and video processing. Typical neural network architectures include fully connected networks, convolutional neural networks, recurrent neural networks, and transformer structures with self-attention mechanisms, which can accomplish different task objectives.
[0057] Furthermore, semantic communication technology, which combines AI with the development of future wireless communication systems, has received widespread attention in recent years. In contrast, the traditional communication paradigm is called "syntactic communication," which uses traditional information theory methods such as Shannon's formula, information entropy, and the Nyquist sampling theorem to compress source information and then achieve efficient and reliable transmission through channel coding. "Semantic communication," on the other hand, primarily studies the extraction and representation of semantic information from source information, as well as the design of semantic communication systems integrated with AI, thereby coupling various modules together to achieve the transmission of semantic information.
[0058] A key difference between traditional grammatical communication systems and modern semantic communication systems lies in the design of source coding and channel coding. In traditional grammatical communication systems, source coding and channel coding are separate. Specifically, at the transmitter, source coding first compresses source information such as images, speech, and video into the original bitstream to be transmitted. This bitstream is then transmitted using channel coding methods such as Low Density Parity Check (LDPC) codes and Polar codes. At the receiver, the original bitstream is first recovered through channel decoding, and then the original source is reconstructed through source decoding. This separation of source and channel coding facilitates modular optimization and design. Source coding compresses information redundancy, saving transmission bandwidth, while channel decoding enhances redundancy, improving the information's ability to resist noise and interference, and enabling error detection and correction.
[0059] In modern semantic communication systems, source coding and channel coding are jointly designed. Specifically, on the transmitter side, a joint source-channel encoder is designed to directly encode the original source information into the bit stream to be transmitted; on the receiver side, a joint source-channel decoder is designed to directly recover the original source information from the received bit stream. This method of coupling source and channel can achieve joint optimization in source compression redundancy and channel redundancy. Furthermore, since both the encoder and decoder are designed using AI models, end-to-end training and deployment can be achieved, making the AI model more adaptable to the current wireless channel characteristics and noise conditions, thus optimizing system performance. Simultaneously, joint source-channel coding can further incorporate modulation and demodulation functions, further optimizing system performance.
[0060] • Channel State Information (CSI) feedback;
[0061] CSI feedback schemes typically employ codebook-based eigenvector feedback to enable the base station to acquire downlink CSI. Specifically, the base station sends a downlink CSI Reference Signal (CSI-RS) to the terminal (user). The terminal uses the downlink CSI-RS to estimate the CSI of the downlink channel and performs eigenvalue decomposition on the estimated downlink channel to obtain the eigenvector corresponding to the downlink channel.
[0062] Specifically, the relevant technologies provide two codebook design schemes: Type I and Type II. The Type I codebook is used for CSI feedback with normal precision and is mainly used for transmission in Single-User Multiple-Input Multiple-Output (SU-MIMO) scenarios. The Type II codebook is mainly used to improve the transmission performance of Multi-User MIMO (MU-MIMO).
[0063] For both Type I and Type II codebooks, a two-level codebook feedback system W = W1W2 is used. W1 describes the wideband and long-term characteristics of the channel and determines a set of L Discrete Fourier Transform (DFT) beams. W2 describes the subband and short-term characteristics of the channel.
[0064] Specifically, for the Type I codebook, W2 selects one beam from L DFT beams; for the Type II codebook, W2 linearly combines the L DFT beams in W1, providing feedback in the form of amplitude and phase. Generally, the Type II codebook utilizes a higher number of feedback bits to achieve higher precision CSI feedback performance.
[0065] In NR, the CSI reporting process is carried by Uplink Control Information (UCI) on the Physical Uplink Control Channel (PUCCH) or the Physical Uplink Shared Channel (PUSCH). This includes periodic and semi-persistent reporting on the PUCCH, and aperiodic and semi-persistent reporting on the PUSCH. When the receiver fails to receive the UCI information correctly based on Cyclic Redundancy Code (CRC) verification, causing CSI feedback failure, for periodic and semi-persistent reporting, the network side does not perform any other operations, continues to use the reporting result from the previous CSI feedback cycle, and waits for the CSI to be reported in the next cycle. For aperiodic reporting, the network side reschedules a set of CSI-RS transmission and user CSI feedback resources and re-performs CSI feedback.
[0066] • AI-based CSI feedback;
[0067] Given the tremendous success of AI technology, especially deep learning, in areas such as computer vision and natural language processing, the communications field has begun to explore the use of deep learning to solve technical challenges that are difficult to address using traditional communication methods.
[0068] The neural network architecture commonly used in deep learning is nonlinear and data-driven. It can extract features from actual channel matrix data and reconstruct the compressed channel matrix information from the UE at the base station as much as possible. While ensuring the restoration of channel information, it also provides the possibility of reducing CSI feedback overhead at the UE. Deep learning-based CSI feedback treats channel information as an image to be compressed, uses a deep learning autoencoder to compress the input channel information, and reconstructs the compressed channel image at the transmitting end, which can preserve channel information to a greater extent.
[0069] In the discussion of relevant standards, AI-based CSI feedback, as one of the main use cases of AI projects, has undergone multiple rounds of discussion on simulation results and potential standard impact. Its main implementation framework is shown in Figure 2, which illustrates a schematic diagram of the autoencoder framework provided by the relevant technology.
[0070] The entire feedback system consists of an encoder and a decoder, deployed at the user transmitter (terminal side) and the base station receiver (network side), respectively. After obtaining channel information through channel estimation, it serves as the input to the encoder. The encoder's neural network compresses and encodes the channel information to obtain a compressed bitstream (feedback bitstream). This feedback bitstream is then fed back to the base station via an air interface feedback link. The base station uses the decoder to recover the channel information based on the feedback bitstream and outputs complete feedback channel information. The neural networks of the encoder and decoder shown in Figure 2 can employ deep neural networks (DNNs) composed of multiple fully connected layers, convolutional neural networks (CNNs) composed of multiple convolutional layers, recurrent neural networks (RNNs) with structures such as Long Short-Term Memory (LSTM) and Gated Recurrent Units (GRUs), or residual neural network architectures with self-attention mechanisms to improve the performance of the encoder and decoder.
[0071] The CSI inputs and outputs mentioned above can both be full-channel information or feature vector information obtained based on full-channel information. Therefore, current deep learning-based channel information feedback methods are mainly divided into full-channel information feedback and feature vector feedback. While the former can achieve compression and feedback of full-channel information, it has high feedback bitstream overhead and is not supported in existing NR systems. Feature vector-based feedback methods are the feedback architecture currently supported in NR systems. AI-based feature vector feedback methods can achieve higher CSI feedback accuracy with the same feedback bit overhead, or significantly reduce feedback overhead while achieving the same CSI feedback accuracy.
[0072] In subsequent standard discussions, to further enhance the performance of AI-based CSI feedback, the temporal correlation between CSI at different times will be utilized to improve the compressed feedback performance of CSI. Specifically, the implementation methods are mainly divided into the following two types: AI-based joint spatial-frequency-temporal CSI compressed feedback; and AI-based joint CSI prediction and compression.
[0073] (1) AI-based spatial-frequency-temporal joint CSI compression feedback;
[0074] Figure 3 shows a schematic diagram of AI-based spatial-frequency-time joint CSI compression feedback provided by related technologies.
[0075] The encoder input on the user side (terminal side) includes not only the CSI measurement information at the current time (time t), but also historical CSI information from past times (time t-1). This historical CSI information is represented by the feature output of the latent vector space from the encoder at past times and does not have explicit physical meaning. Similarly, the decoder input on the network side includes not only the feedback information at the current time, but also historical CSI information from past times. This historical CSI information is represented by the feature output of the latent space from the decoder at past times and also does not have explicit physical meaning.
[0076] (2) AI-based joint CSI prediction and compression;
[0077] Figure 4 illustrates a schematic diagram of AI-based joint CSI prediction and compression provided by related technologies.
[0078] The user-side (terminal-side) CSI prediction model takes CSI measurement information as input and outputs at least one predicted CSI as the encoder input (or a pre-processed predicted CSI as the encoder input; for example, the predicted CSI is full-channel information, and singular value decomposition can be used to extract feature vectors from different layers for the CSI compression process). In this case, when the encoder input is multiple predicted CSIs, the encoder output is joint feedback information, and the decoder can simultaneously recover multiple predicted CSIs. By extracting the correlation information of multiple predicted CSIs, the performance of CSI feedback can be further enhanced. Furthermore, the user-side CSI prediction and CSI encoding compression processes can be either separated as shown in Figure 4 or coupled, i.e., completed through a single AI model, as indicated by the dashed box.
[0079] • Quantization and dequantization;
[0080] Quantization is the process of converting a continuous analog signal into a discrete digital signal. Specifically, quantization divides the amplitude range of a signal into several discrete level values and uses these discrete levels to approximate the amplitude of the sampled signal.
[0081] Dequantization is the inverse process of quantization, which restores a discrete digital signal to a continuous analog signal. During dequantization, the discrete digital signal is mapped back to a continuous amplitude range based on the characteristics of the quantizer and the quantization level. The goal of dequantization is to restore the original signal as closely as possible while minimizing the impact of quantization errors.
[0082] • CSI feedback method based on source-channel joint coding;
[0083] Besides AI-based CSI feedback, there is another type of CSI feedback method that draws on semantic communication ideas, based on joint source-channel coding (also known as joint source-channel coding). Figure 5 shows a schematic diagram of the CSI feedback method based on joint source-channel coding provided by related technologies.
[0084] Figure 5 illustrates the first method. In this method, the output of the joint source-channel coding is a quantized binary bit stream (quantized bits), and the demodulation output is the unquantized bits. The joint source-channel coding module can be decomposed into a first AI model (joint source-channel coding feature extraction module) for feature extraction and a quantization module. The joint source-channel decoding module can be decomposed into a dequantization module and a second AI model (joint source-channel decoding feature extraction module) for feature extraction.
[0085] At the transmitting end, a joint source-channel encoder (also known as a source-channel joint encoder, or simply encoder) is designed to encode the first CSI (measured CSI) as the original source information into a bit stream to be transmitted (quantized bits), and then map it onto physical resources through a modulation module, such as a Quadrature Phase Shift Keying (QPSK) module, a 16-Quadrature Amplitude Modulation (16QAM) module, or a 64QAM module.
[0086] At the receiver, a joint source-channel decoder (also called a joint source-channel decoder, or simply decoder) is designed to restore the demodulated log-likelihood ratio (the quantized bits to be dequantized) to a second CSI, which is the same as or similar to the first CSI. This method of coupling source and channel coding can achieve joint optimization in source compression redundancy and channel redundancy. Furthermore, since both the encoder and decoder are designed using AI models, end-to-end training and deployment can be achieved, making the AI models more adaptable to the current wireless channel characteristics and noise conditions. This further optimizes the performance of the CSI encoder and CSI decoder against noise and fading channels under non-ideal channel conditions, achieving optimal system performance.
[0087] Furthermore, source-channel joint coding can also include modulation and demodulation functions. Figure 6 shows a schematic diagram of the CSI feedback method based on source-channel joint coding provided by related technologies.
[0088] Figure 6 illustrates the second method. At the transmitting end (user side), the output of the source-channel joint encoder is not a bitstream, but rather complex symbols (complex sequences) that can be directly mapped to physical resources. Power constraints are applied to these complex symbols to satisfy certain conditions. At the receiving end (network side), the source-channel joint decoder takes the received symbols (the received complex sequences) after channel equalization and symbol detection as input and outputs the original information (i.e., recovered CSI). This end-to-end design approach achieves a balance between source coding, channel coding, and modulation, further optimizing end-to-end link performance.
[0089] In the joint encoding and decoding process of the source and channel, the parameters of the encoder (quantization module) and decoder (dequantization module) need to be aligned, i.e., they must use the same parameters; otherwise, decoding errors may occur. Related technologies rely on fixed parameters for encoding and decoding, but if the parameters change and cause misalignment between the two sides, it will affect decoding performance and even lead to decoding errors.
[0090] To address the aforementioned problems, embodiments of this application provide a method for quantizing information transmission and reception. Figure 7 illustrates a schematic diagram of a quantizing information transmission and reception method provided in an exemplary embodiment of this application, wherein the method is executed by a first device 110 and a second device 120.
[0091] This quantitative information transmission and reception method can be described in conjunction with the structure provided in the embodiment of Figure 5, or in conjunction with the structure provided in the embodiment of Figure 6. This application does not limit it in this regard, but only uses the embodiment of Figure 7 (corresponding to the structure provided in the embodiment of Figure 5) as an example for description.
[0092] First information;
[0093] In some embodiments, the second device 120 sends first information to the first device 110; correspondingly, the first device 110 receives the first information sent by the second device 120; wherein the first information is related to quantization and / or dequantization between the first device 110 and the second device 120, the quantization is related to the encoder, and the dequantization is related to the decoder.
[0094] As an example and not a limitation, the encoder can be one or more of the following: source-channel joint coding method, source-channel coding method, channel coding method, first model, first AI / Machine Learning (ML) model, encoder end, coding function, coding model;
[0095] The decoder can be one or more of the following: source-channel joint decoding method, source-channel decoding method, channel decoding method, second model, second AI / ML model, decoding end, decoding function, and decoding model.
[0096] In some embodiments, the combination of encoder and decoder includes one or more of the following:
[0097] The encoder is a source-channel joint coding method, and the decoder is a source-channel joint decoding method; the encoder is a source-channel coding method, and the decoder is a source-channel decoding method; the encoder is a channel coding method, and the decoder is a channel decoding method; the encoder is a first model, and the decoder is a second model; the encoder is a first AI / ML model, and the decoder is a second AI / ML model; the encoder is the encoding end, and the decoder is the decoding end; the encoder is the encoding function, and the decoder is the decoding function; the encoder is the encoding model, and the decoder is the decoding model.
[0098] The encoder and decoder can be understood as equivalent to the above description. In the embodiments of this application, the encoder and decoder are usually used as examples for illustration.
[0099] In some embodiments, the first information includes one or more of the following: first indication information; second indication information; third indication information; fourth indication information; fifth indication information; sixth indication information; seventh indication information; eighth indication information;
[0100] The first indication information is used to determine the association between the encoder related to uplink transmission and the first associated ID, and / or the association between the decoder related to uplink transmission and the second associated ID, where the first associated ID is used to identify the encoder and the second associated ID is used to identify the decoder; the second indication information is used to determine the association between the encoder related to uplink transmission and the quantization parameters, and / or the association between the decoder related to uplink transmission and the quantization parameters; the third indication information is used to determine the quantization parameters related to uplink transmission; the fourth indication information is used to indicate that the quantization mode related to uplink transmission be reverted to the first quantization mode, where the first quantization mode is a preset quantization mode; the fifth indication information is used to determine the association between the encoder related to downlink transmission and the first associated ID, and / or the association between the decoder related to downlink transmission and the second associated ID; the sixth indication information is used to determine the association between the encoder related to downlink transmission and the dequantization parameters, and / or the association between the decoder related to downlink transmission and the dequantization parameters, where the dequantization parameters correspond to the quantization parameters; the seventh indication information is used to determine the dequantization parameters related to downlink transmission; and the eighth indication information is used to indicate that the quantization mode related to downlink transmission be reverted to the first quantization mode.
[0101] • First instruction information;
[0102] In some embodiments, the association between the encoder and the first associated ID includes the association between the encoder's input and the first associated ID.
[0103] In some embodiments, the association between the decoder and the second associated ID includes the association between the decoder's input and the second associated ID.
[0104] The associated ID can be an identifier or a number, used to associate an encoder or decoder.
[0105] For example, the encoder input includes: a first association ID, a feature vector corresponding to the channel, information about the channel in the angle-delay domain, one or more of the Precoding Matrix Indicator (PMI) and Rank Indication (RI); the encoder output includes: a sequence of floating-point numbers to be quantized;
[0106] The decoder's input includes: the second association ID, one or more items from the dequantized floating-point sequence; the decoder's output includes: the feature vector corresponding to the recovered channel, the channel's information in the angle-delay domain, one or more items from PMI and RI.
[0107] The input and output of the encoder and decoder may contain other information, which is only illustrated here for illustrative purposes.
[0108] The association between the encoder and the first associated ID is shown in Table 1, the association between the encoder input and the first associated ID is shown in Table 2, the association between the decoder and the second associated ID is shown in Table 3, and the association between the decoder input and the second associated ID is shown in Table 4.
[0109] Table 1
[0110] Table 2
[0111] Table 3
[0112] Table 4
[0113] Second instruction information;
[0114] In some embodiments, the quantization parameters include one or more of the following: quantization step size; quantization bit depth; quantization range; quantization type; quantization module parameters.
[0115] For example, the quantization type is one or more of uniform quantization, non-uniform quantization, vector quantization, and scalar quantization; the quantization module parameters include encoder-based model parameters and structure.
[0116] Taking the second indication information used to determine the association between the encoder and quantization parameters related to uplink transmission as an example, the association between the encoder and quantization parameters is shown in Table 5.
[0117] Table 5
[0118] For example, quantization parameter 1 includes: quantization step size = 0.1, quantization type = uniform quantization, then encoding model 1 and quantization step size = 0.1, quantization type = uniform quantization association.
[0119] For example, the second indication information indicates the quantization parameters: {quantization step size = 0.1, quantization type = uniform quantization}, and is associated with the source-channel joint coding feature extraction module_v1 (encoder). For the first device 110, after obtaining the PMI value or RI value, it obtains the floating-point sequence to be quantized through the source-channel joint coding feature extraction module_v1, and outputs the quantized bits using a quantization step size of 0.1 and uniform quantization. After receiving the quantized bits, the second device uses the source-channel joint decoding feature extraction module_v1 (decoder) to recover the original PMI value or RI value based on the quantization step size of 0.1.
[0120] In actual implementation, it may not be implemented exactly as described in the above embodiments. As long as the second indication information can represent the association between the encoder and quantization parameters related to uplink transmission, and / or the association between the decoder and quantization parameters related to uplink transmission, the first device 110 can understand the associated quantization parameters based on the second indication information.
[0121] • Third instruction information;
[0122] In some embodiments, the third indication information includes quantization parameters; or, the third indication information includes an encoder and / or a decoder, and the second indication information and the third indication information are used to jointly determine the quantization parameters; or, the third indication information includes a first association ID and / or a second association ID, and the first indication information, the second indication information and the third indication information are used to jointly determine the quantization parameters.
[0123] When the third indication information includes quantization parameters, the relationship between the third indication information and the quantization parameters is shown in Table 6; taking the third indication information including the encoder as an example, the relationship between the third indication information and the quantization parameters is shown in Table 7; taking the third indication information including the first associated ID as an example, the relationship between the third indication information and the quantization parameters is shown in Table 8.
[0124] Table 6
[0125] Table 7
[0126] Table 8
[0127] The third indication information can directly include quantization parameters, thus enabling the direct determination of quantization parameters.
[0128] The third indication information may also include the encoder. In this case, the second indication information is used to determine the association between the encoder and the quantization parameters related to the uplink transmission, so that both are used to jointly determine the quantization parameters. For example, if the third indication information indicates that the encoder is a first AI / ML model (encoding model 1), then the first device determines quantization parameter 1 based on the association between the first AI / ML model and the quantization parameters in the second indication information.
[0129] The third indication information may also include a first association ID. The first indication information is used to determine the association between the encoder related to uplink transmission and the first association ID. The second indication information is used to determine the association between the encoder related to uplink transmission and the quantization parameters. These three information are then used together to determine the quantization parameters. For example, if the third indication information indicates that the first association ID is 001, then the first indication information is used to determine that the encoder associated with "001" is the first AI / ML model (encoding model 1). The first device then determines quantization parameter 1 based on the association between the first AI / ML model and the quantization parameters in the second indication information.
[0130] • Fourth instruction message;
[0131] In some embodiments, the first device 110 falls back from the quantization mode related to uplink transmission to the first quantization mode (also referred to as the basic quantization mode) according to the fourth instruction information.
[0132] Basic quantization is a quantization method with lower precision (e.g., some quantization parameters are below a preset threshold), thus ensuring transmission reliability by sacrificing precision. For example, it may have a larger quantization step size or a lower quantization bit depth.
[0133] When the second device 120 detects a significant deterioration in the channel quality related to the first device 110, for example, when the Channel Quality Indicator (CQI) value reported by the first device 110 is lower than a preset threshold, such as when the CQI value is lower than 3, the second device 120 will trigger the first device 110 to fall back to the first quantization mode through the fourth indication information, thereby ensuring the reliability of transmission. For example, the first quantization mode is set to 2-bit scalar uniform quantization, and the quantization parameters include: quantization type is uniform quantization, quantization bit depth is 2 bits, and quantization range is [-1, 1].
[0134] In some embodiments, the first device 110 switches to a second quantization mode, which is indicated by a third indication information;
[0135] The third indication information is sent by the second device 120 when the parameter used to indicate channel quality is higher than the first threshold, where the first threshold is a preset threshold.
[0136] When the channel quality associated with the first device 110 is restored, for example, when the CQI value is higher than 3, the second device 120 re-instructs the first device 110 to switch to a higher precision second quantization mode (e.g., switch to 8-bit vector quantization) through the third indication information, that is, to cancel the first quantization mode.
[0137] • Fifth to eighth instruction messages;
[0138] The first to fourth indication messages are related to uplink transmission, and the fifth to eighth indication messages are related to downlink transmission. The specific implementation principle is the same as that of the first to fourth indication messages, and will not be repeated here.
[0139] • First ability information;
[0140] In some embodiments, the first device 110 sends first capability information to the second device 120. The first capability information is used to indicate the capability of the first device 110, and the capability of the first device 110 is related to the quantization and / or dequantization between the first device 110 and the second device 120.
[0141] In some embodiments, the capabilities of the first device 110 include one or more of the following: supported encoders and version numbers; supported quantization method types; supported quantization parameter configuration capabilities; whether it supports switching to the first quantization mode; the ability to bind encoders and quantization parameters; the current resource status of the first device 110; and whether it supports receiving and processing first indication information to eighth indication information.
[0142] For example, taking the encoder as the first AI / ML model, the capabilities of the first device 110 may include model name, architecture type, corresponding training set, etc.; it may also include quantization step size, quantization bit depth, quantization range, quantization type, etc.; it may also include the current computing power and power consumption of the first device 110, etc., which are not limited in this application embodiment.
[0143] In some embodiments, the first capability information includes first reporting information and / or second reporting information, wherein the first reporting information is reported according to frequency band dimension, carrier dimension, or UE dimension; and the second reporting information is reported according to serving cell dimension.
[0144] In some embodiments, the first reporting information is reported by frequency band; or, the first reporting information is reported by combination of frequency bands; or, the first reporting information is reported by each frequency band in the combination of frequency bands; or, the first reporting information is reported by each carrier in each frequency band in the combination of frequency bands; or, the first reporting information is reported by frequency band range; or, the first reporting information is reported by UE.
[0145] In some embodiments, the second reporting information does not carry serving cell indication information, and the second reporting information is applicable to the serving cell currently transmitting the second reporting information; or, the second reporting information is applicable to one or more serving cells currently associated with the first device 110.
[0146] The second reported information does not carry serving cell indication information, that is, it does not indicate that the second reported information applies to a specific serving cell or certain serving cells. In this case, the second reported information applies to the serving cell currently transmitting the second reported information; or it applies to one or more serving cells currently associated with the first device 110, such as all serving cells currently associated with the first device 110.
[0147] In some embodiments, the second reporting information carries serving cell indication information, and the second reporting information is applicable to the serving cell indicated by the serving cell indication information.
[0148] • Joint coding of source and channel and joint decoding of source and channel;
[0149] In some embodiments, the first device 110 includes an encoder that encodes a first CSI into quantized bits based on first information, wherein the first CSI is a measured CSI.
[0150] In some embodiments, the second device 120 includes a decoder that recovers the quantized bit to be dequantized into a second CSI, the second CSI being the same as or similar to the first CSI, and the quantized bit to be dequantized being associated with the quantized bit.
[0151] At the transmitting end (first device 110 side), the encoder treats the first CSI (measured CSI) as the raw source information, encodes it into a bit stream to be transmitted (quantized bits), and maps it to physical resources through a modulation module, such as a QPSK module, 16QAM module, or 64QAM module. The modulated bits are then transmitted to the second device 120 via the uplink channel. At the receiving end (second device 120 side), the demodulation module receives the modulated bits and demodulates them into bits to be dequantized. The decoder recovers the second CSI from the dequantized bits; the second CSI is the same as or similar to the first CSI.
[0152] Figure 8 shows a flowchart of a quantization information receiving method provided in an exemplary embodiment of this application. The method is performed by a first device and includes:
[0153] Step 810: Receive the first information sent by the second device.
[0154] Among them, the first information is related to the quantization and / or dequantization between the first device and the second device, the quantization is related to the encoder, and the dequantization is related to the decoder.
[0155] By way of example and not limitation, the encoder is one or more of the following: source-channel joint coding method, source-channel coding method, channel coding method, first model, first AI / ML model, encoder end, coding function, coding model; the decoder is one or more of the following: source-channel joint decoding method, source-channel decoding method, channel decoding method, second model, second AI / ML model, decoder end, decoding function, decoding model.
[0156] In some embodiments, the combination of encoder and decoder includes one or more of the following: the encoder is a source-channel joint coding method, and the decoder is a source-channel joint decoding method; the encoder is a source-channel coding method, and the decoder is a source-channel decoding method; the encoder is a channel coding method, and the decoder is a channel decoding method; the encoder is a first model, and the decoder is a second model; the encoder is a first AI / ML model, and the decoder is a second AI / ML model; the encoder is an encoding end, and the decoder is a decoding end; the encoder is an encoding function, and the decoder is a decoding function; the encoder is an encoding model, and the decoder is a decoding model.
[0157] The encoder and decoder can be understood as equivalent to the above description. In the embodiments of this application, the encoder and decoder are usually used as examples for illustration.
[0158] In some embodiments, the first device is a terminal device and the second device is a network device; or, the first device is a network device and the second device is a terminal device; or the first device is a first terminal device and the second device is a second terminal device.
[0159] This application can be applied to centimeter-wave communication scenarios, millimeter-wave communication scenarios, low-frequency communication scenarios, channel semantic communication scenarios, and source semantic communication scenarios (channel semantic communication scenarios refer to scenarios where terminal devices send CSI information to network devices, while source semantic communication scenarios can refer to scenarios where terminal devices send non-bitstream information sources such as images, videos, and text to network devices, or scenarios where network devices send non-bitstream information sources such as images, videos, and text to terminal devices), and can also be applied to various other communication scenarios without limitation.
[0160] In this application, the technical solution is typically described using the example of communication between a terminal device and a network device. This application can also be used in scenarios where a first terminal device communicates with a second terminal device, such as sidelink transmission, device-to-device (D2D) transmission, etc.
[0161] In this application, "input" (In) refers to the input of an encoder (or decoder), or the meaning of "input is the input of an encoder (or decoder)," and includes one or more of the following meanings:
[0162] (1) The input is directly used as the input content of the encoder (or decoder);
[0163] (2) After the input is processed, it becomes the input content of the encoder (or decoder). That is, the input content of the encoder (or decoder) is determined based on the input (In).
[0164] Output (Out) refers to the output of the encoder (or decoder), or the meaning of output as the output of the encoder (or decoder), including one or more of the following meanings:
[0165] (1) The output is the direct output content of the encoder (or decoder);
[0166] (2) The output of the encoder (or decoder) is processed to obtain the output, that is, the output (Out) is determined based on the output of the encoder (or decoder).
[0167] The above input (In) and output (Out) are used to represent the various input and output information in the following text.
[0168] In some embodiments, the first information is associated with CSI transmission.
[0169] In other words, the quantization information reception method is associated with CSI transmission, and the encoder and decoder are applied to CSI transmission.
[0170] 1.1 First instruction information to eighth instruction information;
[0171] In some embodiments, the first information includes one or more of the following: first indication information; second indication information; third indication information; fourth indication information; fifth indication information; sixth indication information; seventh indication information; eighth indication information;
[0172] The first indication information is used to determine the association between the encoder related to uplink transmission and the first associated ID, and / or the association between the decoder related to uplink transmission and the second associated ID, where the first associated ID is used to identify the encoder and the second associated ID is used to identify the decoder; the second indication information is used to determine the association between the encoder related to uplink transmission and the quantization parameters, and / or the association between the decoder related to uplink transmission and the quantization parameters; the third indication information is used to determine the quantization parameters related to uplink transmission; the fourth indication information is used to indicate that the quantization mode related to uplink transmission be reverted to the first quantization mode, where the first quantization mode is a preset quantization mode; the fifth indication information is used to determine the association between the encoder related to downlink transmission and the first associated ID, and / or the association between the decoder related to downlink transmission and the second associated ID; the sixth indication information is used to determine the association between the encoder related to downlink transmission and the dequantization parameters, and / or the association between the decoder related to downlink transmission and the dequantization parameters, where the dequantization parameters correspond to the quantization parameters; the seventh indication information is used to determine the dequantization parameters related to downlink transmission; and the eighth indication information is used to indicate that the quantization mode related to downlink transmission be reverted to the first quantization mode.
[0173] In some embodiments, the first to eighth indication information is transmitted via one or more of Downlink Control Information (DCI) signaling, Radio Resource Control (RRC) signaling, Medium Access Control Control Element (MAC CE) signaling, system information, and broadcast information.
[0174] 1.1.1 First instruction information;
[0175] In some embodiments, the association between the encoder and the first associated ID includes the association between the encoder's input and the first associated ID; the association between the decoder and the second associated ID includes the association between the decoder's input and the second associated ID.
[0176] The associated ID can be an identifier or a number, used to associate an encoder or decoder.
[0177] By way of example and not limitation, the encoder input includes one or more of the following: a first association ID, a feature vector corresponding to the channel, information about the channel in the angle-delay domain, PMI, and RI; the encoder output includes: a sequence of floating-point numbers to be quantized; the decoder input includes one or more of the following: a second association ID, a sequence of dequantized floating-point numbers; the decoder output includes one or more of the following: a feature vector corresponding to the channel, information about the channel in the angle-delay domain, PMI, and RI; or,
[0178] The encoder input includes one or more of the following: a first association ID and a non-bitstream source; the encoder output includes: a sequence of floating-point numbers to be quantized; the decoder input includes one or more of the following: a second association ID and a dequantized sequence of floating-point numbers; the decoder output includes: a non-bitstream source.
[0179] For example, the encoder input includes: the feature vector corresponding to the channel, the channel information in the angle-delay domain, PMI, and RI; the encoder output includes: the floating-point number sequence to be quantized; the decoder input includes: the dequantized floating-point number sequence; the decoder output includes: the recovered feature vector corresponding to the channel, the channel information in the angle-delay domain, PMI, and RI.
[0180] Alternatively, the encoder's input includes: non-bitstream information sources such as images, videos, and text; the encoder's output includes: a sequence of floating-point numbers to be quantized; the decoder's input includes: a sequence of floating-point numbers after dequantization; and the decoder's output includes: the recovered non-bitstream information sources such as images, videos, and text.
[0181] Alternatively, the encoder input includes: a first association ID, a feature vector corresponding to the channel, information about the channel in the angle-delay domain, PMI, and RI; the encoder output includes: a floating-point sequence to be quantized; the decoder input includes: a second association ID, a dequantized floating-point sequence; the decoder output includes: a recovered feature vector corresponding to the channel, information about the channel in the angle-delay domain, PMI, and RI.
[0182] Alternatively, the encoder's input includes: a first association ID, non-bitstream information sources such as images, videos, and text; the encoder's output includes: a floating-point sequence to be quantized; the decoder's input includes: a second association ID, a dequantized floating-point sequence; the decoder's output includes: the recovered non-bitstream information sources such as images, videos, and text.
[0183] The input and output of the encoder and decoder may contain other information, which is only an example here.
[0184] In some embodiments, the first indication information is transmitted via one or more of DCI signaling, RRC signaling, MAC CE signaling, system information, and broadcast information.
[0185] 1.1.2 Second instruction information;
[0186] In some embodiments, the encoder includes a source-channel joint coding module, and the quantization step corresponding to the quantization parameter is the quantization step that follows the source-channel joint coding module.
[0187] The purpose of the quantization step is to convert the continuous signal values output by the source-channel joint coding module into discrete values for subsequent coding and transmission.
[0188] In some embodiments, the quantization parameters and / or dequantization parameters include one or more of the following: quantization step size; quantization bit depth; quantization range; quantization type; quantization module parameters.
[0189] For example, the quantization type can be one of uniform quantization, non-uniform quantization, vector quantization, or scalar quantization; the quantization module parameters include encoder-based model parameters and structure.
[0190] For example, the second indication information indicates the quantization parameters: {quantization step size = 0.1, quantization type = uniform quantization}, and is associated with the source-channel joint coding feature extraction module_v1 (encoder). For the first device, it obtains the PMI value or RI value, obtains the floating-point sequence to be quantized through the source-channel joint coding feature extraction module_v1, and outputs the quantized bits using a quantization step size of 0.1 and uniform quantization.
[0191] Alternatively, the second indication information indicates the quantization parameters: {quantization depth = 8 bits, quantization range = [0, 255]}, and is associated with the encoded image feature extraction module _v2 (encoder). For the first device, the original image is acquired, the floating-point sequence to be quantized is obtained through the encoded image feature extraction module _v2, the floating-point sequence to be quantized is mapped to an integer sequence using 8-bit quantization, and the quantized bits are output.
[0192] Alternatively, the second indication information indicates the quantization parameters: {quantization type = vector quantization}, and associates the Vector Quantized-Variational AutoEncoder (VQ-VAE) encoder_v3 with the VQ-VAE decoder_v3. For the first device, the information of the channel in the angle-delay domain (e.g., a 1024-dimensional vector) is acquired and mapped to a predefined codebook index via the VQ-VAE encoder_v3. Each discrete vector in the codebook corresponds to a unique codebook index, and these codebook indices can be encoded into a fixed-length bit sequence, thereby outputting the bit sequence corresponding to the information of the channel in the angle-delay domain.
[0193] In actual implementation, it may not be implemented exactly as described in the above embodiments. As long as the second indication information can represent the association between the encoder and quantization parameters related to uplink transmission, and / or the association between the decoder and quantization parameters related to uplink transmission, the first device can understand the associated quantization parameters based on the second indication information.
[0194] In some embodiments, the second indication information is transmitted via one or more of DCI signaling, RRC signaling, MAC CE signaling, system information, and broadcast information.
[0195] 1.1.3 Third instruction information;
[0196] In some embodiments, the third indication information includes quantization parameters; or, the third indication information includes an encoder and / or a decoder, and the second indication information and the third indication information are used to jointly determine the quantization parameters; or, the third indication information includes a first association ID and / or a second association ID, and the first indication information, the second indication information and the third indication information are used to jointly determine the quantization parameters.
[0197] The third indication information can directly include quantization parameters, thus enabling the direct determination of these parameters. For example, quantization parameters include quantization step size, quantization depth, quantization range, quantization type, and quantization module parameters; where the quantization type can be one of uniform quantization, non-uniform quantization, vector quantization, or scalar quantization; and the quantization module parameters include encoder-based model parameters and structure.
[0198] Alternatively, if the third indication information includes the encoder, then the second indication information is used to determine the association between the encoder and quantization parameters related to uplink transmission, so that both are used to jointly determine the quantization parameters. For example, if the third indication information indicates that the encoder is a first AI / ML model, then the first device determines the quantization parameters based on the association between the first AI / ML model and the quantization parameters in the second indication information. As another example, if the third indication information indicates that the encoder input includes the feature vector corresponding to the channel, the channel information in the angle-delay domain, PMI, RI, and that the encoder output includes the floating-point sequence to be quantized, then the encoder is indicated to be a first AI / ML model.
[0199] Alternatively, if the third indication information includes a first association ID, then the first indication information is used to determine the association between the encoder related to uplink transmission and the first association ID, and the second indication information is used to determine the association between the encoder related to uplink transmission and the quantization parameters. Thus, all three are used to jointly determine the quantization parameters. For example, if the third indication information indicates that the first association ID is 001, then the first indication information is used to determine that the encoder associated with "001" is the first AI / ML model. The first device then determines the quantization parameters based on the association between the first AI / ML model and the quantization parameters in the second indication information.
[0200] In some embodiments, the third indication information is transmitted via one or more of DCI signaling, RRC signaling, MAC CE signaling, system information, and broadcast information.
[0201] 1.1.4 Fourth instruction information;
[0202] In some embodiments, the first device falls back from the quantization mode associated with uplink transmission to the first quantization mode (also referred to as the basic quantization mode) according to the fourth instruction information.
[0203] When the second device detects a significant deterioration in channel quality related to the first device, such as when the CQI value reported by the first device is lower than a preset threshold (e.g., a CQI value lower than 3), the second device will trigger the first device to fall back to the first quantization mode via the fourth indication information, thereby ensuring the reliability of transmission. For example, the first quantization mode is set to 2-bit scalar uniform quantization, and the quantization parameters include: uniform quantization type, 2-bit quantization depth, and quantization range [-1, 1].
[0204] In some embodiments, the method further includes: switching to a second quantization mode, wherein the second quantization mode is indicated by third indication information;
[0205] The third indication information is sent by the second device when the parameter used to indicate channel quality is higher than the first threshold, where the first threshold is a preset threshold.
[0206] When the channel quality associated with the first device is restored, for example, when the CQI value is higher than 3, the second device re-instructs the first device to switch to a higher precision second quantization mode (e.g., switch to 8-bit vector quantization) through the third indication information, that is, cancel the first quantization mode.
[0207] In some embodiments, the fourth indication information is transmitted via one or more of DCI signaling, RRC signaling, MAC CE signaling, system information, and broadcast information.
[0208] 1.1.5 Fifth instruction information;
[0209] In some embodiments, the association between the encoder and the first associated ID includes the association between the encoder's input and the first associated ID; the association between the decoder and the second associated ID includes the association between the decoder's input and the second associated ID.
[0210] The associated ID can be an identifier or a number, used to associate an encoder or decoder.
[0211] For example, the input of the encoder includes: non-bitstream information sources such as images, videos, and text; the output of the encoder includes: a floating-point number sequence to be quantized; the input of the decoder includes: a dequantized floating-point number sequence; the output of the decoder includes: the recovered non-bitstream information sources such as images, videos, and text.
[0212] Alternatively, the encoder's input includes: a first association ID, non-bitstream information sources such as images, videos, and text; the encoder's output includes: a floating-point sequence to be quantized; the decoder's input includes: a second association ID, a dequantized floating-point sequence; the decoder's output includes: the recovered non-bitstream information sources such as images, videos, and text.
[0213] The input and output of the encoder and decoder may contain other information, which is only an example here.
[0214] In some embodiments, the fifth indication information is transmitted via one or more of DCI signaling, RRC signaling, MAC CE signaling, system information, and broadcast information.
[0215] 1.1.6 Sixth instruction information;
[0216] In some embodiments, the dequantization parameters include one or more of the following: quantization step size; quantization bit depth; quantization range; quantization type; quantization module parameters.
[0217] Dequantization parameters are parameters that correspond to quantization parameters. Dequantization parameters are the same as quantization parameters, or they have a predefined relationship with quantization parameters.
[0218] For example, the sixth indication information indicates the dequantization parameters: {quantization step size = 0.1, quantization type = uniform quantization}, and is associated with the source-channel joint decoding feature extraction module_v1 (decoder). For the first device, it obtains the PMI value or RI value, obtains the floating-point sequence to be quantized through the source-channel joint coding feature extraction module_v1 (encoder), and outputs the quantized bits using a quantization step size of 0.1 and uniform quantization. After receiving the quantized bits, the second device recovers the original PMI value or RI value using the source-channel joint decoding feature extraction module_v1 based on the quantization step size of 0.1.
[0219] Alternatively, the sixth indication information specifies the dequantization parameters: {quantization bit depth = 8 bits, quantization range = [0, 255]}, and associates it with the decoding image feature extraction module _v2 (decoder). For the first device, it acquires the original image, obtains the floating-point sequence to be quantized through the encoding image feature extraction module _v2 (encoder), maps the floating-point sequence to an integer sequence using 8-bit quantization, and outputs the quantized bits. After receiving the quantized bits, the second device uses the decoding image feature extraction module _v2 to perform image reconstruction based on the quantization bit depth and quantization range.
[0220] Alternatively, the sixth indication information specifies the dequantization parameters: {quantization type = vector quantization}, and associates the VQ-VAE encoder_v3 with the VQ-VAE decoder_v3. For the first device, the channel information in the angle-delay domain (e.g., a 1024-dimensional vector) is acquired and mapped to a predefined codebook index using the VQ-VAE encoder_v3. Each discrete vector in the codebook corresponds to a unique codebook index, which can be encoded into a fixed-length bit sequence, thus outputting the bit sequence corresponding to the channel information in the angle-delay domain. The second device receives the bit sequence corresponding to the 1024-dimensional vector and uses the VQ-VAE decoder_v3 and the same codebook to recover the channel information in the angle-delay domain.
[0221] In some embodiments, the sixth indication information is transmitted via one or more of DCI signaling, RRC signaling, MAC CE signaling, system information, and broadcast information.
[0222] 1.1.7 Seventh instruction message;
[0223] In some embodiments, the seventh indication information includes dequantization parameters; or, the seventh indication information includes an encoder and / or a decoder, and the sixth and seventh indication information are used to jointly determine the dequantization parameters; or, the seventh indication information includes a first association ID and / or a second association ID, and the fifth, sixth, and seventh indication information are used to jointly determine the dequantization parameters.
[0224] The seventh indication information can directly include dequantization parameters, thus enabling direct determination of these parameters. For example, dequantization parameters include quantization step size, quantization depth, quantization range, quantization type, and quantization module parameters; wherein, the quantization type can be one of uniform quantization, non-uniform quantization, vector quantization, or scalar quantization; and the quantization module parameters include decoder-based model parameters and structure.
[0225] Alternatively, if the seventh indication information includes the decoder, then the sixth indication information is used to determine the association between the decoder and dequantization parameters related to downlink transmission, so that both are used to jointly determine the dequantization parameters. For example, if the seventh indication information indicates that the decoder is a second AI / ML model, then the first device determines the dequantization parameters based on the association between the second AI / ML model in the sixth indication information and the dequantization parameters. As another example, if the seventh indication information indicates that the decoder's input is a dequantized floating-point sequence and the decoder's output is a recovered image, video, text, or other non-bitstream information source, then the decoder is indicated to be a second AI / ML model.
[0226] Alternatively, if the seventh indication information includes a second association ID, then the fifth indication information is used to determine the association between the decoder related to downlink transmission and the second association ID, and the sixth indication information is used to determine the association between the decoder related to downlink transmission and the dequantization parameters. These three information are then used together to determine the dequantization parameters. For example, if the seventh indication information indicates that the second association ID is 001, then the fifth indication information is used to determine that the decoder associated with "001" is the second AI / ML model. The first device then determines the dequantization parameters based on the association between the second AI / ML model and the dequantization parameters in the sixth indication information.
[0227] In some embodiments, the seventh indication information is transmitted via one or more of DCI signaling, RRC signaling, MAC CE signaling, system information, and broadcast information.
[0228] 1.1.8 Eighth Instruction Message;
[0229] In some embodiments, the first device reverts from a quantization mode related to downlink transmission to a first quantization mode (also referred to as the basic quantization mode) according to the eighth indication information.
[0230] When the second device detects a significant deterioration in the channel quality related to the first device, such as when the CQI value reported by the first device is lower than a preset threshold (e.g., when the CQI value is lower than 3), the second device will trigger the first device to fall back to the first quantization mode through the eighth indication information, thereby ensuring the reliability of transmission.
[0231] In some embodiments, the method further includes: switching to a second quantization mode, the second quantization mode being indicated by a seventh indication information;
[0232] The seventh indication information is sent by the second device when the parameter used to indicate channel quality is higher than the first threshold, which is a preset threshold.
[0233] The second quantization method can be the same as or different from the second quantization method indicated in 1.1.4. This application embodiment does not limit this.
[0234] In some embodiments, the eighth indication information is transmitted via one or more of DCI signaling, RRC signaling, MAC CE signaling, system information, and broadcast information.
[0235] 1.2 Primary capability information;
[0236] In some embodiments, the method further includes: sending first capability information to a second device, the first capability information indicating the capability of the first device, the capability of the first device being related to quantization and / or dequantization between the first device and the second device.
[0237] In some embodiments, the capabilities of the first device include one or more of the following: supported encoders and version numbers; supported quantization method types; supported quantization parameter configuration capabilities; whether it supports switching to the first quantization method; the ability to bind encoders and quantization parameters; the current resource status of the first device; and whether it supports receiving and processing first indication information to eighth indication information.
[0238] For example, taking the encoder as the first AI / ML model, the capabilities of the first device may include model name, architecture type, corresponding training set, etc.; it may also include quantization step size, quantization bit depth, quantization range, quantization type, etc.; it may also include the current computing power and power consumption of the first device, etc., which are not limited in this embodiment of the application.
[0239] In some embodiments, the first capability information includes first reporting information and / or second reporting information, wherein the first reporting information is reported according to frequency band dimension, carrier dimension, or UE dimension, and the second reporting information is reported according to serving cell dimension.
[0240] 1.2.1 First reported information;
[0241] In some embodiments, the first capability information includes first reporting information, which is reported according to frequency band dimension, carrier dimension, or UE dimension.
[0242] In some embodiments, the first reporting information is reported by frequency band; or, the first reporting information is reported by combination of frequency bands; or, the first reporting information is reported by each frequency band in the combination of frequency bands; or, the first reporting information is reported by each carrier in each frequency band in the combination of frequency bands; or, the first reporting information is reported by frequency band range; or, the first reporting information is reported by UE.
[0243] The first reported information is reported per band, indicating that the capabilities corresponding to different frequency bands can be reported independently; the first reported information is reported per band combination, indicating that the capabilities corresponding to different frequency band combinations can be reported independently; the first reported information is reported per band per band combination, indicating that the capabilities corresponding to different frequency bands within different frequency band combinations can be reported independently; the first reported information is reported per component carrier per band per band combination, indicating that the capabilities corresponding to different carriers within different frequency bands within different frequency band combinations can be reported independently; the first reported information is reported per frequency range (per FR), indicating that the capabilities corresponding to different FRs can be reported independently, for example, the capabilities corresponding to FR1 and FR2 can be reported independently; the first reported information is reported per UE (per UE), indicating that if the UE reports a capability, then that capability is supported on all frequency bands.
[0244] The instructions or configurations of the second device need to match the capabilities reported by the first device. For example, after receiving the first report from the first device, the second device can instruct the first device to support the joint source-channel coding process with quantization and dequantization configurations within its capabilities.
[0245] In some embodiments, the first reporting information is transmitted via one or more of UE capability signaling, RRC signaling, and MAC CE signaling.
[0246] 1.2.2 Second reported information;
[0247] In some embodiments, the first capability information includes second reporting information, which is reported according to the serving cell dimension.
[0248] Optionally, the second reporting information does not carry serving cell indication information, and the second reporting information is applicable to the serving cell currently transmitting the second reporting information; or, the second reporting information is applicable to one or more serving cells currently associated with the first device.
[0249] The second reported information does not carry serving cell indication information, that is, it does not indicate that the second reported information applies to a specific serving cell or certain serving cells. In this case, the second reported information applies to the serving cell currently transmitting the second reported information; or it applies to one or more serving cells currently associated with the first device, such as all serving cells currently associated with the first device.
[0250] Optionally, the second reported information carries serving cell indication information, and the second reported information is applicable to the serving cell indicated by the serving cell indication information.
[0251] For example, the serving cell indication information indicates one serving cell (cell A), or the serving cell indication information indicates multiple serving cells (3 cells, cell A, cell B and cell C).
[0252] The instructions or configurations of the second device need to match the capabilities reported by the first device. For example, after receiving the second reporting information from the first device, the second device can provide the first device with quantization and dequantization configurations within its capabilities to support the joint source-channel coding process.
[0253] In some embodiments, the second reporting information is transmitted via one or more of UE assistance information, RRC signaling, and MAC CE signaling.
[0254] In this application, the relevant embodiments in 1.1 (including 1.1.1 to 1.1.8) and 1.2 (including 1.2.1 and 1.2.2) can be implemented individually or in combination. For example, the first device only receives the first information sent by the second device and does not send the first capability information to the second device. The first information includes the first indication information to the eighth indication information. Or, for another example, the first device sends the first capability information to the second device. The first capability information includes the first reporting information. The first device receives the first information sent by the second device. The first information includes the first indication information to the eighth indication information. The embodiments of this application do not limit this.
[0255] In summary, the method provided in this embodiment achieves information consistency between the first device and the second device for quantization and / or dequantization by receiving first information sent by the second device. The first information is related to quantization and / or dequantization between the first device and the second device. Quantization is related to the encoder, and dequantization is related to the decoder.
[0256] The method provided in this embodiment also determines one or more of the following by providing a variety of first information: quantization parameters, dequantization parameters, encoder, decoder, and first quantization mode. This facilitates parameter alignment between the first device and the second device, thereby improving encoding and decoding performance.
[0257] The method provided in this embodiment also switches to a second quantization mode, which is indicated by a third indication information or a seventh indication information. The third indication information or the seventh indication information is sent by the second device when the parameter used to indicate the channel quality is higher than a first threshold. The first threshold is a preset threshold, which enables the first device to switch to a more suitable quantization mode when the channel quality is good, thereby improving coding performance.
[0258] The method provided in this embodiment also sends first capability information to the second device. The first capability information is used to indicate the capability of the first device. The capability of the first device is related to the quantization and / or dequantization between the first device and the second device, so that the second device can understand the capability of the first device and send first information matching the capability of the first device to the first device.
[0259] Figure 9 illustrates a flowchart of a quantitative information transmission method provided in an exemplary embodiment of this application. The method is executed by a second device and includes:
[0260] Step 910: Send the first information to the first device.
[0261] Among them, the first information is related to the quantization and / or dequantization between the first device and the second device, the quantization is related to the encoder, and the dequantization is related to the decoder.
[0262] By way of example and not limitation, the encoder is one or more of the following: source-channel joint coding method, source-channel coding method, channel coding method, first model, first AI / ML model, encoder end, coding function, coding model; the decoder is one or more of the following: source-channel joint decoding method, source-channel decoding method, channel decoding method, second model, second AI / ML model, decoder end, decoding function, decoding model.
[0263] In some embodiments, the combination of encoder and decoder includes one or more of the following:
[0264] The encoder is a source-channel joint coding method, and the decoder is a source-channel joint decoding method; the encoder is a source-channel coding method, and the decoder is a source-channel decoding method; the encoder is a channel coding method, and the decoder is a channel decoding method; the encoder is a first model, and the decoder is a second model; the encoder is a first AI / ML model, and the decoder is a second AI / ML model; the encoder is the encoding end, and the decoder is the decoding end; the encoder is the encoding function, and the decoder is the decoding function; the encoder is the encoding model, and the decoder is the decoding model.
[0265] In some embodiments, the first device is a terminal device and the second device is a network device; or, the first device is a network device and the second device is a terminal device; or the first device is a first terminal device and the second device is a second terminal device.
[0266] In some embodiments, the first information is associated with CSI transmission.
[0267] In other words, the quantization information transmission method is associated with CSI transmission, and the encoder and decoder are applied to CSI transmission.
[0268] For specific implementation details, please refer to step 810 of the embodiment in Figure 8, which will not be repeated here.
[0269] 2.1 First instruction information to eighth instruction information;
[0270] In some embodiments, the first information includes one or more of the following: first indication information; second indication information; third indication information; fourth indication information; fifth indication information; sixth indication information; seventh indication information; eighth indication information;
[0271] The first indication information is used to determine the association between the encoder related to uplink transmission and the first associated identifier ID, and / or the association between the decoder related to uplink transmission and the second associated ID, where the first associated ID identifies the encoder and the second associated ID identifies the decoder; the second indication information is used to determine the association between the encoder related to uplink transmission and the quantization parameters, and / or the association between the decoder related to uplink transmission and the quantization parameters; the third indication information is used to determine the quantization parameters related to uplink transmission; the fourth indication information is used to indicate that the quantization mode related to uplink transmission be reverted to the first quantization mode, where the first quantization mode is a preset quantization mode; the fifth indication information is used to determine the association between the encoder related to downlink transmission and the first associated ID, and / or the association between the decoder related to downlink transmission and the second associated ID; the sixth indication information is used to determine the association between the encoder related to downlink transmission and the dequantization parameters, and / or the association between the decoder related to downlink transmission and the dequantization parameters, where the dequantization parameters correspond to the quantization parameters; the seventh indication information is used to determine the dequantization parameters related to downlink transmission; and the eighth indication information is used to indicate that the quantization mode related to downlink transmission be reverted to the first quantization mode.
[0272] For specific implementation details, please refer to section 1.1 of the embodiment shown in Figure 8, which will not be repeated here.
[0273] 2.1.1 First instruction information;
[0274] In some embodiments, the association between the encoder and the first associated ID includes the association between the encoder's input and the first associated ID;
[0275] The association between the decoder and the second associated ID, including the association between the decoder's input and the second associated ID.
[0276] In some embodiments, the encoder input includes one or more of the following: a first association ID, a feature vector corresponding to the channel, information about the channel in the angle-delay domain, PMI, and RI; the encoder output includes: a sequence of floating-point numbers to be quantized; the decoder input includes one or more of the following: a second association ID, a dequantized sequence of floating-point numbers; the decoder output includes one or more of the following: a feature vector corresponding to the channel, information about the channel in the angle-delay domain, PMI, and RI; or,
[0277] The encoder input includes one or more of the following: a first association ID and a non-bitstream source; the encoder output includes: a sequence of floating-point numbers to be quantized; the decoder input includes one or more of the following: a second association ID and a dequantized sequence of floating-point numbers; the decoder output includes: a non-bitstream source.
[0278] For specific implementation details, please refer to section 1.1.1 of the embodiment shown in Figure 8, which will not be repeated here.
[0279] 2.1.2 Second instruction information;
[0280] In some embodiments, the quantization parameters and / or dequantization parameters include one or more of the following: quantization step size; quantization bit depth; quantization range; quantization type; quantization module parameters.
[0281] For specific implementation details, please refer to section 1.1.2 of the embodiment shown in Figure 8, which will not be repeated here.
[0282] 2.1.3 Third instruction information;
[0283] In some embodiments, the third indication information includes quantization parameters; or, the third indication information includes an encoder and / or a decoder, and the second indication information and the third indication information are used to jointly determine the quantization parameters; or, the third indication information includes a first association ID and / or a second association ID, and the first indication information, the second indication information and the third indication information are used to jointly determine the quantization parameters.
[0284] For specific implementation details, please refer to section 1.1.3 of the embodiment shown in Figure 8, which will not be repeated here.
[0285] 2.1.4 Fourth instruction information;
[0286] In some embodiments, the method further includes: sending a third instruction signal;
[0287] The third indication information is used to instruct the first device to switch to the second quantization mode. The third indication information is sent by the second device when the parameter used to indicate channel quality is higher than the first threshold, and the first threshold is a preset threshold.
[0288] For specific implementation details, please refer to section 1.1.4 of the embodiment shown in Figure 8, which will not be repeated here.
[0289] 2.1.5 Fifth instruction information;
[0290] For specific implementation details, please refer to section 1.1.5 of the embodiment shown in Figure 8, which will not be repeated here.
[0291] 2.1.6 Sixth instruction information;
[0292] For specific implementation details, please refer to section 1.1.6 of the embodiment shown in Figure 8, which will not be repeated here.
[0293] 2.1.7 Seventh instruction information;
[0294] In some embodiments, the seventh indication information includes dequantization parameters; or, the seventh indication information includes an encoder and / or a decoder, and the sixth and seventh indication information are used to jointly determine the dequantization parameters; or, the seventh indication information includes a first association ID and / or a second association ID, and the fifth, sixth, and seventh indication information are used to jointly determine the dequantization parameters.
[0295] For specific implementation details, please refer to section 1.1.7 of the embodiment shown in Figure 8, which will not be repeated here.
[0296] 2.1.8 Eighth Instruction Message;
[0297] In some embodiments, the method further includes: sending a seventh indication message;
[0298] The seventh indication information is used to instruct the first device to switch to the second quantization mode. The seventh indication information is sent by the second device when the parameter used to indicate channel quality is higher than the first threshold, and the first threshold is a preset threshold.
[0299] In some embodiments, the first to eighth indication information is transmitted via one or more of DCI signaling, RRC signaling, MAC CE signaling, system information, and broadcast information.
[0300] For specific implementation details, please refer to section 1.1.8 of the embodiment shown in Figure 8, which will not be repeated here.
[0301] 2.2 Primary capability information;
[0302] In some embodiments, the method further includes: receiving first capability information sent by a first device, the first capability information indicating the capability of the first device, the capability of the first device being related to quantization and / or dequantization between the first device and the second device.
[0303] In some embodiments, the first capability information includes first reporting information and / or second reporting information, wherein the first reporting information is reported according to frequency band dimension, carrier dimension, or UE dimension, and the second reporting information is reported according to serving cell dimension.
[0304] For specific implementation details, please refer to section 1.2 of the embodiment shown in Figure 8, which will not be repeated here.
[0305] 2.2.1 First reported information;
[0306] In some embodiments, the first capability information includes first reporting information, which is reported according to frequency band dimension, carrier dimension, or UE dimension.
[0307] As an example and not a limitation, the first reported information is reported by frequency band; or, the first reported information is reported by combination of frequency bands; or, the first reported information is reported by each frequency band in the combination of frequency bands; or, the first reported information is reported by each carrier in each frequency band in the combination of frequency bands; or, the first reported information is reported by frequency band range; or, the first reported information is reported by UE.
[0308] In some embodiments, the first reporting information is transmitted via one or more of UE capability signaling, RRC signaling, and MAC CE signaling.
[0309] For specific implementation details, please refer to section 1.2.1 of the embodiment shown in Figure 8, which will not be repeated here.
[0310] 2.2.2 Second reported information;
[0311] In some embodiments, the first capability information includes second reporting information, which is reported according to the serving cell dimension.
[0312] Optionally, the second reporting information does not carry serving cell indication information, and the second reporting information is applicable to the serving cell currently transmitting the second reporting information; or, the second reporting information is applicable to one or more serving cells currently associated with the first device.
[0313] Optionally, the second reported information carries serving cell indication information, and the second reported information is applicable to the serving cell indicated by the serving cell indication information.
[0314] In some embodiments, the second reporting information is transmitted via one or more of UE auxiliary information, RRC signaling, and MAC CE signaling.
[0315] In some embodiments, the capabilities of the first device include one or more of the following: supported encoders and version numbers; supported quantization method types; supported quantization parameter configuration capabilities; whether it supports switching to the first quantization method; the ability to bind encoders and quantization parameters; the current resource status of the first device; and whether it supports receiving and processing first indication information to eighth indication information.
[0316] For specific implementation details, please refer to section 1.2.2 of the embodiment shown in Figure 8, which will not be repeated here.
[0317] In this application, the relevant embodiments in 2.1 (including 2.1.1 to 2.1.8) and 2.2 (including 2.2.1 and 2.2.2) can be implemented individually or in combination. For example, the second device sends first information to the first device, the first information including first indication information to eighth indication information. If the first device does not send first capability information, the second device naturally does not receive the first capability information. Or, for example, the second device receives first capability information, the first capability information including first reporting information, and the second device sends first information to the first device, the first information including first indication information to eighth indication information. The embodiments of this application do not limit this.
[0318] In summary, the method provided in this embodiment achieves information consistency between the first device and the second device for quantization and / or dequantization by sending first information to the first device. The first information is related to quantization and / or dequantization between the first device and the second device. Quantization is related to the encoder, and dequantization is related to the decoder.
[0319] The method provided in this embodiment also determines one or more of the following by providing a variety of first information: quantization parameters, dequantization parameters, encoder, decoder, and first quantization mode. This facilitates parameter alignment between the first device and the second device, thereby improving encoding and decoding performance.
[0320] The method provided in this embodiment also sends a third indication message or a seventh indication message; wherein the third indication message or the seventh indication message is used to instruct the first device to switch to the second quantization mode. The third indication message or the seventh indication message is sent by the second device when the parameter used to indicate the channel quality is higher than a first threshold. The first threshold is a preset threshold, which enables the first device to switch to a more suitable quantization mode when the channel quality is better, thereby improving the coding performance.
[0321] The method provided in this embodiment also receives first capability information sent by a first device. The first capability information is used to indicate the capability of the first device. The capability of the first device is related to the quantization and / or dequantization between the first device and the second device, so that the second device can understand the capability of the first device and send first information matching the capability of the first device to the first device.
[0322] In the above embodiments, the embodiments corresponding to FIG8 and FIG9 can be implemented individually or in combination, and this application does not limit them.
[0323] Figure 10 illustrates a flowchart of a quantitative information transmission and reception method provided in an exemplary embodiment of this application. The method is executed by a first device 110 and a second device 120, and includes at least one of the following steps:
[0324] Step 1010: The first device 110 sends the first capability information to the second device 120; correspondingly, the second device 120 receives the first capability information sent by the first device 110.
[0325] The first capability information is used to indicate the capability of the first device 110, and the capability of the first device 110 is related to the quantization and / or dequantization between the first device 110 and the second device 120.
[0326] In some embodiments, the capabilities of the first device 110 include one or more of the following: supported encoders and version numbers; supported quantization method types; supported quantization parameter configuration capabilities; whether it supports switching to the first quantization mode; the ability to bind encoders and quantization parameters; the current resource status of the first device 110; and whether it supports receiving and processing first indication information to eighth indication information.
[0327] For example, the first device 110 supports models such as the first decoding model 111 and the second encoding model 112. Correspondingly, the second device 120 supports models such as the first encoding model 121 and the second decoding model 122.
[0328] For specific implementation details of the first to eighth instruction messages, please refer to the embodiment in Figure 8, which will not be repeated here.
[0329] Step 1020: The second device 120 sends first information to the first device 110 (the first information includes first indication information to third indication information); correspondingly, the first device 110 receives the first information sent by the second device 120.
[0330] The first information is related to the quantization and / or dequantization between the first device 110 and the second device 120. Quantization is related to the encoder, and dequantization is related to the decoder. In this embodiment, the encoder is used as the encoding model and the decoder is used as the decoding model for illustration.
[0331] For example, the first information includes first indication information to third indication information. The third indication information indicates that the first associated ID is 001. Then the first indication information is used to determine that the encoder associated with "001" is the first encoding model 121 and the decoder associated with "001" is the first decoding model 111. Then the first device 110 determines the quantization parameters according to the association relationship between the model and the quantization parameters in the second indication information.
[0332] Step 1030: The first device 110 sends a first CSI-RS to the second device 120; correspondingly, the second device 120 receives the first CSI-RS sent by the first device 110.
[0333] The first device 110 enables the second device 120 to estimate the channel quality of the uplink transmission channel by sending a reference signal (first CSI-RS) to the second device 120.
[0334] Step 1040: The second device 120 measures and encodes the first CSI-RS.
[0335] After receiving the first CSI-RS, the second device 120 measures the first CSI-RS and encodes it using the first coding model 121 to obtain first coding information related to the channel quality of the uplink transmission channel.
[0336] Step 1050: The second device 120 sends the first encoded information to the first device 110; correspondingly, the first device 110 receives the first encoded information sent by the second device 120.
[0337] After the second device 120 obtains the first encoded information, it sends the first encoded information to the first device 110.
[0338] Step 1060: The first device 110 decodes the first encoded information.
[0339] After receiving the first encoded information, the first device 110 uses the first decoding model 111 to decode the first encoded information and obtain the feedback information of the second device 120 to the first CSI-RS.
[0340] Step 1070: The second device 120 sends first information to the first device 110 (the first information includes the fifth to the seventh indication information); correspondingly, the first device 110 receives the first information sent by the second device 120.
[0341] For example, the first information includes the fifth to the seventh indication information. The seventh indication information indicates that the second associated ID is 002. Then the fifth indication information is used to determine that the encoder associated with "002" is the second encoding model 112 and the decoder associated with "002" is the second decoding model 122. Then the first device 110 determines the dequantization parameters according to the association between the model and the dequantization parameters in the sixth indication information.
[0342] Step 1080: The second device 120 sends a second CSI-RS to the first device 110; correspondingly, the first device 110 receives the second CSI-RS sent by the second device 120.
[0343] The second device 120 enables the first device 110 to estimate the channel quality of the downlink transmission channel by sending a reference signal (second CSI-RS) to the first device 110.
[0344] Step 1090: The first device 110 measures and encodes the second CSI-RS.
[0345] After receiving the second CSI-RS, the first device 110 measures the second CSI-RS and encodes it using the second coding model 112 to obtain second coding information related to the channel quality of the downlink transmission channel.
[0346] Step 10100: The first device 110 sends the second encoded information to the second device 120; correspondingly, the second device 120 receives the second encoded information sent by the first device 110.
[0347] After the first device 110 obtains the second encoded information, it sends the second encoded information to the second device 120.
[0348] Step 10110: The second device 120 decodes the second encoded information.
[0349] After receiving the second encoded information, the second device 120 uses the second decoding model 122 to decode the second encoded information to obtain the feedback information of the first device 110 to the second CSI-RS.
[0350] Step 10120: The second device 120 sends first information (which includes fourth indication information) to the first device 110; correspondingly, the first device 110 receives the first information sent by the second device 120.
[0351] When the second device 120 detects a significant deterioration in the channel quality related to the first device 110, for example, when the CQI value reported by the first device 110 is lower than a preset threshold, such as when the CQI value is lower than 3, the second device 120 will trigger the first device 110 to switch to the first quantization mode through the fourth indication information, thereby ensuring the reliability of transmission. For example, the first quantization mode is set to 2-bit scalar uniform quantization, and the quantization parameters include: quantization type is uniform quantization, quantization bit depth is 2 bits, and quantization range is [-1, 1].
[0352] Optionally, the first information includes the eighth instruction information. For specific implementation details, please refer to the embodiment in Figure 8. It will not be repeated here, but only the example of the first information including the fourth instruction information will be used for explanation.
[0353] Step 10130: The first device 110 switches to the first quantization mode.
[0354] For example, the first device 110 switches to 2-bit scalar uniform quantization according to the fourth instruction information.
[0355] Step 10140: The second device 120 sends first information (which includes third indication information) to the first device 110; correspondingly, the first device 110 receives the first information sent by the second device 120.
[0356] When the channel quality associated with the first device 110 is restored, for example, when the CQI value is higher than 3, the second device 120 re-instructs the first device 110 to switch to a higher precision second quantization mode (e.g., switch to 8-bit vector quantization) through the third indication information, that is, to cancel the first quantization mode.
[0357] Optionally, the first information includes the seventh instruction information. For specific implementation details, please refer to the embodiment in Figure 8. It will not be repeated here, but only the example of the first information including the third instruction information will be used for explanation.
[0358] Step 10150: The first device 110 switches to the second quantization mode.
[0359] For example, the first device 110 switches to 8-bit vector quantization according to the third instruction information.
[0360] Figure 11 shows a block diagram of a first device provided in an exemplary embodiment of this application. The device can be implemented as a first device or as part of a first device by software or hardware or a combination of both. The device includes:
[0361] The receiving module 1110 is used to receive first information sent by the second device. The first information is related to the quantization and / or dequantization between the first device and the second device. The quantization is related to the encoder, and the dequantization is related to the decoder.
[0362] In one possible design of this embodiment, the encoder is one or more of the following: source-channel joint coding method, source-channel coding method, channel coding method, first model, first AI / ML model, encoding end, coding function, coding model; the decoder is one or more of the following: source-channel joint decoding method, source-channel decoding method, channel decoding method, second model, second AI / ML model, decoding end, decoding function, decoding model.
[0363] In one possible design of this embodiment, the combination of encoder and decoder includes one or more of the following:
[0364] The encoder is a source-channel joint coding method, and the decoder is a source-channel joint decoding method; the encoder is a source-channel coding method, and the decoder is a source-channel decoding method; the encoder is a channel coding method, and the decoder is a channel decoding method; the encoder is a first model, and the decoder is a second model; the encoder is a first AI / ML model, and the decoder is a second AI / ML model; the encoder is the encoding end, and the decoder is the decoding end; the encoder is the encoding function, and the decoder is the decoding function; the encoder is the encoding model, and the decoder is the decoding model.
[0365] In one possible design of this embodiment, the first device is a terminal device and the second device is a network device; or, the first device is a network device and the second device is a terminal device; or the first device is a first terminal device and the second device is a second terminal device.
[0366] In one possible design of this embodiment, the first information is associated with CSI transmission.
[0367] For specific implementation details, please refer to step 810 of the embodiment in Figure 8, which will not be repeated here.
[0368] 3.1 First to eighth instruction messages;
[0369] In one possible design of this embodiment, the first information includes one or more of the following: first indication information; second indication information; third indication information; fourth indication information; fifth indication information; sixth indication information; seventh indication information; eighth indication information;
[0370] The first indication information is used to determine the association between the encoder related to uplink transmission and the first associated ID, and / or the association between the decoder related to uplink transmission and the second associated ID, where the first associated ID is used to identify the encoder and the second associated ID is used to identify the decoder; the second indication information is used to determine the association between the encoder related to uplink transmission and the quantization parameters, and / or the association between the decoder related to uplink transmission and the quantization parameters; the third indication information is used to determine the quantization parameters related to uplink transmission; the fourth indication information is used to indicate that the quantization mode related to uplink transmission be reverted to the first quantization mode, where the first quantization mode is a preset quantization mode; the fifth indication information is used to determine the association between the encoder related to downlink transmission and the first associated ID, and / or the association between the decoder related to downlink transmission and the second associated ID; the sixth indication information is used to determine the association between the encoder related to downlink transmission and the dequantization parameters, and / or the association between the decoder related to downlink transmission and the dequantization parameters, where the dequantization parameters correspond to the quantization parameters; the seventh indication information is used to determine the dequantization parameters related to downlink transmission; and the eighth indication information is used to indicate that the quantization mode related to downlink transmission be reverted to the first quantization mode.
[0371] For specific implementation details, please refer to section 1.1 of the embodiment shown in Figure 8, which will not be repeated here.
[0372] 3.1.1 First instruction information;
[0373] In one possible design of this embodiment, the association between the encoder and the first associated ID includes the association between the encoder's input and the first associated ID; the association between the decoder and the second associated ID includes the association between the decoder's input and the second associated ID.
[0374] In one possible design of this embodiment, the encoder input includes one or more of the following: a first association ID, a feature vector corresponding to the channel, channel information in the angle-delay domain, PMI, and RI; the encoder output includes: a sequence of floating-point numbers to be quantized; the decoder input includes one or more of the following: a second association ID, a dequantized sequence of floating-point numbers; the decoder output includes one or more of the following: a feature vector corresponding to the channel, channel information in the angle-delay domain, PMI, and RI; or,
[0375] The encoder input includes one or more of the following: a first association ID and a non-bitstream source; the encoder output includes: a sequence of floating-point numbers to be quantized; the decoder input includes one or more of the following: a second association ID and a dequantized sequence of floating-point numbers; the decoder output includes: a non-bitstream source.
[0376] For specific implementation details, please refer to section 1.1.1 of the embodiment shown in Figure 8, which will not be repeated here.
[0377] 3.1.2 Second instruction information;
[0378] In one possible design of this embodiment, the quantization parameters and / or dequantization parameters include one or more of the following: quantization step size; quantization bit depth; quantization range; quantization type; quantization module parameters.
[0379] For specific implementation details, please refer to section 1.1.2 of the embodiment shown in Figure 8, which will not be repeated here.
[0380] 3.1.3 Third instruction information;
[0381] In one possible design of this embodiment, the third indication information includes quantization parameters; or, the third indication information includes an encoder and / or a decoder, and the second indication information and the third indication information are used to jointly determine the quantization parameters; or, the third indication information includes a first association ID and / or a second association ID, and the first indication information, the second indication information and the third indication information are used to jointly determine the quantization parameters.
[0382] For specific implementation details, please refer to section 1.1.3 of the embodiment shown in Figure 8, which will not be repeated here.
[0383] 3.1.4 Fourth instruction information;
[0384] In one possible design of this embodiment, the processing module 1120 is used to switch to the second quantization mode, which is indicated by the third indication information;
[0385] The third indication information is sent by the second device when the parameter used to indicate channel quality is higher than the first threshold, where the first threshold is a preset threshold.
[0386] For specific implementation details, please refer to section 1.1.4 of the embodiment shown in Figure 8, which will not be repeated here.
[0387] 3.1.5 Fifth instruction information;
[0388] For specific implementation details, please refer to section 1.1.5 of the embodiment shown in Figure 8, which will not be repeated here.
[0389] 3.1.6 Sixth instruction information;
[0390] For specific implementation details, please refer to section 1.1.6 of the embodiment shown in Figure 8, which will not be repeated here.
[0391] 3.1.7 Seventh instruction information;
[0392] In one possible design of this embodiment, the seventh indication information includes dequantization parameters; or, the seventh indication information includes an encoder and / or a decoder, and the sixth and seventh indication information are used to jointly determine the dequantization parameters; or, the seventh indication information includes a first association ID and / or a second association ID, and the fifth, sixth, and seventh indication information are used to jointly determine the dequantization parameters.
[0393] For specific implementation details, please refer to section 1.1.7 of the embodiment shown in Figure 8, which will not be repeated here.
[0394] 3.1.8 Eighth instruction information;
[0395] In one possible design of this embodiment, the processing module 1120 is used to switch to the second quantization mode, which is indicated by the seventh indication information;
[0396] The seventh indication information is sent by the second device when the parameter used to indicate channel quality is higher than the first threshold, where the first threshold is a preset threshold.
[0397] In one possible design of this embodiment, the first to eighth indication information are transmitted through one or more of DCI signaling, RRC signaling, MAC CE signaling, system information, and broadcast information.
[0398] For specific implementation details, please refer to section 1.1.8 of the embodiment shown in Figure 8, which will not be repeated here.
[0399] 3.2 Primary capability information;
[0400] In one possible design of this embodiment, the sending module 1130 is used to send first capability information to the second device. The first capability information is used to indicate the capability of the first device. The capability of the first device is related to the quantization and / or dequantization between the first device and the second device.
[0401] In one possible design of this embodiment, the first capability information includes first reporting information and / or second reporting information. The first reporting information is reported according to the frequency band dimension, carrier dimension, or UE dimension, and the second reporting information is reported according to the serving cell dimension.
[0402] For specific implementation details, please refer to section 1.2 of the embodiment shown in Figure 8, which will not be repeated here.
[0403] 3.2.1 First reported information;
[0404] In one possible design of this embodiment, the first capability information includes first reporting information, which is reported according to frequency band dimension, carrier dimension, or UE dimension.
[0405] In one possible design of this embodiment, the first reporting information is reported according to frequency band; or, the first reporting information is reported according to frequency band combination; or, the first reporting information is reported according to each frequency band in the frequency band combination; or, the first reporting information is reported according to each carrier in each frequency band in the frequency band combination; or, the first reporting information is reported according to frequency band range; or, the first reporting information is reported according to UE.
[0406] In one possible design of this embodiment, the first reporting information is transmitted through one or more of UE capability signaling, RRC signaling, and MAC CE signaling.
[0407] For specific implementation details, please refer to section 1.2.1 of the embodiment shown in Figure 8, which will not be repeated here.
[0408] 3.2.2 Second reported information;
[0409] In one possible design of this embodiment, the first capability information includes the second reporting information, which is reported according to the serving cell dimension.
[0410] Optionally, the second reporting information does not carry serving cell indication information, and the second reporting information is applicable to the serving cell currently transmitting the second reporting information; or, the second reporting information is applicable to one or more serving cells currently associated with the first device.
[0411] Optionally, the second reported information carries serving cell indication information, and the second reported information is applicable to the serving cell indicated by the serving cell indication information.
[0412] In one possible design of this embodiment, the second reporting information is transmitted through one or more of UE auxiliary information, RRC signaling, and MAC CE signaling.
[0413] In one possible design of this embodiment, the capabilities of the first device include one or more of the following: supported encoders and version numbers; supported quantization method types; supported quantization parameter configuration capabilities; whether it supports switching to the first quantization mode; the ability to bind encoders and quantization parameters; the current resource status of the first device; and whether it supports receiving and processing the first instruction information to the eighth instruction information.
[0414] For specific implementation details, please refer to section 1.2.2 of the embodiment shown in Figure 8, which will not be repeated here.
[0415] This embodiment uses a receiving module 1110, a processing module 1120 and a sending module 1130 as an example for illustration. The number of receiving modules 1110, processing modules 1120 and sending modules 1130 is not limited.
[0416] For a description of the function of the receiving module 1110, please refer to step 810 in the embodiment shown in Figure 8. For a description of the function of the processing module 1120, please refer to step 810 in the embodiment shown in Figure 8. For a description of the function of the sending module 1130, please refer to step 810 in the embodiment shown in Figure 8.
[0417] Figure 12 shows a block diagram of a second device provided in an exemplary embodiment of this application. The device can be implemented as a second device, or as part of a second device, by software or hardware, or a combination of both. The device includes:
[0418] The transmitting module 1210 is used to transmit first information to the first device. The first information is related to the quantization and / or dequantization between the first device and the second device. The quantization is related to the encoder, and the dequantization is related to the decoder.
[0419] In one possible design of this embodiment, the encoder is one or more of the following: source-channel joint coding method, source-channel coding method, channel coding method, first model, first AI / ML model, encoding end, coding function, coding model; the decoder is one or more of the following: source-channel joint decoding method, source-channel decoding method, channel decoding method, second model, second AI / ML model, decoding end, decoding function, decoding model.
[0420] In one possible design of this embodiment, the combination of encoder and decoder includes one or more of the following:
[0421] The encoder is a source-channel joint coding method, and the decoder is a source-channel joint decoding method; the encoder is a source-channel coding method, and the decoder is a source-channel decoding method; the encoder is a channel coding method, and the decoder is a channel decoding method; the encoder is a first model, and the decoder is a second model; the encoder is a first AI / ML model, and the decoder is a second AI / ML model; the encoder is the encoding end, and the decoder is the decoding end; the encoder is the encoding function, and the decoder is the decoding function; the encoder is the encoding model, and the decoder is the decoding model.
[0422] In one possible design of this embodiment, the first device is a terminal device and the second device is a network device; or, the first device is a network device and the second device is a terminal device; or the first device is a first terminal device and the second device is a second terminal device.
[0423] In one possible design of this embodiment, the first information is associated with CSI transmission.
[0424] For specific implementation details, please refer to step 810 of the embodiment in Figure 8, which will not be repeated here.
[0425] 4.1 First instruction information to eighth instruction information;
[0426] In one possible design of this embodiment, the first information includes one or more of the following: first indication information; second indication information; third indication information; fourth indication information; fifth indication information; sixth indication information; seventh indication information; eighth indication information;
[0427] The first indication information is used to determine the association between the encoder related to uplink transmission and the first associated identifier ID, and / or the association between the decoder related to uplink transmission and the second associated ID, where the first associated ID identifies the encoder and the second associated ID identifies the decoder; the second indication information is used to determine the association between the encoder related to uplink transmission and the quantization parameters, and / or the association between the decoder related to uplink transmission and the quantization parameters; the third indication information is used to determine the quantization parameters related to uplink transmission; the fourth indication information is used to indicate that the quantization mode related to uplink transmission be reverted to the first quantization mode, where the first quantization mode is a preset quantization mode; the fifth indication information is used to determine the association between the encoder related to downlink transmission and the first associated ID, and / or the association between the decoder related to downlink transmission and the second associated ID; the sixth indication information is used to determine the association between the encoder related to downlink transmission and the dequantization parameters, and / or the association between the decoder related to downlink transmission and the dequantization parameters, where the dequantization parameters correspond to the quantization parameters; the seventh indication information is used to determine the dequantization parameters related to downlink transmission; and the eighth indication information is used to indicate that the quantization mode related to downlink transmission be reverted to the first quantization mode.
[0428] For specific implementation details, please refer to section 1.1 of the embodiment shown in Figure 8, which will not be repeated here.
[0429] 4.1.1 First instruction information;
[0430] In one possible design of this embodiment, the association between the encoder and the first associated ID includes the association between the encoder's input and the first associated ID; the association between the decoder and the second associated ID includes the association between the decoder's input and the second associated ID.
[0431] In one possible design of this embodiment, the encoder input includes one or more of the following: a first association ID, a feature vector corresponding to the channel, channel information in the angle-delay domain, PMI, and RI; the encoder output includes: a sequence of floating-point numbers to be quantized; the decoder input includes one or more of the following: a second association ID, a dequantized sequence of floating-point numbers; the decoder output includes one or more of the following: a feature vector corresponding to the channel, channel information in the angle-delay domain, PMI, and RI; or,
[0432] The encoder input includes one or more of the following: a first association ID and a non-bitstream source; the encoder output includes: a sequence of floating-point numbers to be quantized; the decoder input includes one or more of the following: a second association ID and a dequantized sequence of floating-point numbers; the decoder output includes: a non-bitstream source.
[0433] For specific implementation details, please refer to section 1.1.1 of the embodiment shown in Figure 8, which will not be repeated here.
[0434] 4.1.2 Second instruction information;
[0435] In one possible design of this embodiment, the quantization parameters and / or dequantization parameters include one or more of the following: quantization step size; quantization bit depth; quantization range; quantization type; quantization module parameters.
[0436] For specific implementation details, please refer to section 1.1.2 of the embodiment shown in Figure 8, which will not be repeated here.
[0437] 4.1.3 Third instruction information;
[0438] In one possible design of this embodiment, the third indication information includes quantization parameters; or, the third indication information includes an encoder and / or a decoder, and the second indication information and the third indication information are used to jointly determine the quantization parameters; or, the third indication information includes a first association ID and / or a second association ID, and the first indication information, the second indication information and the third indication information are used to jointly determine the quantization parameters.
[0439] For specific implementation details, please refer to section 1.1.3 of the embodiment shown in Figure 8, which will not be repeated here.
[0440] 4.1.4 Fourth instruction information;
[0441] In one possible design of this embodiment, the sending module 1210 is used to send third indication information;
[0442] The third indication information is used to instruct the first device to switch to the second quantization mode. The third indication information is sent by the second device when the parameter used to indicate channel quality is higher than the first threshold, and the first threshold is a preset threshold.
[0443] For specific implementation details, please refer to section 1.1.4 of the embodiment shown in Figure 8, which will not be repeated here.
[0444] 4.1.5 Fifth instruction information;
[0445] For specific implementation details, please refer to section 1.1.5 of the embodiment shown in Figure 8, which will not be repeated here.
[0446] 4.1.6 Sixth instruction information;
[0447] For specific implementation details, please refer to section 1.1.6 of the embodiment shown in Figure 8, which will not be repeated here.
[0448] 4.1.7 Seventh instruction message;
[0449] In one possible design of this embodiment, the seventh indication information includes dequantization parameters; or, the seventh indication information includes an encoder and / or a decoder, and the sixth and seventh indication information are used to jointly determine the dequantization parameters; or, the seventh indication information includes a first association ID and / or a second association ID, and the fifth, sixth, and seventh indication information are used to jointly determine the dequantization parameters.
[0450] For specific implementation details, please refer to section 1.1.7 of the embodiment shown in Figure 8, which will not be repeated here.
[0451] 4.1.8 Eighth Instruction Message;
[0452] In one possible design of this embodiment, the sending module 1210 is used to send the seventh indication information;
[0453] The seventh indication information is used to instruct the first device to switch to the second quantization mode. The seventh indication information is sent by the second device when the parameter used to indicate channel quality is higher than the first threshold, and the first threshold is a preset threshold.
[0454] In one possible design of this embodiment, the first to eighth indication information are transmitted through one or more of DCI signaling, RRC signaling, MAC CE signaling, system information, and broadcast information.
[0455] For specific implementation details, please refer to section 1.1.8 of the embodiment shown in Figure 8, which will not be repeated here.
[0456] 4.2 Primary Capability Information;
[0457] In one possible design of this embodiment, the receiving module 1220 is used to receive first capability information sent by the first device. The first capability information is used to indicate the capability of the first device, and the capability of the first device is related to the quantization and / or dequantization between the first device and the second device.
[0458] In one possible design of this embodiment, the first capability information includes first reporting information and / or second reporting information. The first reporting information is reported according to the frequency band dimension, carrier dimension, or UE dimension, and the second reporting information is reported according to the serving cell dimension.
[0459] For specific implementation details, please refer to section 1.2 of the embodiment shown in Figure 8, which will not be repeated here.
[0460] 4.2.1 First reported information;
[0461] In one possible design of this embodiment, the first capability information includes first reporting information, which is reported according to frequency band dimension, carrier dimension, or UE dimension.
[0462] As an example and not a limitation, the first reported information is reported by frequency band; or, the first reported information is reported by combination of frequency bands; or, the first reported information is reported by each frequency band in the combination of frequency bands; or, the first reported information is reported by each carrier in each frequency band in the combination of frequency bands; or, the first reported information is reported by frequency band range; or, the first reported information is reported by UE.
[0463] In one possible design of this embodiment, the first reporting information is transmitted through one or more of UE capability signaling, RRC signaling, and MAC CE signaling.
[0464] For specific implementation details, please refer to section 1.2.1 of the embodiment shown in Figure 8, which will not be repeated here.
[0465] 4.2.2 Second reported information;
[0466] In one possible design of this embodiment, the first capability information includes the second reporting information, which is reported according to the serving cell dimension.
[0467] Optionally, the second reporting information does not carry serving cell indication information, and the second reporting information is applicable to the serving cell currently transmitting the second reporting information; or, the second reporting information is applicable to one or more serving cells currently associated with the first device.
[0468] Optionally, the second reported information carries serving cell indication information, and the second reported information is applicable to the serving cell indicated by the serving cell indication information.
[0469] In one possible design of this embodiment, the second reporting information is transmitted through one or more of UE auxiliary information, RRC signaling, and MAC CE signaling.
[0470] In one possible design of this embodiment, the capabilities of the first device include one or more of the following: supported encoders and version numbers; supported quantization method types; supported quantization parameter configuration capabilities; whether it supports switching to the first quantization mode; the ability to bind encoders and quantization parameters; the current resource status of the first device; and whether it supports receiving and processing the first instruction information to the eighth instruction information.
[0471] For specific implementation details, please refer to section 1.2.2 of the embodiment shown in Figure 8, which will not be repeated here.
[0472] This embodiment uses one transmitting module 1210 and one receiving module 1220 as an example for illustration, and the number of transmitting modules 1210 and receiving modules 1220 is not limited.
[0473] For a description of the function of the sending module 1210, please refer to step 910 in the embodiment shown in Figure 9. For a description of the function of the receiving module 1220, please refer to step 910 in the embodiment shown in Figure 9.
[0474] Figure 13 shows a schematic diagram of the structure of a first device provided in an exemplary embodiment of this application. The first device 1300 may include a processor 1301, a transceiver 1302, and a memory 1303. The processor 1301 can be used to control transmission and / or reception, such as to implement the functions of the processing module 1120 described above. The transceiver 1302 can be used to implement transmission and / or reception functions, such as to implement the functions of at least one of the receiving module 1110 and the transmitting module 1130 described above.
[0475] The processor 1301 includes one or more processing cores. The processor 1301 executes various functional applications and information processing by running software programs and modules.
[0476] The transceiver 1302 may include a receiver and a transmitter. For example, the receiver and transmitter may be implemented as the same wireless communication component, which may include a wireless communication chip and a radio frequency antenna.
[0477] Transceiver 1302 is used to receive first information sent by a second device. The first information is related to quantization and / or dequantization between the first device and the second device. Quantization is related to the encoder, and dequantization is related to the decoder.
[0478] The memory 1303 can be connected to the processor 1301 and the transceiver 1302.
[0479] The memory 1303 can be used to store a computer program executed by the processor, and the processor 1301 is used to execute the computer program to implement the various steps in the above method embodiments.
[0480] Furthermore, the memory 1303 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, including but not limited to: magnetic disks or optical disks, electrically erasable programmable read-only memory, erasable programmable read-only memory, static on-demand memory, read-only memory, magnetic memory, flash memory, and programmable read-only memory.
[0481] For details not described in this embodiment, please refer to the method-side embodiment above, which will not be repeated here.
[0482] Figure 14 shows a schematic diagram of the structure of a second device provided in an exemplary embodiment of this application. The second device 1400 may include a processor 1401, a transceiver 1402, and a memory 1403. The processor 1401 can be used to control transmission and / or reception. The transceiver 1402 can be used to implement transmission and / or reception functions, such as implementing the functions of at least one of the aforementioned transmission module 1210 and reception module 1220.
[0483] The processor 1401 includes one or more processing cores, and the processor 1401 executes various functional applications and information processing by running software programs and modules.
[0484] Transceiver 1402 may include a receiver and a transmitter. For example, transceiver 1402 may include a wired communication component, which may include a wired communication chip and a wired interface (such as a fiber optic interface). Optionally, transceiver 1402 may also include a wireless communication component, which may include a wireless communication chip and a radio frequency antenna.
[0485] Transceiver 1402 is used to send first information to a first device. The first information is related to quantization and / or dequantization between the first device and the second device. Quantization is related to the encoder, and dequantization is related to the decoder.
[0486] The memory 1403 can be connected to the processor 1401 and the transceiver 1402.
[0487] The memory 1403 can be used to store a computer program executed by the processor, and the processor 1401 is used to execute the computer program to implement the various steps in the above method embodiments.
[0488] Furthermore, the memory 1403 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, including but not limited to: magnetic disks or optical disks, electrically erasable programmable read-only memory, erasable programmable read-only memory, static on-demand memory, read-only memory, magnetic memory, flash memory, and programmable read-only memory.
[0489] For details not described in this embodiment, please refer to the method-side embodiment above, which will not be repeated here.
[0490] This application also provides a computer-readable storage medium storing a computer program for execution by a processor to implement the above-described quantized information receiving method and / or quantized information transmitting method. In some embodiments, the computer-readable storage medium may include: read-only memory (ROM), random-access memory (RAM), solid-state drives (SSDs), or optical discs, etc. The random-access memory may include resistive random-access memory (ReRAM) and dynamic random-access memory (DRAM).
[0491] This application also provides a chip, which includes programmable logic circuits and / or program instructions. When the chip is running on a first device, it is used to implement the above-described quantization information receiving method on the first device side.
[0492] This application also provides a chip, which includes programmable logic circuits and / or program instructions. When the chip is running on a second device, it is used to implement the above-described method for transmitting quantized information on the second device side.
[0493] This application embodiment also provides a first chip, which includes programmable logic circuits and / or program instructions. When the first chip is running on a first device, it is used to "receive first information sent by a second device, wherein the first information is related to quantization and / or dequantization between the first device and the second device, the quantization is related to the encoder, and the dequantization is related to the decoder".
[0494] This application embodiment also provides a second chip, which includes programmable logic circuits and / or program instructions. When the second chip is running on a second device, it is used to "send first information to a first device, wherein the first information is related to quantization and / or dequantization between the first device and the second device, the quantization is related to the encoder, and the dequantization is related to the decoder".
[0495] This application also provides a computer program product, which includes a computer program stored in a computer-readable storage medium. A processor reads from the computer-readable storage medium and executes the computer program to implement the above-described quantization information receiving method and / or quantization information sending method.
[0496] It should be understood that the term "instruction" mentioned in the embodiments of this application can be a direct instruction, an indirect instruction, or an indication of a relationship. For example, A instructing B can mean that A directly instructs B, such as B being able to obtain information through A; it can also mean that A indirectly instructs B, such as A instructing C, so B can obtain information through C; or it can mean that there is a relationship between A and B.
[0497] In the description of the embodiments of this application, the term "correspondence" may indicate that there is a direct or indirect correspondence between two things, or that there is an association between two things, or that there is a relationship of instruction and being instructed, configuration and being configured, etc.
[0498] In some embodiments of this application, "predefined" can be implemented by pre-storing corresponding codes, tables, or other means that can be used to indicate relevant information in the device, and this application does not limit the specific implementation method. For example, predefined can refer to what is defined in the protocol.
[0499] In some embodiments of this application, "protocol" may refer to standard protocols in the field of communications, such as LTE protocol, NR protocol and related protocols applied to future communication systems, and this application does not limit it.
[0500] In this article, "multiple" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. The character " / " generally indicates that the preceding and following related objects have an "or" relationship.
[0501] In this article, "greater than or equal to" can mean greater than or equal to, and "less than or equal to" can mean less than or equal to.
[0502] Furthermore, the step numbers described herein are merely illustrative of one possible execution order between steps. In some other embodiments, the steps may not be executed in the order of their numbers, such as two steps with different numbers being executed simultaneously, or two steps with different numbers being executed in the reverse order of the illustration. This application does not limit this.
[0503] Those skilled in the art will recognize that the functions described in the embodiments of this application in one or more of the above examples can be implemented using hardware, software, firmware, or any combination thereof. When implemented using software, these functions can be stored in a computer-readable medium or transmitted as one or more instructions or code on a computer-readable medium. Computer-readable media include computer storage media and communication media, wherein communication media include any medium that facilitates the transfer of a computer program from one place to another. Storage media can be any available medium that can be accessed by a general-purpose or special-purpose computer.
[0504] The above are merely exemplary embodiments of this application and are not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application shall be included within the protection scope of this application.
Claims
1. A method for receiving quantized information, characterized in that, The method is performed by a first device, and the method includes: The device receives first information sent by a second device, the first information being related to quantization and / or dequantization between the first device and the second device, the quantization being related to the encoder and the dequantization being related to the decoder.
2. The method according to claim 1, characterized in that, The combination of the encoder and the decoder includes one or more of the following: The encoder is a source-channel joint coding method, and the decoder is a source-channel joint decoding method; the encoder is a source-channel coding method, and the decoder is a source-channel decoding method; the encoder is a channel coding method, and the decoder is a channel decoding method; the encoder is a first model, and the decoder is a second model; the encoder is a first artificial intelligence (AI) / machine learning (ML) model, and the decoder is a second AI / ML model; the encoder is an encoding end, and the decoder is a decoding end; the encoder is an encoding function, and the decoder is a decoding function; the encoder is an encoding model, and the decoder is a decoding model.
3. The method according to claim 1 or 2, characterized in that, The first information includes one or more of the following: First instruction; Second instruction; Third instruction; Fourth instruction; Fifth instruction; Sixth instruction; Seventh instruction; Eighth instruction message; The first indication information is used to determine the association between the encoder related to uplink transmission and the first associated identifier ID, and / or the association between the decoder related to uplink transmission and the second associated ID, wherein the first associated ID is used to identify the encoder and the second associated ID is used to identify the decoder; the second indication information is used to determine the association between the encoder related to uplink transmission and the quantization parameter, and / or the association between the decoder related to uplink transmission and the quantization parameter; the third indication information is used to determine the quantization parameter related to uplink transmission; the fourth indication information is used to indicate that the quantization mode related to uplink transmission be reverted to the first quantization mode, wherein the first quantization mode is a preset quantization mode; the fifth indication information is used to determine the association between the encoder related to downlink transmission and the first associated ID, and / or the association between the decoder related to downlink transmission and the second associated ID; the sixth indication information is used to determine the association between the encoder related to downlink transmission and the dequantization parameter, and / or the association between the decoder related to downlink transmission and the dequantization parameter, wherein the dequantization parameter corresponds to the quantization parameter; The seventh indication information is used to determine the dequantization parameters related to downlink transmission; The eighth indication information is used to indicate that the quantization mode related to downlink transmission should be reverted to the first quantization mode.
4. The method according to claim 3, characterized in that, The association between the encoder and the first associated ID includes the association between the encoder's input and the first associated ID; the association between the decoder and the second associated ID includes the association between the decoder's input and the second associated ID.
5. The method according to claim 4, characterized in that, The encoder's input includes one or more of the following: the first association ID, the feature vector corresponding to the channel, the channel information in the angle-delay domain, the precoding matrix indicator (PMI), and the rank indicator (RI); the encoder's output includes: the floating-point sequence to be quantized; the decoder's input includes one or more of the following: the second association ID, and the dequantized floating-point sequence; the decoder's output includes one or more of the following: the feature vector corresponding to the channel, the channel information in the angle-delay domain, the PMI, and the RI; or, The input to the encoder includes one or more of the following: the first associated ID, and a non-bitstream source; the output of the encoder includes: the floating-point sequence to be quantized; the input to the decoder includes one or more of the following: the second associated ID, and the dequantized floating-point sequence; the output of the decoder includes: the non-bitstream source.
6. The method according to any one of claims 3 to 5, characterized in that, The quantization parameters and / or the dequantization parameters include one or more of the following: quantization step size; quantization bit depth; quantization range; quantization type; quantization module parameters.
7. The method according to claim 3 or 6, characterized in that, The third indication information includes the quantization parameter; or, the third indication information includes the encoder and / or the decoder, and the second indication information and the third indication information are used to jointly determine the quantization parameter; or, the third indication information includes the first association ID and / or the second association ID, and the first indication information, the second indication information and the third indication information are used to jointly determine the quantization parameter.
8. The method according to claim 3, characterized in that, The method further includes: Switch to the second quantization mode, which is indicated by the third indication information or the seventh indication information; The third indication information or the seventh indication information is sent by the second device when the parameter used to indicate channel quality is higher than a first threshold, where the first threshold is a preset threshold.
9. The method according to claim 3 or 6, characterized in that, The seventh indication information includes the dequantization parameter; or, the seventh indication information includes the encoder and / or the decoder, and the sixth indication information and the seventh indication information are used to jointly determine the dequantization parameter; or, the seventh indication information includes the first association ID and / or the second association ID, and the fifth indication information, the sixth indication information and the seventh indication information are used to jointly determine the dequantization parameter.
10. The method according to any one of claims 3 to 9, characterized in that, The first indication information to the eighth indication information are transmitted through one or more of the following: downlink control information (DCI) signaling, radio resource control (RRC) signaling, media access control unit (MAC CE) signaling, system information, and broadcast information.
11. The method according to any one of claims 3 to 10, characterized in that, The method further includes: Send first capability information to the second device, the first capability information being used to indicate the capability of the first device, the capability of the first device being related to the quantization and / or the dequantization between the first device and the second device.
12. The method according to claim 11, characterized in that, The first capability information includes first reporting information, which is reported according to frequency band dimension, carrier dimension, or user equipment (UE) dimension.
13. The method according to claim 12, characterized in that, The first reported information is reported by frequency band; or, the first reported information is reported by combination of frequency bands; or, the first reported information is reported by each frequency band in the combination of frequency bands; or, the first reported information is reported by each carrier in each frequency band in the combination of frequency bands; or, the first reported information is reported by frequency band range; or, the first reported information is reported by UE.
14. The method according to claim 12 or 13, characterized in that, The first reported information is transmitted through one or more of the following: UE capability signaling, RRC signaling, and MAC CE signaling.
15. The method according to claim 11, characterized in that, The first capability information includes the second reported information, which is reported according to the serving cell dimension.
16. The method according to claim 15, characterized in that, The second reporting information does not carry serving cell indication information, and the second reporting information is applicable to the serving cell currently transmitting the second reporting information; or, the second reporting information is applicable to one or more serving cells currently associated with the first device.
17. The method according to claim 15, characterized in that, The second reported information carries serving cell indication information, and the second reported information is applicable to the serving cell indicated by the serving cell indication information.
18. The method according to any one of claims 15 to 17, characterized in that, The second reported information is transmitted through one or more of the following: UE auxiliary information, RRC signaling, and MAC CE signaling.
19. The method according to any one of claims 11 to 18, characterized in that, The capabilities of the first device include one or more of the following: Supported encoders and version numbers; supported quantization method types; supported quantization parameter configuration capabilities; whether switching to the first quantization method is supported; and the binding capability between the encoder and the quantization parameters. The current resource status of the first device; Whether it supports receiving and processing the first instruction information to the eighth instruction information.
20. The method according to any one of claims 1 to 19, characterized in that, The first device is a terminal device, and the second device is a network device; or, the first device is the network device, and the second device is the terminal device; or the first device is a first terminal device, and the second device is a second terminal device.
21. The method according to any one of claims 1 to 20, characterized in that, The first information is associated with the Channel State Information (CSI) transmission.
22. A method for transmitting quantitative information, characterized in that, The method is performed by a second device, and the method includes: Send first information to a first device, the first information being related to quantization and / or dequantization between the first device and the second device, the quantization being related to the encoder and the dequantization being related to the decoder.
23. The method according to claim 22, characterized in that, The combination of the encoder and the decoder includes one or more of the following: The encoder is a source-channel joint coding method, and the decoder is a source-channel joint decoding method; the encoder is a source-channel coding method, and the decoder is a source-channel decoding method; the encoder is a channel coding method, and the decoder is a channel decoding method; the encoder is a first model, and the decoder is a second model; the encoder is a first artificial intelligence (AI) / machine learning (ML) model, and the decoder is a second AI / ML model; the encoder is an encoding end, and the decoder is a decoding end; the encoder is an encoding function, and the decoder is a decoding function; the encoder is an encoding model, and the decoder is a decoding model.
24. The method according to claim 22 or 23, characterized in that, The first information includes one or more of the following: First instruction; Second instruction; Third instruction; Fourth instruction; Fifth instruction; Sixth instruction; Seventh instruction; Eighth instruction message; The first indication information is used to determine the association between the encoder related to uplink transmission and the first associated identifier ID, and / or the association between the decoder related to uplink transmission and the second associated ID, wherein the first associated ID is used to identify the encoder and the second associated ID is used to identify the decoder; the second indication information is used to determine the association between the encoder related to uplink transmission and the quantization parameter, and / or the association between the decoder related to uplink transmission and the quantization parameter; the third indication information is used to determine the quantization parameter related to uplink transmission; the fourth indication information is used to indicate that the quantization mode related to uplink transmission be reverted to the first quantization mode, wherein the first quantization mode is a preset quantization mode; the fifth indication information is used to determine the association between the encoder related to downlink transmission and the first associated ID, and / or the association between the decoder related to downlink transmission and the second associated ID; the sixth indication information is used to determine the association between the encoder related to downlink transmission and the dequantization parameter, and / or the association between the decoder related to downlink transmission and the dequantization parameter, wherein the dequantization parameter corresponds to the quantization parameter; The seventh indication information is used to determine the dequantization parameters related to downlink transmission; The eighth indication information is used to indicate that the quantization mode related to downlink transmission should be reverted to the first quantization mode.
25. The method according to claim 24, characterized in that, The association between the encoder and the first associated ID includes the association between the encoder's input and the first associated ID; the association between the decoder and the second associated ID includes the association between the decoder's input and the second associated ID.
26. The method according to claim 25, characterized in that, The encoder's input includes one or more of the following: the first association ID, the feature vector corresponding to the channel, the channel information in the angle-delay domain, the precoding matrix indicator (PMI), and the rank indicator (RI); the encoder's output includes: the floating-point sequence to be quantized; the decoder's input includes one or more of the following: the second association ID, and the dequantized floating-point sequence; the decoder's output includes one or more of the following: the feature vector corresponding to the channel, the channel information in the angle-delay domain, the PMI, and the RI; or, The input to the encoder includes one or more of the following: the first associated ID, and a non-bitstream source; the output of the encoder includes: the floating-point sequence to be quantized; the input to the decoder includes one or more of the following: the second associated ID, and the dequantized floating-point sequence; the output of the decoder includes: the non-bitstream source.
27. The method according to any one of claims 24 to 26, characterized in that, The quantization parameters and / or the dequantization parameters include one or more of the following: quantization step size; quantization bit depth; quantization range; quantization type; quantization module parameters.
28. The method according to claim 24 or 27, characterized in that, The third indication information includes the quantization parameter; or, the third indication information includes the encoder and / or the decoder, and the second indication information and the third indication information are used to jointly determine the quantization parameter; or, the third indication information includes the first association ID and / or the second association ID, and the first indication information, the second indication information and the third indication information are used to jointly determine the quantization parameter.
29. The method according to claim 24, characterized in that, The method further includes: Send the third instruction message or the seventh instruction message; The third indication information or the seventh indication information is used to instruct the first device to switch to the second quantization mode. The third indication information or the seventh indication information is sent by the second device when the parameter used to indicate channel quality is higher than a first threshold, and the first threshold is a preset threshold.
30. The method according to claim 24 or 27, characterized in that, The seventh indication information includes the dequantization parameter; or, the seventh indication information includes the encoder and / or the decoder, and the sixth indication information and the seventh indication information are used to jointly determine the dequantization parameter; or, the seventh indication information includes the first association ID and / or the second association ID, and the fifth indication information, the sixth indication information and the seventh indication information are used to jointly determine the dequantization parameter.
31. The method according to any one of claims 24 to 30, characterized in that, The first indication information to the eighth indication information are transmitted through one or more of the following: downlink control information (DCI) signaling, radio resource control (RRC) signaling, media access control unit (MAC CE) signaling, system information, and broadcast information.
32. The method according to any one of claims 24 to 31, characterized in that, The method further includes: The first capability information sent by the first device is received. The first capability information is used to indicate the capability of the first device. The capability of the first device is related to the quantization and / or the dequantization between the first device and the second device.
33. The method according to claim 32, characterized in that, The first capability information includes first reporting information, which is reported according to frequency band dimension, carrier dimension, or user equipment (UE) dimension.
34. The method according to claim 33, characterized in that, The first reported information is reported by frequency band; or, the first reported information is reported by combination of frequency bands; or, the first reported information is reported by each frequency band in the combination of frequency bands; or, the first reported information is reported by each carrier in each frequency band in the combination of frequency bands; or, the first reported information is reported by frequency band range; or, the first reported information is reported by UE.
35. The method according to claim 33 or 34, characterized in that, The first reported information is transmitted through one or more of the following: UE capability signaling, RRC signaling, and MAC CE signaling.
36. The method according to claim 32, characterized in that, The first capability information includes the second reported information, which is reported according to the serving cell dimension.
37. The method according to claim 36, characterized in that, The second reporting information does not carry serving cell indication information, and the second reporting information is applicable to the serving cell currently transmitting the second reporting information; or, the second reporting information is applicable to one or more serving cells currently associated with the first device.
38. The method according to claim 36, characterized in that, The second reported information carries serving cell indication information, and the second reported information is applicable to the serving cell indicated by the serving cell indication information.
39. The method according to any one of claims 36 to 38, characterized in that, The second reported information is transmitted through one or more of the following: UE auxiliary information, RRC signaling, and MAC CE signaling.
40. The method according to any one of claims 32 to 39, characterized in that, The capabilities of the first device include one or more of the following: Supported encoders and version numbers; supported quantization method types; supported quantization parameter configuration capabilities; whether switching to the first quantization method is supported; and the binding capability between the encoder and the quantization parameters. The current resource status of the first device; Whether it supports receiving and processing the first instruction information to the eighth instruction information.
41. The method according to any one of claims 22 to 40, characterized in that, The first device is a terminal device, and the second device is a network device; or, the first device is the network device, and the second device is the terminal device; or the first device is a first terminal device, and the second device is a second terminal device.
42. The method according to any one of claims 22 to 41, characterized in that, The first information is associated with the Channel State Information (CSI) transmission.
43. A first device, characterized in that, The first device includes: A receiving module is configured to receive first information sent by a second device, the first information being related to quantization and / or dequantization between the first device and the second device, wherein the quantization is related to an encoder and the dequantization is related to a decoder.
44. A second device, characterized in that, The second device includes: A sending module is configured to send first information to a first device, the first information being related to quantization and / or dequantization between the first device and the second device, wherein the quantization is related to an encoder and the dequantization is related to a decoder.
45. A first device, characterized in that, The first device includes: A processor; a transceiver connected to the processor; a memory for storing executable instructions of the processor; wherein the processor is configured to load and execute the executable instructions to implement the quantized information receiving method as described in any one of claims 1 to 21.
46. A second device, characterized in that, The second device includes: A processor; a transceiver connected to the processor; a memory for storing executable instructions of the processor; wherein the processor is configured to load and execute the executable instructions to implement the quantized information transmission method as described in any one of claims 22 to 42.
47. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores at least one program, which is loaded and executed by a processor to implement the quantized information receiving method as described in any one of claims 1 to 21; and / or the quantized information transmitting method as described in any one of claims 22 to 42.
48. A chip, characterized in that, The chip includes programmable logic circuits and / or program instructions. When the chip is running on a first device, it is used to implement the quantized information receiving method according to any one of claims 1 to 21; when the chip is running on a second device, it is used to implement the quantized information transmitting method according to any one of claims 22 to 42.
49. A computer program product, characterized in that, The computer program product includes computer instructions stored in a computer-readable storage medium, a processor retrieves the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions to implement the quantized information receiving method as described in any one of claims 1 to 21. And / or the quantization information transmission method according to any one of claims 22 to 42.
