WO2024007957A1 - Csi反馈的方法、装置、设备及可读存储介质 - Google Patents

Csi反馈的方法、装置、设备及可读存储介质 Download PDF

Info

Publication number
WO2024007957A1
WO2024007957A1 PCT/CN2023/103970 CN2023103970W WO2024007957A1 WO 2024007957 A1 WO2024007957 A1 WO 2024007957A1 CN 2023103970 W CN2023103970 W CN 2023103970W WO 2024007957 A1 WO2024007957 A1 WO 2024007957A1
Authority
WO
WIPO (PCT)
Prior art keywords
csi
reference signal
reply
terminal
correlation
Prior art date
Application number
PCT/CN2023/103970
Other languages
English (en)
French (fr)
Inventor
温子睿
李刚
韩双锋
Original Assignee
中国移动通信有限公司研究院
中国移动通信集团有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 中国移动通信有限公司研究院, 中国移动通信集团有限公司 filed Critical 中国移动通信有限公司研究院
Publication of WO2024007957A1 publication Critical patent/WO2024007957A1/zh

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04BTRANSMISSION
    • H04B7/00Radio transmission systems, i.e. using radiation field
    • H04B7/02Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas
    • H04B7/04Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas
    • H04B7/06Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station
    • H04B7/0613Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station using simultaneous transmission
    • H04B7/0615Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station using simultaneous transmission of weighted versions of same signal
    • H04B7/0619Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station using simultaneous transmission of weighted versions of same signal using feedback from receiving side
    • H04B7/0621Feedback content
    • H04B7/0626Channel coefficients, e.g. channel state information [CSI]
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04BTRANSMISSION
    • H04B7/00Radio transmission systems, i.e. using radiation field
    • H04B7/02Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas
    • H04B7/04Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas
    • H04B7/0413MIMO systems
    • H04B7/0456Selection of precoding matrices or codebooks, e.g. using matrices antenna weighting
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04BTRANSMISSION
    • H04B7/00Radio transmission systems, i.e. using radiation field
    • H04B7/02Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas
    • H04B7/04Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas
    • H04B7/06Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station

Definitions

  • the embodiments of the present application relate to the field of communication technology, and specifically relate to a method, device, equipment and readable storage medium for channel state information (Channel State Information, CSI) feedback.
  • CSI Channel State Information
  • the channel prediction scheme in the related art only performs channel prediction on the network side to reduce the CSI feedback frequency of the terminal.
  • the real-time performance of the current actual CSI acquisition on the network side is reduced. If the channel changes drastically, This will lead to a significant decline in network-side prediction performance.
  • Embodiments of the present application provide a CSI feedback method, device, equipment and readable storage medium to solve the technical problem of how to reduce the frequency of CSI feedback while ensuring the accuracy of CSI prediction.
  • a method for CSI feedback is provided, applied to a terminal, and the method includes:
  • first CSI and second CSI which are obtained by the terminal performing measurement processing based on the first reference signal and the second reference signal respectively;
  • the first reply is used to indicate that the terminal has received to the second reference signal without feeding back the second CSI.
  • obtaining the first CSI and the second CSI includes:
  • the first CSI is obtained according to the plurality of third CSIs, and the first CSI is CSI in the delay Doppler domain;
  • the second CSI is obtained according to the plurality of fourth CSIs, and the second CSI is CSI in the delay Doppler domain.
  • the first reference signal includes a first CSI-RS group
  • the second reference signal includes a second CSI-RS group.
  • the second reference signal is sent by the network device according to the first beamforming
  • the first beamforming is determined by the network device according to the precoding matrix
  • the precoding matrix is the The device generates the fifth CSI based on the fifth CSI, which is predicted by the network device based on the first CSI and the AI model.
  • obtaining the first CSI according to the plurality of third CSIs includes:
  • Obtaining the second CSI according to the plurality of fourth CSIs includes:
  • performing a channel comparison according to the first CSI and the second CSI, and determining to feed back the second CSI or the first reply includes:
  • the correlation threshold configured by the network device is obtained.
  • the correlation threshold represents the minimum CSI delay Doppler spectrum correlation that the network device can accept according to the service quality requirements.
  • the method also includes:
  • the second CSI is sent to the network device.
  • a CSI feedback method applied to network equipment, and the method includes:
  • the first CSI is obtained by the terminal according to the first reference signal measurement processing
  • the first reply or the second CSI is fed back by the terminal based on comparison of the first CSI and the second CSI.
  • the second CSI is measured and processed by the terminal based on the second reference signal. It is obtained that the first reply is used to indicate that the terminal has received the second reference signal and does not need to feed back the second CSI.
  • sending a second reference signal includes:
  • the first CSI and AI model perform CSI prediction to obtain the predicted fifth CSI
  • the second reference signal is transmitted through the first beamforming.
  • the method further includes:
  • a third reference signal is transmitted through the second beamforming.
  • the method further includes:
  • the second CSI and AI model perform CSI prediction to obtain the predicted seventh CSI
  • a fourth reference signal is transmitted through the third beamforming.
  • the method also includes:
  • a device for CSI feedback applied to terminals, including:
  • a first acquisition module configured to acquire first CSI and second CSI, which are obtained by the terminal through measurement processing based on the first reference signal and the second reference signal respectively;
  • a first determination module configured to perform channel comparison according to the first CSI and the second CSI, and determine to feed back the second CSI or a first reply, where the first reply is used to indicate that the terminal has received the first reply.
  • two reference signals and there is no need to feed back the second CSI.
  • the fourth aspect provides a device for CSI feedback, applied to network equipment, including:
  • the second sending module is used to send the first reference signal
  • a first receiving module configured to receive first CSI, where the first CSI is obtained by the terminal based on measurement processing of the first reference signal;
  • the third sending module is used to send the second reference signal
  • the second receiving module is configured to receive a first reply or a second CSI.
  • the first reply or the second CSI is fed back by the terminal based on comparison of the first CSI and the second CSI.
  • the second CSI is fed back by the terminal based on the first CSI and the second CSI. Obtained from the second reference signal measurement process, the first reply is used to indicate that the terminal has received the second reference signal and does not need to feed back the second CSI.
  • a communication device including a processor, a memory, and a program or instructions stored on the memory and executable on the processor.
  • the program or instructions When the program or instructions are executed by the processor, the following is implemented: The steps of the method described in the first aspect or the second aspect.
  • a readable storage medium In a sixth aspect, a readable storage medium is provided. Programs or instructions are stored on the readable storage medium. When the programs or instructions are executed by a processor, the steps of the method described in the first aspect or the second aspect are implemented.
  • the terminal acquires the first CSI and the second CSI, which are obtained by the terminal through measurement processing based on the first reference signal and the second reference signal respectively; and then the terminal obtains the first CSI and the second CSI according to
  • the first CSI and the second CSI are compared to determine whether to feed back the second CSI or the first reply, where the comparison result is used to indicate whether the network device is capable of predicting the second CSI. If the comparison result indicates that the network device is not capable of predicting second CSI, the terminal feeds back the second CSI to the network device. If the comparison result indicates that the network device is capable of predicting the second CSI, the terminal does not feed back the second CSI to the network device, thereby reducing the error while ensuring the accuracy of CSI prediction.
  • Small CSI feedback frequency Rate is used to indicate whether the network device is capable of predicting the second CSI.
  • Figure 1 is a schematic diagram of the codebook evolution process
  • Figure 2 is a schematic diagram of time domain channel compression feedback
  • Figure 3 is a schematic diagram of a communication system provided by an embodiment of the present application.
  • Figure 4 is one of the flow charts of a CSI feedback method provided by an embodiment of the present application.
  • Figure 5 is a second flowchart of a CSI feedback method provided by an embodiment of the present application.
  • Figure 6a is one of the schematic diagrams of channel feedback time domain dynamic compression provided by the embodiment of the present application.
  • Figure 6b is the second schematic diagram of channel feedback time domain dynamic compression provided by the embodiment of the present application.
  • Figure 7 is a schematic diagram of the LSTM network unit structure
  • Figure 8 is a schematic diagram of the prediction method based on LSTM network
  • Figure 9 is one of the schematic diagrams of a CSI feedback device provided by an embodiment of the present application.
  • Figure 10 is a second schematic diagram of a CSI feedback device provided by an embodiment of the present application.
  • Figure 11 is a schematic diagram of a communication device provided by an embodiment of the present application.
  • the precoding vector of each layer of the Type 1 codebook is passed through a two-dimensional discrete Fourier Transform (DFT) vector And its phase rotation quantization, the quantization accuracy is not high.
  • DFT discrete Fourier Transform
  • Type I terminal processing is the simplest and has the lowest overhead, so it is considered the most basic Precoding Matrix Indicator (PMI) feedback codebook in NR, which is NR Required technology for terminals;
  • PMI Precoding Matrix Indicator
  • Type II Type II
  • the PMI fed back by the terminal includes the information of the basis vector, the amplitude and phase information of the weighting coefficient.
  • multipaths with strong energy contribution usually have sparse characteristics, that is, they only occupy a small subspace in the entire N-dimensional space. Therefore, the number of basis vectors required to represent the subspace where these multipaths are located can be greatly reduced, thereby achieving the purpose of compression.
  • the evolution from Type I to Type II is based on selecting basis vectors in the spatial domain for compression, while eType II further selects basis vectors in the frequency subband based on Type II to achieve compression in the frequency domain.
  • channel compression As shown in Figure 2, time domain correlation is used for channel feedback compression. Among them, high update rate refers to high update rate, low update rate w/o Doppler information refers to low update rate w/o Doppler information, and low update rate w/Doppler information refers to low update rate w/Doppler information. .
  • the wireless communication system may include: a network device 31 and a terminal 32, and the terminal 32 may communicate with the network device 31 (transmit signaling or data).
  • the connection between the above devices can be wireless Connection, in order to conveniently and intuitively represent the connection relationship between various devices, solid lines are used in Figure 3.
  • the terminal involved in this application can be a mobile phone, a tablet computer (Tablet Personal Computer), a laptop computer (Laptop Computer), or a notebook computer, a personal digital assistant (Personal Digital Assistant, PDA), a handheld computer, a netbook, or a super mobile personal computer.
  • Tablet Personal Computer Tablet Personal Computer
  • laptop computer laptop computer
  • PDA Personal Digital Assistant
  • the network equipment involved in this application may include access network equipment, where the access network equipment may also be called wireless access network equipment, radio access network (Radio Access Network, RAN), wireless access network function or wireless access network unit.
  • Access network equipment can include base stations, Wireless Local Area Network (WLAN) access points or Wireless Fidelity (WiFi) nodes, etc.
  • WLAN Wireless Local Area Network
  • WiFi Wireless Fidelity
  • the base station can be called Node B, Evolved Node B (eNB), Access Point Access point, Base Transceiver Station (BTS), Radio Base Station, Radio Transceiver, Basic Service Set (BSS), Extended Service Set (ESS), Home B-node, Home Evolved B node, Transmitting Receiving Point (TRP) or some other suitable terminology in the field, as long as the same technical effect is achieved, the base station is not limited to specific technical terms. It should be noted that in the implementation of this application This example only takes the base station in the NR system as an example, and does not limit the specific type of base station.
  • an embodiment of the present application provides a CSI feedback method, which is applied to a terminal. Specific steps include: step 401 and step 402.
  • Step 401 Obtain the first CSI and the second CSI, which are obtained by the terminal performing measurement processing based on the first reference signal and the second reference signal respectively;
  • the first and second reference signals may include CSI reference signals (Channel-State-Information Reference Signal, CSI-RS), or CSI-RS group. If the first and second reference signals include CSI-RS groups, compared to the continuously transmitted CSI-RS, the network device will The transmission mode is changed to pulse group, and the channel feedback of the vacant part in the middle of adjacent CSI-RS groups is completed with the help of channel time domain correlation, thus reducing the channel feedback overhead.
  • CSI reference signals Channel-State-Information Reference Signal, CSI-RS
  • CSI-RS group Channel-State-Information Reference Signal
  • Step 402 Compare the first CSI and the second CSI to determine whether to feed back the second CSI or the first reply.
  • the first reply is used to indicate that the terminal has received the second reference signal, and There is no need to feed back the second CSI.
  • the above comparison result is used to indicate whether the network device is capable of predicting the second CSI. If the comparison result indicates that the network device is not capable of predicting the second CSI, the terminal feeds back the second CSI to the network device. If the comparison result indicates that the network device is capable of predicting the second CSI, CSI, the terminal does not feed back the second CSI to the network device, and reduces the CSI feedback frequency while ensuring the accuracy of CSI prediction.
  • obtaining the first CSI and the second CSI includes:
  • the first CSI is obtained according to the plurality of third CSIs, and the first CSI is CSI in the delay Doppler domain;
  • the second CSI is obtained according to the plurality of fourth CSIs, and the second CSI is CSI in the delay Doppler domain;
  • the first reference signal and the second reference signal are two adjacent reference signals sent by the network device.
  • the above-mentioned first reference signal includes a first CSI-RS group
  • the second reference signal includes a second CSI-RS group.
  • the plurality of third CSI and the plurality of fourth CSI can also be called CSI time-frequency spatial domain results, and the first CSI and second CSI can also be called CSI actual measurement results in the delay Doppler domain.
  • CSI compression is mainly concentrated in the spatial domain and frequency domain at the same time.
  • CSI is compressed in the time domain to further reduce feedback overhead.
  • the second reference signal is sent by the network device according to the first beamforming, and the first beamforming is determined by the network device according to the precoding matrix,
  • the precoding matrix is generated by the network device based on the fifth CSI, and the fifth CSI is predicted by the network device based on the received first CSI and an artificial intelligence (Artificial Intelligence, AI) model.
  • AI Artificial Intelligence
  • the second CSI is obtained by the terminal performing measurement processing based on the second reference signal.
  • the terminal needs to feed back the actual measurement result of the second reference signal to the network device.
  • the network device Since the network device has the capability of channel prediction, the network device can predict the fifth CSI at the subsequent time based on the previously received first CSI and AI model.
  • the network device may not receive the second reference signal from the terminal. CSI feedback (that is, the second CSI may not be received), and the predicted fifth CSI is used as the CSI feedback of the second reference signal.
  • obtaining the first CSI according to the plurality of third CSIs includes:
  • Obtaining the second CSI according to the plurality of fourth CSIs includes:
  • performing a channel comparison according to the first CSI and the second CSI, and determining to feed back the second CSI or the first reply includes:
  • performing channel correlation calculation based on the first CSI and second CSI, and determining whether to feed back the second CSI based on the correlation calculation result includes:
  • the above correlation threshold may also be called an error threshold.
  • the second CSI is sent to the network device.
  • the method further includes:
  • the correlation threshold configured by the network device
  • the correlation threshold represents the network
  • the network equipment obtains the minimum acceptable CSI delay Doppler spectrum correlation according to the service quality requirements.
  • the network device is used to perform channel prediction and the correlation threshold set by the network device, and the terminal determines whether the channel characteristics have undergone changes that are unacceptable to the AI model of the network device, so that the terminal can operate at a lower frequency Feedback CSI, thereby reducing feedback overhead, and through correlation calculation and correlation threshold comparison performed at the terminal, the terminal can feed back the measured CSI in a more timely manner when the channel status changes significantly, increasing the reliability of the channel measurement mechanism.
  • an embodiment of the present application provides a CSI feedback method, which is applied to network equipment. Specific steps include: step 501, step 502, step 503, and step 504.
  • Step 501 Send the first reference signal
  • Step 502 Receive the first CSI, which is obtained by the terminal according to the first reference signal measurement process
  • Step 503 Send the second reference signal
  • Step 504 Receive the first reply or the second CSI.
  • the first reply or the second CSI is fed back by the terminal based on comparison of the first CSI and the second CSI.
  • the second CSI is fed back by the terminal based on the second reference. Obtained through signal measurement processing, the first reply is used to indicate that the terminal has received the second reference signal and does not need to feed back the second CSI.
  • the first reference signal and the second reference signal are two adjacent reference signals sent by the network device.
  • sending the second reference signal includes:
  • the first CSI and AI model perform CSI prediction to obtain the predicted fifth CSI
  • the second reference signal is transmitted through the first beamforming.
  • the method further includes:
  • a third reference signal is transmitted through the second beamforming.
  • the terminal feeds back the first reply to the base station, it means that the current channel characteristics are highly relevant and the network equipment CSI can continue to be predicted.
  • the method further includes:
  • the second CSI and AI model perform CSI prediction to obtain the predicted seventh CSI
  • a fourth reference signal is transmitted through the third beamforming.
  • the method further includes:
  • the correlation threshold is configured, and the correlation threshold represents the minimum CSI delay Doppler spectrum correlation that the network device can accept according to the service quality requirements.
  • the network device is used to perform channel prediction and the correlation threshold set by the network device, and the terminal determines whether the channel characteristics have undergone changes that are unacceptable to the AI model of the network device, so that the terminal can operate at a lower frequency Feedback CSI, thereby reducing feedback overhead, and through correlation calculation and correlation threshold comparison performed at the terminal, the terminal can feed back the measured CSI in a more timely manner when the channel status changes significantly, increasing the reliability of the channel measurement mechanism.
  • Step 1 The base station configures and sends relevant parameters of the CSI-RS group to the terminal.
  • the relevant parameters of the CSI-RS group include one or more of the following:
  • the CSI correlation threshold in this article can also be called correlation threshold.
  • the base station Before actual deployment, some actual measurement or simulation work may be required to determine the impact of the correlation between different delay Doppler spectra on the performance of the channel prediction model under different environmental motion states, so that the base station can determine the performance of the channel prediction model according to its Quality of Service (QoS). ) needs to obtain the minimum acceptable CSI delay Doppler spectrum correlation, that is, the CSI correlation threshold E.
  • QoS Quality of Service
  • Step 2 The base station sends the first CSI-RS group to the terminal.
  • the first CSI-RS group herein may also be called the first reference signal.
  • Step 3 The terminal performs measurement estimation on the received first CSI-RS group to obtain multiple third CSIs, And perform symplectic Fourier transform on it to convert it into the first CSI in the delay Doppler domain.
  • the measurement estimation in this article may also be called CSI-RS pulse measurement estimation.
  • the plurality of third CSIs in this article may also be called third CSI time-frequency spatial domain information, or third CSI time-frequency spatial domain results, or third CSI feedback, or third CSI actual measurement results.
  • the first CSI in this article may also be called the first CSI measured result, or the first CSI measured Doppler information, or the first CSI feedback.
  • Step 4 The terminal compresses and feeds back the first CSI in the delay Doppler domain to the base station.
  • Step 5 The base station decompresses the CSI sent by the terminal and performs inverse symplectic Fourier transform to restore it to the first CSI.
  • the base station predicts the fifth CSI at subsequent times based on the first CSI and the AI model, and generates the corresponding precoding matrix.
  • the AI model in this article can also be called an AI pre-training model.
  • the fifth CSI in this article may also be called fifth CSI time-frequency spatial domain information, or CSI prediction information, or CSI prediction results.
  • Step 6 The base station performs beamforming based on the precoding matrix obtained from the fifth CSI.
  • Step 7 After sending the first CSI-RS group, the base station sends the second CSI-RS group to the terminal at an inter-CSI-RS group interval (T).
  • T inter-CSI-RS group interval
  • Step 8 The terminal performs measurement estimation on the second CSI-RS group, obtains multiple fourth CSIs, and performs symplectic Fourier transform on them to convert them into second CSIs in the delay Doppler domain.
  • the plurality of fourth CSIs in this article may also be called fourth CSI time-frequency spatial domain information, or fourth CSI time-frequency spatial domain results, or fourth CSI feedback, or fourth CSI actual measurement results.
  • the second CSI in this article may also be called the second CSI actual measurement result, or the second CSI actual measurement Doppler information, or the second CSI feedback.
  • Step 9 The terminal performs channel correlation calculation on the first CSI obtained in step 3 and the second CSI in step 8. If the calculated channel correlation is greater than the CSI correlation threshold (E), perform step 10a. If the calculated channel correlation If the correlation is less than the CSI correlation threshold (E), perform step 10b.
  • the terminal has the ability to perform correlation calculations on the delay Doppler spectrum at different times to determine whether the relative motion state between the terminal and the surrounding environment has changed significantly.
  • the terminal uses the latest delay Doppler spectrum fed back to the base station as Referring to the results, the correlation operation is performed with the subsequent delay Doppler spectrum obtained by CSI-RS analysis.
  • the correlation coefficient of the matrix can be used to represent the correlation between different delay Doppler spectra. The calculation method is as follows:
  • a and B are the delay Doppler spectrum matrices, m and n are the rows and columns of the matrix, r is the matrix correlation coefficient.
  • the terminal determines whether the base station is capable of predicting subsequent CSI based on the CSI obtained by analyzing the periodically delivered CSI-RS, and decides whether to feed back the latest CSI measurement results to the base station.
  • the CSI feedback frequency can be reduced while ensuring the accuracy of CSI prediction.
  • Step 10a The terminal feeds back a first reply to the base station.
  • the first reply is used to indicate that the terminal has received the second reference signal but does not need to feed back other information, and then performs step 11a. Refer to Figure 6a.
  • the first reply in this article may also be called judgment result reply signaling, or first reply (Acknowledge character, ACK).
  • the delay Doppler spectrum correlation is less than the CSI correlation threshold. At this time, it is similar to the existing process and feeds back the CSI to the base station; the other is that the delay Doppler spectrum correlation is less than the CSI correlation threshold.
  • the delay Doppler spectrum correlation is greater than the CSI correlation threshold, using the base station's predictive CSI capability, the terminal does not need to feed back CSI. However, if the terminal does not feed back any information to the base station, the base station cannot determine that it is due to the current channel characteristics. There is no feedback because the CSI-RS or CSI feedback is lost during the transmission process.
  • the first reply is introduced in this embodiment.
  • the reply indicates that the current channel characteristics are highly relevant and the base station can continue to predict CSI.
  • Step 10b The terminal compresses and feeds back the second CSI in the delay Doppler domain to the base station, and then performs step 11b, refer to Figure 6b.
  • the second CSI in this article may also be called the CSI actual measurement result, or the CSI actual measurement Doppler information.
  • Step 11a After receiving the first reply, the base station continues to predict the CSI at subsequent times, and Generate the corresponding precoding matrix, and then perform step 12, refer to Figure 6a.
  • Step 11b The base station receives the second CSI sent by the terminal, uses the second CSI and the AI model to predict the CSI at subsequent times, and generates the corresponding precoding matrix, and then performs step 12, refer to Figure 6b.
  • Step 12 The base station performs beamforming based on the precoding matrix obtained from the CSI prediction information, and then returns to step 7.
  • the base station has the ability to perform CSI prediction (i.e., timing sequence prediction) based on the AI model and CSI time-frequency spatial domain information (i.e., measured channel information).
  • the current AI model for predicting the timing sequence can be a recurrent neural network (Recurrent Neural Network, RNN) network.
  • the RNN network includes a long short-term memory (LSTM) network.
  • LSTM long short-term memory
  • Figure 7 The structure of the LSTM is shown in Figure 7.
  • x t is the network input at time t
  • y t is the network output at time t
  • h t and c t represent the status of the network unit at time t.
  • the LSTM network needs to add three gates to the network unit.
  • the first is the forgetting gate.
  • the forgetting gate mainly selectively forgets the input passed in from the previous node.
  • the calculated z f is used as the forgetting gate.
  • Gate control is used to control which of the previous state c t-1 needs to be retained and which needs to be forgotten; followed by the memory gate, the memory gate will selectively memorize the input x t , and the current input content is represented by z calculated previously. .
  • the gating signal that selects what kind of information to memorize is controlled by z i ; finally, there is the output gate, which determines which information will be treated as the output of the current state.
  • the LSTM network can control the transmission state through the gating state, remember information that requires long-term memory, and discard unimportant information.
  • GRU Gated Recurrent Unit
  • the measured CSI results are first used as model input to obtain a CSI prediction result.
  • the CSI prediction results obtained in the previous round can be regarded as the new known CSI, and the earliest measured data Eliminate and add the previous round of CSI prediction results from the model input as the model input so that the shape of the input matrix remains unchanged, and so on.
  • this prediction process can continue, but in fact, considering the changes in the prediction results in the model input In order to ensure the quality of the prediction results due to the increase in the number of channels and the changes in the characteristics of the channel itself, other mechanisms need to be added to limit the time to end the prediction.
  • an embodiment of the present application provides a device for CSI feedback, which is applied to a terminal.
  • the device 900 includes:
  • the first acquisition module 901 is used to acquire the first CSI and the second CSI, which are obtained by the terminal through measurement processing based on the first reference signal and the second reference signal respectively;
  • the first determining module 902 is configured to perform channel comparison according to the first CSI and the second CSI, and determine to feed back the second CSI or a first reply, where the first reply is used to indicate that the terminal has received the The second reference signal does not need to feed back the second CSI.
  • the first acquisition module 901 is further used to:
  • the plurality of third CSIs obtain the first CSI in the delay Doppler domain
  • a second CSI in the delay Doppler domain is obtained.
  • the first reference signal includes a first CSI-RS group
  • the second reference signal includes a second CSI-RS group
  • the second reference signal is sent by the network device according to the first beamforming, the first beamforming is determined by the network device according to the precoding matrix, and the precoding matrix
  • the coding matrix is generated by the network device based on the fifth CSI, and the fifth CSI is predicted by the network device based on the first CSI and the AI model.
  • the first acquisition module 901 is further used to:
  • the first determination module 902 is further configured to: perform channel correlation calculation according to the first CSI and the second CSI, and determine to feed back the second CSI or the second CSI according to the correlation calculation result. First reply.
  • the first determining module 902 is further used to:
  • a first reply is sent to the network device, where the first reply is used to indicate that the terminal has received the second reference signal without feedback of the third reference signal. 2 CSI.
  • the device further includes:
  • the second acquisition module is used to acquire the correlation threshold configured by the network device.
  • the correlation threshold represents the minimum CSI delay Doppler spectrum correlation that the network device can accept according to the service quality requirements.
  • the device further includes:
  • a first sending module configured to send the second CSI to a network device if the correlation calculation result is less than the correlation threshold.
  • the device provided by the embodiment of the present application can implement each process implemented by the method embodiment shown in Figure 4 and achieve the same technical effect. To avoid duplication, the details will not be described here.
  • an embodiment of the present application provides a device for CSI feedback, which is applied to network equipment.
  • the device 1000 includes:
  • the second sending module 1001 is used to send the first reference signal
  • the first receiving module 1002 is configured to receive the first CSI, which is obtained by the terminal according to the first reference signal measurement processing;
  • the third sending module 1003 is used to send the second reference signal
  • the second receiving module 1004 is configured to receive a first reply or a second CSI.
  • the first reply or the second CSI is fed back by the terminal based on comparing the first CSI and the second CSI.
  • the second CSI is The first reply is obtained by the terminal according to the second reference signal measurement process, and is used to indicate that the terminal has received the second reference signal and does not need to feed back the second CSI.
  • the third sending module 1003 is further used for:
  • the first CSI and AI model perform CSI prediction to obtain the predicted fifth CSI
  • the second reference signal is transmitted through the first beamforming.
  • the device further includes: a first processing module, configured to:
  • a third reference signal is transmitted through the second beamforming.
  • the device further includes: a second processing module, used for:
  • the second CSI and AI model perform CSI prediction to obtain the predicted seventh CSI
  • a fourth reference signal is transmitted through the third beamforming.
  • the device further includes:
  • a configuration module configured to configure the correlation threshold, where the correlation threshold represents the minimum CSI delay Doppler spectrum correlation that the network device can accept according to the service quality requirements.
  • the device provided by the embodiment of the present application can implement each process implemented by the method embodiment shown in Figure 5 and achieve the same technical effect. To avoid duplication, the details will not be described here.
  • the embodiment of the present application also provides a communication device 1100, including a processor 1101, a memory 1102, and a program or instruction stored on the memory 1102 and executable on the processor 1101.
  • the program or instruction When executed by the processor 1101, each process of the above method embodiment in Figure 4 or Figure 5 is implemented, and the same technical effect can be achieved. To avoid repetition, they will not be repeated here.
  • Embodiments of the present application also provide a readable storage medium, with programs or instructions stored on the readable storage medium.
  • program or instructions When the program or instructions are executed by a processor, each process of the method embodiment shown in Figure 4 or Figure 5 is implemented. , and can achieve the same technical effect, so to avoid repetition, they will not be described again here.
  • the processor is the processor in the terminal described in the above embodiment.
  • the readable storage media includes computer-readable storage media, such as computer read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disks or optical disks, etc.
  • the steps of the method or algorithm described in conjunction with the disclosure of this application can be implemented in hardware or by executing software instructions on a processor.
  • Software instructions can be composed of corresponding software modules.
  • Software modules can be stored in RAM, flash memory, ROM, Erasable Programmable Read-Only Memory (EPROM), and electrically erasable programmable read-only memory. (Electrically Erasable Programmable Read-Only Memory, EEPROM), register, hard disk, mobile hard disk, read-only optical disk or any other form of storage media well known in the art.
  • An exemplary storage medium is coupled to the processor such that the processor can read information from the storage medium and write information to the storage medium.
  • the storage medium can also be an integral part of the processor.
  • the processor and storage medium can be carried in an Application Specific Integrated Circuit (ASIC).
  • the ASIC can be carried in the core network interface device.
  • the processor and the storage medium can also exist as discrete components in the core network interface device.
  • Computer-readable media includes computer storage media and communication media including any medium that facilitates transfer of a computer program from one place to another.
  • Storage media can be any available media that can be accessed by a general purpose or special purpose computer.
  • embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, embodiments of the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment that combines software and hardware aspects. Furthermore, embodiments of the present application may employ computer-usable storage media (including but not limited to It is not limited to the form of computer program products implemented on magnetic disk storage, Compact Disc Read-Only Memory (CD-ROM), optical storage, etc.).
  • CD-ROM Compact Disc Read-Only Memory
  • Embodiments of the present application are described with reference to flowcharts and/or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present application. It will be understood that each process and/or block in the flowchart illustrations and/or block diagrams, and combinations of processes and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing device to produce a machine, such that the instructions executed by the processor of the computer or other programmable data processing device produce a use A device for realizing the functions specified in one process or multiple processes of the flowchart and/or one block or multiple blocks of the block diagram.
  • These computer program instructions may also be stored in a computer-readable memory that causes a computer or other programmable data processing apparatus to operate in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including the instruction means, the instructions
  • the device implements the functions specified in a process or processes of the flowchart and/or a block or blocks of the block diagram.
  • These computer program instructions may also be loaded onto a computer or other programmable data processing device, causing a series of operating steps to be performed on the computer or other programmable device to produce computer-implemented processing, thereby executing on the computer or other programmable device.
  • Instructions provide steps for implementing the functions specified in a process or processes of a flowchart diagram and/or a block or blocks of a block diagram.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Signal Processing For Digital Recording And Reproducing (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

本申请实施例提供一种CSI反馈的方法、装置、设备及可读存储介质,该方法包括:获取第一CSI和第二CSI,所述第一CSI和第二CSI是所述终端分别基于第一参考信号和第二参考信号进行测量处理得到的;根据所述第一CSI和第二CSI进行比较,确定反馈所述第二CSI或第一回复,所述第一回复用于表示所述终端已接收到所述第二参考信号,且无需反馈所述第二CSI。

Description

CSI反馈的方法、装置、设备及可读存储介质
相关申请的交叉引用
本申请主张在2022年07月05日在中国提交的中国专利申请No.202210846696.4的优先权,其全部内容通过引用包含于此。
技术领域
本申请实施例涉及通信技术领域,具体涉及一种信道状态信息(Channel State Information,CSI)反馈的方法、装置、设备及可读存储介质。
背景技术
相关技术中的信道预测方案是仅在网络侧进行信道预测以降低终端的CSI反馈频率,但由于CSI反馈频率的降低,导致网络侧对当前实际CSI获取的实时性降低,如果信道发生剧烈变化,会导致网络侧预测性能明显下降。
因此,如何在保证CSI预测准确性的情况下减小CSI反馈频率是亟待解决的问题。
发明内容
本申请实施例在于提供一种CSI反馈的方法、装置、设备及可读存储介质,解决如何在保证CSI预测准确性的情况下减小CSI反馈频率的技术问题。
第一方面,提供一种CSI反馈的方法,应用于终端,所述方法包括:
获取第一CSI和第二CSI,所述第一CSI和第二CSI是所述终端分别基于第一参考信号和第二参考信号进行测量处理得到的;
根据所述第一CSI和第二CSI进行信道比较相关性计算,并根据相关性计算结果,确定是否反馈所述第二CSI或第一回复,所述第一回复用于表示所述终端已接收到所述第二参考信号,且无需反馈所述第二CSI。
可选地,获取第一CSI和第二CSI,包括:
对第一参考信号进行测量,根据所述第一参考信号的测量结果得到多个第三CSI;
根据所述多个第三CSI得到所述第一CSI,所述第一CSI为时延多普勒域上的CSI;
对第二参考信号进行测量,根据所述第二参考信号的测量结果得到多个第四CSI;
根据所述多个第四CSI得到所述第二CSI,所述第二CSI为时延多普勒域上的CSI。
可选地,所述第一参考信号包括第一CSI-RS组,所述第二参考信号包括第二CSI-RS组。
可选地,所述第二参考信号是网络设备根据第一波束赋形发送的,所述第一波束赋形是所述网络设备根据预编码矩阵确定的,所述预编码矩阵是所述网络设备根据第五CSI生成的,所述第五CSI是所述网络设备根据第一CSI和AI模型预测得到的。
可选地,根据所述多个第三CSI得到所述第一CSI,包括:
对所述多个第三CSI进行辛傅里叶变换,获得所述第一CSI;
根据所述多个第四CSI得到所述第二CSI,包括:
对所述多个第四CSI进行辛傅里叶变换,获得所述第二CSI。
可选地,根据所述第一CSI和第二CSI进行信道比较,确定反馈所述第二CSI或第一回复,包括:
根据所述第一CSI和第二CSI进行信道相关性计算,并根据相关性计算结果,确定反馈所述第二CSI或第一回复。
可选地,根据所述第一CSI和第二CSI进行信道相关性计算,并根据相关性计算结果,确定是否反馈所述第二CSI,包括:
根据所述第一CSI和第二CSI进行信道相关性计算,得到相关性计算结果;
将所述相关性计算结果与相关性阈值进行比较;
如果所述相关性计算结果大于或等于所述相关性阈值,则向网络设备发送第一回复;
获取所述网络设备配置的所述相关性阈值,所述相关性阈值表示所述网络设备根据服务质量需求得到所能接受的最小CSI时延多普勒谱相关性。
可选地,所述方法还包括:
如果所述相关性计算结果小于所述相关性阈值,则向网络设备发送所述第二CSI。
第二方面,提供一种CSI反馈的方法,应用于网络设备,所述方法包括:
发送第一参考信号;
接收第一CSI,所述第一CSI是终端根据所述第一参考信号测量处理得到的;
发送第二参考信号;
接收第一回复或者第二CSI,所述第一回复或者第二CSI是终端根据第一CSI和第二CSI进行比较确定反馈的,所述第二CSI是终端根据所述第二参考信号测量处理得到的,所述第一回复用于表示所述终端已接收到第二参考信号,且无需反馈所述第二CSI。
可选地,发送第二参考信号,包括:
根据所述第一CSI和AI模型,进行CSI预测,得到预测的第五CSI;
根据所述第五CSI,生成相应的预编码矩阵;
根据所述预编码矩阵,确定第一波束赋形;
通过所述第一波束赋形发送第二参考信号。
可选地,在接收第一回复之后,所述方法还包括:
根据所述第一CSI和AI模型,进行CSI预测,得到预测的第六CSI;
根据所述第六CSI,生成相应的预编码矩阵;
根据所述预编码矩阵,确定第二波束赋形;
通过所述第二波束赋形发送第三参考信号。
可选地,在接收第二CSI之后,所述方法还包括:
根据所述第二CSI和AI模型,进行CSI预测,得到预测的第七CSI;
根据所述第七CSI,生成相应的预编码矩阵;
根据所述预编码矩阵,确定第三波束赋形;
通过所述第三波束赋形发送第四参考信号。
可选地,所述方法还包括:
配置所述相关性阈值,所述相关性阈值表示所述网络设备根据服务质量 需求得到所能接受的最小CSI时延多普勒谱相关性。
第三方面,提供一种CSI反馈的装置,应用于终端,包括:
第一获取模块,用于获取第一CSI和第二CSI,所述第一CSI和第二CSI是所述终端分别基于第一参考信号和第二参考信号进行测量处理得到的;
第一确定模块,用于根据所述第一CSI和第二CSI进行信道比较,确定反馈所述第二CSI或第一回复,所述第一回复用于表示所述终端已接收到所述第二参考信号,且无需反馈所述第二CSI。
第四方面,提供一种CSI反馈的装置,应用于网络设备,包括:
第二发送模块,用于发送第一参考信号;
第一接收模块,用于接收第一CSI,所述第一CSI是终端根据所述第一参考信号测量处理得到的;
第三发送模块,用于发送第二参考信号;
第二接收模块,用于接收第一回复或者第二CSI,所述第一回复或者第二CSI是终端根据第一CSI和第二CSI进行比较确定反馈的,所述第二CSI是终端根据所述第二参考信号测量处理得到的,所述第一回复用于表示所述终端已接收到所述第二参考信号,且无需反馈所述第二CSI。
第五方面,提供一种通信设备,包括处理器,存储器及存储在所述存储器上并可在所述处理器上运行的程序或指令,所述程序或指令被所述处理器执行时实现如第一方面或第二方面所述的方法的步骤。
第六方面,提供一种可读存储介质,所述可读存储介质上存储程序或指令,所述程序或指令被处理器执行时实现如第一方面或第二方面所述的方法的步骤。
在本申请实施例中,终端获取第一CSI和第二CSI,所述第一CSI和第二CSI是所述终端分别基于第一参考信号和第二参考信号进行测量处理得到的;然后终端根据所述第一CSI和第二CSI进行比较,确定反馈所述第二CSI或第一回复,其中,比较结果用于表示网络设备是否有能力预测第二CSI,如果比较结果表示网络设备没有能力预测第二CSI,则终端向网络设备反馈第二CSI,如果比较结果表示网络设备有能力预测第二CSI,则终端不向网络设备反馈第二CSI,从而在在保证CSI预测准确性的情况下减小CSI反馈频 率。
附图说明
通过阅读下文优选实施方式的详细描述,各种其他的优点和益处对于本领域普通技术人员将变得清楚明了。附图仅用于示出优选实施方式的目的,而并不认为是对本申请的限制。而且在整个附图中,用相同的参考符号表示相同的部件。在附图中:
图1是码本演化过程的示意图;
图2是时域信道压缩反馈的示意图;
图3是本申请实施例提供的一种通信系统的示意图;
图4是本申请实施例提供的一种CSI反馈的方法的流程图之一;
图5是本申请实施例提供的一种CSI反馈的方法的流程图之二;
图6a是本申请实施例提供的信道反馈时域动态压缩的示意图之一;
图6b是本申请实施例提供的信道反馈时域动态压缩的示意图之二;
图7是LSTM网络单元结构的示意图;
图8是基于LSTM网络的预测方法的示意图;
图9是本申请实施例提供的一种CSI反馈的装置的示意图之一;
图10是本申请实施例提供的一种CSI反馈的装置的示意图之二;
图11是本申请的实施例提供的通信设备的示意图。
具体实施方式
下面将结合本申请实施例中的附图,对本申请实施例中的技术方案进行清楚、完整地描述,显然,所描述的实施例是本申请一部分实施例,而不是全部的实施例。基于本申请中的实施例,本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例,都属于本申请保护的范围。
本申请的说明书和权利要求书中的术语“包括”以及它的任何变形,意图在于覆盖不排他的包含,例如,包含了一系列步骤或单元的过程、方法、系统、产品或设备不必限于清楚地列出的那些步骤或单元,而是可包括没有清楚地列出的或对于这些过程、方法、产品或设备固有的其它步骤或单元。此 外,说明书以及权利要求中使用“和/或”表示所连接对象的至少其中之一,例如A和/或B,表示包含单独A,单独B,以及A和B都存在三种情况。
在本申请实施例中,“示例性的”或者“例如”等词用于表示作例子、例证或说明。本申请实施例中被描述为“示例性的”或者“例如”的任何实施例或设计方案不应被解释为比其它实施例或设计方案更优选或更具优势。确切而言,使用“示例性的”或者“例如”等词旨在以具体方式呈现相关概念。
在新空口(New Radio,NR)支持的不同信道码本类型中,类型1(Type I)码本每一层的预编码矢量通过一个二维离散傅里叶变换(Discrete Fourier Transform,DFT)矢量及其相位旋转量化,量化精度不高,然而,Type I终端处理最简单、开销最低,因此在NR中被认为是最基本的预编码矩阵指示(Precoding Matrix Indicator,PMI)反馈码本,是NR的终端必选技术;类型2(Type II)码本的基本结构是在每层的每个极化方向上使用一组基矢量的线性加权组合作为预编码矢量。终端反馈的PMI包括基矢量的信息,加权系数的幅度和相位信息。从无线信道的特征上来看,具有较强能量贡献的多径通常具有稀疏的特性,即它们在整个N维空间中只占很小的一个子空间。因此,表示这些多径所在的子空间,所需要的基矢量个数就可以大幅度减小,从而达到压缩的目的。
由于不同子带的幅度和相位之间存在一定的相关性,尤其是相位,不同子带的相位变化是由于无线信道中多径的时延变化到频域上的不同位置带来的。这样的相关性使得进一步压缩Type II码本的开销成为可能。
如图1所示,从Type I到演进到Type II是通过在空域上选取基向量进行压缩,而eType II则是在Type II的基础上在频率子带上进一步选取基向量,实现频域上的信道压缩。如图2所示,利用时域相关性进行信道反馈压缩。其中,高更新率是指high update rate,低更新率w/o多普勒信息是指low update rate w/o Doppler information,低更新率w/多普勒信息是指low update rate w/Doppler information。
参见图3,为本发明实施例提供的一种无线通信系统的架构示意图。该无线通信系统可以包括:网络设备31和终端32,终端32可以与网络设备31通信(传输信令或传输数据)。在实际应用中上述各个设备之间的连接可以为无线 连接,为了方便直观地表示各个设备之间的连接关系,图3中采用实线示意。
本申请涉及的终端可以是手机、平板电脑(Tablet Personal Computer)、膝上型电脑(Laptop Computer)或称为笔记本电脑、个人数字助理(Personal Digital Assistant,PDA)、掌上电脑、上网本、超级移动个人计算机(ultra-mobile personal computer,UMPC)、移动上网装置(Mobile Internet Device,MID)、增强现实(augmented reality,AR)/虚拟现实(virtual reality,VR)设备、机器人、可穿戴式设备(Wearable Device)、车载设备(Vehicle User Equipment,VUE)、行人终端(Pedestrian User Equipment,PUE)、智能家居(具有无线通信功能的家居设备,如冰箱、电视、洗衣机或者家具等)、游戏机、个人计算机(personal computer,PC)、柜员机或者自助机等终端侧设备,可穿戴式设备包括:智能手表、智能手环、智能耳机、智能眼镜、智能首饰(智能手镯、智能手链、智能戒指、智能项链、智能脚镯、智能脚链等)、智能腕带、智能服装等。需要说明的是,在本申请实施例并不限定终端的具体类型。
本申请涉及的网络设备可以包括接入网设备,其中,接入网设备也可以称为无线接入网设备、无线接入网(Radio Access Network,RAN)、无线接入网功能或无线接入网单元。接入网设备可以包括基站、无线局域网(Wireless Local Area Network,WLAN)接入点或无线保真(Wireless Fidelity,WiFi)节点等,基站可被称为节点B、演进节点B(eNB)、接入点、基站收发机(Base Transceiver Station,BTS)、无线电基站、无线电收发机、基本服务集(Basic Service Set,BSS)、扩展服务集(Extended Service Set,ESS)、家用B节点、家用演进型B节点、发送接收点(Transmitting Receiving Point,TRP)或所述领域中其他某个合适的术语,只要达到相同的技术效果,所述基站不限于特定技术词汇,需要说明的是,在本申请实施例中仅以NR系统中的基站为例进行介绍,并不限定基站的具体类型。
参见图4,本申请实施例提供一种CSI反馈的方法,应用于终端,具体步骤包括:步骤401和步骤402。
步骤401:获取第一CSI和第二CSI,所述第一CSI和第二CSI是所述终端分别基于第一参考信号和第二参考信号进行测量处理得到的;
可选地,第一、第二参考信号可以包括CSI参考信号 (Channel-State-Information Reference Signal,CSI-RS),或者CSI-RS组,如果第一、第二参考信号包括CSI-RS组,相较于连续发送的CSI-RS,网络设备将参考信号的发送形式改为了脉冲组,相邻的CSI-RS组中间空缺部分的信道反馈借助信道时域相关性进行补全,从而降低了信道反馈开销。
步骤402:根据所述第一CSI和第二CSI进行比较,确定反馈所述第二CSI或第一回复,所述第一回复用于表示所述终端已接收到所述第二参考信号,且无需反馈所述第二CSI。
上述比较结果用于表示网络设备是否有能力预测第二CSI,如果比较结果表示网络设备没有能力预测第二CSI,则终端向网络设备反馈第二CSI,如果比较结果表示网络设备有能力预测第二CSI,则终端不向网络设备反馈第二CSI,在保证CSI预测准确性的情况下减小CSI反馈频率。
在本申请的一种实施方式中,获取第一CSI和第二CSI,包括:
对第一参考信号进行测量,根据所述第一参考信号的测量结果得到多个第三CSI;
根据所述多个第三CSI得到所述第一CSI,所述第一CSI为时延多普勒域上的CSI;
对第二参考信号进行测量,根据所述第二参考信号的测量结果得到多个第四CSI;
根据所述多个第四CSI得到所述第二CSI,所述第二CSI为时延多普勒域上的CSI;
其中,所述第一参考信号和第二参考信号是网络设备发送的相邻的两个参考信号。
上述第一参考信号包括第一CSI-RS组,第二参考信号包括第二CSI-RS组。
上述多个第三CSI、多个第四CSI也可以称为CSI时频空域结果,第一CSI、第二CSI也可以称为时延多普勒域上的CSI实测结果。
相关技术中,对于CSI的压缩主要集中在同一时刻的空域和频域上,在本实施例中,而在时域对CSI进行压缩,进一步减少反馈开销。
在本申请的一种实施方式中,所述第二参考信号是网络设备根据第一波束赋形发送的,所述第一波束赋形是所述网络设备根据预编码矩阵确定的, 所述预编码矩阵是所述网络设备根据第五CSI生成的,所述第五CSI是所述网络设备根据接收到的第一CSI和人工智能(Artificial Intelligence,AI)模型预测得到的。
可以理解的是,第二CSI是终端根据第二参考信号进行测量处理得到的,在现有技术中,终端需要将其作为第二参考信号的实测结果反馈给网络设备,在本实施例中,由于网络设备具有信道预测的能力,该网络设备可以根据之前接收到的第一CSI和AI模型预测后续时刻的第五CSI,对于网络设备而言,该网络设备可以不从终端接收第二参考信号的CSI反馈(即,可以不接收第二CSI),将其预测得到的第五CSI作为第二参考信号的CSI反馈。
在本申请的一种实施方式中,根据所述多个第三CSI得到所述第一CSI,包括:
对所述多个第三CSI进行辛傅里叶变换,获得所述第一CSI;
根据所述多个第四CSI得到所述第二CSI,包括:
对所述多个第四CSI进行辛傅里叶变换,获得所述第二CSI。
可选地,根据所述第一CSI和第二CSI进行信道比较,确定反馈所述第二CSI或第一回复,包括:
根据所述第一CSI和第二CSI进行信道相关性计算,并根据相关性计算结果,确定反馈所述第二CSI或第一回复。
在本申请的一种实施方式中,根据所述第一CSI和第二CSI进行信道相关性计算,并根据相关性计算结果,确定是否反馈所述第二CSI,包括:
根据所述第一CSI和第二CSI进行信道相关性计算,得到相关性计算结果;
将所述相关性计算结果与相关性阈值进行比较;
如果所述相关性计算结果大于或等于所述相关性阈值,则向网络设备发送第一回复;
上述相关性阈值也可以称为误差阈值。
如果所述相关性计算结果小于所述相关性阈值,则向网络设备发送所述第二CSI。
在本申请的一种实施方式中,所述方法还包括:
获取所述网络设备配置的所述相关性阈值,所述相关性阈值表示所述网 络设备根据服务质量需求得到所能接受的最小CSI时延多普勒谱相关性。
在本申请的实施例中,利用网络设备进行信道预测以及网络设备设定的相关性阈值并由终端判别信道特性是否发生了网络设备的AI模型无法接受的改变,使得终端能以更低的频率反馈CSI,从而降低反馈开销,并且通过在终端进行的相关性计算以及相关性门限比较,当信道状态发生明显改变时终端可以更及时地反馈实测的CSI,增加信道测量机制的可靠性。
参见图5,本申请实施例提供一种CSI反馈的方法,应用于网络设备,具体步骤包括:步骤501、步骤502、步骤503、步骤504。
步骤501:发送第一参考信号;
步骤502:接收第一CSI,所述第一CSI是终端根据所述第一参考信号测量处理得到的;
步骤503:发送第二参考信号;
步骤504:接收第一回复或者第二CSI,所述第一回复或者第二CSI是终端根据第一CSI和第二CSI进行比较确定反馈的,所述第二CSI是终端根据所述第二参考信号测量处理得到的,所述第一回复用于表示所述终端已接收到第二参考信号,且无需反馈所述第二CSI。
其中,所述第一参考信号和第二参考信号是网络设备发送的相邻的两个参考信号。
在本申请的一种实施方式中,发送第二参考信号,包括:
根据所述第一CSI和AI模型,进行CSI预测,得到预测的第五CSI;
根据所述第五CSI,生成相应的预编码矩阵;
根据所述预编码矩阵,确定第一波束赋形;
通过所述第一波束赋形发送第二参考信号。
在本申请的一种实施方式中,在接收第一回复之后,所述方法还包括:
根据所述第一CSI和AI模型,进行CSI预测,得到预测的第六CSI;
根据所述第六CSI,生成相应的预编码矩阵;
根据所述预编码矩阵,确定第二波束赋形;
通过所述第二波束赋形发送第三参考信号。
当终端向基站反馈第一回复则代表当前信道特性相关性较强,网络设备 可以继续预测CSI。
在本申请的一种实施方式中,在接收第二CSI之后,所述方法还包括:
根据所述第二CSI和AI模型,进行CSI预测,得到预测的第七CSI;
根据所述第七CSI,生成相应的预编码矩阵;
根据所述预编码矩阵,确定第三波束赋形;
通过所述第三波束赋形发送第四参考信号。
在本申请的一种实施方式中,所述方法还包括:
配置所述相关性阈值,所述相关性阈值表示所述网络设备根据服务质量需求得到所能接受的最小CSI时延多普勒谱相关性。
在本申请的实施例中,利用网络设备进行信道预测以及网络设备设定的相关性阈值并由终端判别信道特性是否发生了网络设备的AI模型无法接受的改变,使得终端能以更低的频率反馈CSI,从而降低反馈开销,并且通过在终端进行的相关性计算以及相关性门限比较,当信道状态发生明显改变时终端可以更及时地反馈实测的CSI,增加信道测量机制的可靠性。
参见图6a和图6b,具体步骤包括:
步骤1:基站配置并向终端发送CSI-RS组的相关参数。
可选地,CSI-RS组的相关参数包括以下一项或多项:
(1)每个CSI-RS组中包含的参考信号数(N);
(2)CSI-RS组内相邻参考信号间隔(t);
(3)CSI-RS组组间间隔(T);
(4)CSI相关性阈值(E)。
本文中的CSI相关性阈值也可以称为相关性阈值。
在实际部署前,可能需要一定实测或仿真工作确定不同环境运动状态下不同时延多普勒谱间的相关性对信道预测模型性能的影响,使基站能够根据其服务质量(Quality of Service,QoS)需求得到所能接受的最小CSI时延多普勒谱相关性,即CSI相关性阈值E。
步骤2:基站向终端发送第一CSI-RS组。
本文中的第一CSI-RS组也可以称为第一参考信号。
步骤3:终端对接收到的第一CSI-RS组进行测量估计,得到多个第三CSI, 并对其进行辛傅里叶变换转化为时延多普勒域上的第一CSI。
本文中的测量估计也可以称为CSI-RS脉冲测量估计。
本文中的多个第三CSI也可以称为第三CSI时频空域信息,或者称为第三CSI时频空域结果,或者称为第三CSI反馈,或者称为第三CSI实测结果。
本文中的第一CSI也可以称为第一CSI实测结果,或者称为第一CSI实测多普勒信息,或者称为第一CSI反馈。
步骤4:终端压缩并向基站反馈时延多普勒域上的第一CSI。
步骤5:基站对终端发送的CSI解压缩,并对其进行逆辛傅里叶变换恢复为第一CSI,基站根据第一CSI和AI模型对后续时刻的第五CSI进行预测,并生成相应的预编码矩阵。
本文中的AI模型也可以称为AI预训练模型。
本文中的第五CSI也可以称为为第五CSI时频空域信息,或者称为CSI预测信息,或者称为CSI预测结果。
步骤6:基站根据第五CSI得到的预编码矩阵进行波束赋形。
步骤7:基站在发送第一CSI-RS组之后,以CSI-RS组组间间隔(T)向终端发送第二CSI-RS组。
步骤8:终端对第二CSI-RS组进行测量估计,得到多个第四CSI并对其进行辛傅里叶变换转化为时延多普勒域上的第二CSI。
本文中的多个第四CSI也可以称为为第四CSI时频空域信息,或者称为第四CSI时频空域结果,或者称为第四CSI反馈,或者称为第四CSI实测结果。
本文中的第二CSI也可以称为第二CSI实测结果,或者称为第二CSI实测多普勒信息,或者称为第二CSI反馈。
步骤9:终端将步骤3得到的第一CSI与步骤8中第二CSI进行信道相关性计算,若计算得到的信道相关性大于CSI相关性阈值(E),执行步骤10a,若计算得到的信道相关性小于CSI相关性阈值(E),执行步骤10b。
终端具有对不同时刻下的时延多普勒谱进行相关性计算以判断终端与周围环境间的相对运动状态是否发生显著改变的能力,终端将最近一次反馈给基站的时延多普勒谱作为参考结果,与后续由CSI-RS解析得到的时延多普勒谱进行相关性运算。
因为CSI时延多普勒谱是一个二维张量,因此可以利用矩阵的相关系数来表示不同时延多普勒谱间的相关性,计算方法如下:
其中,A、B为时延多普勒谱矩阵,m、n为矩阵的行与列, r为矩阵相关系数。
r越接近1说明这两个时延多普勒谱相关性越强,这意味着终端与周围环境间的相对运动状态以及与之对应的信道状态越相近,也表明此时的信道预测结果越准确。
终端根据周期性下发的CSI-RS解析得到的CSI,判定基站是否有能力预测后续CSI,并决定是否向基站反馈最新的CSI实测结果。
通过在基站引入信道预测并在终端加入信道相关性监测机制以在保证CSI预测准确性的情况下减小CSI反馈频率。
步骤10a:终端向基站反馈第一回复,该第一回复用于表示终端已接收到第二参考信号但无需反馈其他信息,然后执行步骤11a,参考图6a。
本文中的第一回复也可以称为判别结果回复信令,或者第一回复(Acknowledge character,ACK)。
本实施例中,终端接收到CSI-RS后存在两种情况,一种是时延多普勒谱相关性小于CSI相关性阈值,此时与现有流程类似,向基站反馈CSI;另一种是,当时延多普勒谱相关性大于CSI相关性阈值时,利用基站的预测CSI能力,终端可以不反馈CSI,但若终端不向基站反馈任何信息,则基站无法确定是由于当前信道特性相关性较强而没有反馈,还是由于CSI-RS或CSI反馈在传输过程中丢失,因此为明确终端未向基站反馈CSI的原因,在本实施例中引入第一回复,当终端向基站反馈第一回复则代表当前信道特性相关性较强,基站可以继续预测CSI。
步骤10b:终端压缩并向基站反馈时延多普勒域上的第二CSI,然后执行步骤11b,参考图6b。
本文中的第二CSI也可以称为CSI实测结果、或者称为CSI实测多普勒信息。
步骤11a:基站接收到第一回复后,继续对后续时刻的CSI进行预测,并 生成相应的预编码矩阵,然后执行步骤12,参考图6a。
比如,根据所述第一CSI和AI模型,进行CSI预测,得到预测的第六CSI;根据所述第六CSI,生成相应的预编码矩阵;根据所述预编码矩阵,确定第二波束赋形;通过所述第二波束赋形发送第三参考信号。
步骤11b:基站接收终端发送的第二CSI,利用第二CSI和AI模型,对后续时刻的CSI进行预测,并生成相应的预编码矩阵,然后执行步骤12,参考图6b。
比如,根据所述第二CSI和AI模型,进行CSI预测,得到预测的第七CSI;根据所述第七CSI,生成相应的预编码矩阵;根据所述预编码矩阵,确定第三波束赋形;通过所述第三波束赋形发送第四参考信号。
步骤12:基站根据CSI预测信息得到的预编码矩阵进行波束赋形,之后返回步骤7。
在实施例中,基站具有基于AI模型和CSI时频空域信息(即实测信道信息)进行CSI预测(即时序序列预测)的能力,当前对于时序序列进行预测的AI模型可以是循环神经网络(Recurrent Neural Network,RNN)网络。
RNN网络包括长短期记忆(Long short-term memory,LSTM)网络,LSTM的结构如图7所示。
其中,xt为t时刻的网络输入,yt为t时刻的网络输出,ht和ct表示t时刻网络单元的状态。
LSTM网络为学习时序特征需要在网络单元中加入三个门,首先是遗忘门,遗忘门主要是对上一个节点传进来的输入进行选择性忘记,具体来说是通过计算得到的zf作为遗忘门控,来控制上一个状态的ct-1哪些需要留哪些需要忘;之后是记忆门,记忆门会将输入xt有选择性地进行记忆,当前的输入内容由前面计算得到的z表示。而选择记忆何种信息的门控信号则由zi进行控制;最后是输出门,输出门将决定哪些信息将会被当成当前状态的输出,主要通过zo进行控制,并且还对记忆门得到输出的co进行了放缩(通过tanh激活函数进行放缩)。因此,LSTM网络可以通过门控状态来控制传输状态,记住需要长时间记忆的信息,舍弃不重要的信息。除经典的LSTM网络外,近年来新出现的门控循环单元(Gated Recurrent Unit,GRU)网络也可以达到类似的效果且引入的新参数更少,更易训练,也可以作为信道预测方法的选项之一。对于 这类AI模型,其可行的预测方式如图8所示。
在预测过程中,首先将实测CSI结果作为模型输入,得到一个CSI预测结果,在下一轮预测过程中,在上一轮得到的CSI预测结果可视为新的已知CSI,将最早的实测数据从模型输入中剔除并加入上一轮的CSI预测结果作为模型输入使得输入矩阵形状保持不变,以此类推,理论上此预测过程可以一直持续下去,但实际上考虑到模型输入中预测结果的增加以及信道本身的特性变化,为保证预测结果的质量还需加入其它机制对结束预测的时间加以限制。
参见图9,本申请实施例提供一种CSI反馈的装置,应用于终端,该装置900包括:
第一获取模块901,用于获取第一CSI和第二CSI,所述第一CSI和第二CSI是所述终端分别基于第一参考信号和第二参考信号进行测量处理得到的;
第一确定模块902,用于根据所述第一CSI和第二CSI进行信道比较,确定反馈所述第二CSI或第一回复,所述第一回复用于表示所述终端已接收到所述第二参考信号,且无需反馈所述第二CSI。
在本申请的一种实施方式中,第一获取模块901进一步用于:
对第一参考信号进行测量,根据所述第一参考信号的测量结果得到多个第三CSI;
根据所述多个第三CSI,得到时延多普勒域上的所述第一CSI;
对第二参考信号进行测量,根据所述第二参考信号的测量结果得到多个第四CSI;
根据所述多个第四CSI,得到时延多普勒域上的第二CSI。
在本申请的一种实施方式中,所述第一参考信号包括第一CSI-RS组,所述第二参考信号包括第二CSI-RS组。
在本申请的一种实施方式中,所述第二参考信号是网络设备根据第一波束赋形发送的,所述第一波束赋形是所述网络设备根据预编码矩阵确定的,所述预编码矩阵是所述网络设备根据第五CSI生成的,所述第五CSI是所述网络设备根据第一CSI和AI模型预测得到的。
在本申请的一种实施方式中,第一获取模块901进一步用于:
对所述多个第三CSI进行辛傅里叶变换,转化为时延多普勒域上的第一 CSI;
对所述多个第四CSI进行辛傅里叶变换,转化为时延多普勒域上的第二CSI。
在本申请的一种实施方式中,第一确定模块902进一步用于:根据所述第一CSI和第二CSI进行信道相关性计算,并根据相关性计算结果,确定反馈所述第二CSI或第一回复。
可选地,第一确定模块902进一步用于:
根据所述第一CSI和第二CSI进行信道相关性计算,得到相关性计算结果;
将所述相关性计算结果与相关性阈值进行比较;
如果所述相关性计算结果大于或等于所述相关性阈值,则向网络设备发送第一回复,所述第一回复用于表示所述终端已接收到第二参考信号,且无需反馈所述第二CSI。
在本申请的一种实施方式中,所述装置还包括:
第二获取模块,用于获取所述网络设备配置的所述相关性阈值,所述相关性阈值表示所述网络设备根据服务质量需求得到所能接受的最小CSI时延多普勒谱相关性。
在本申请的一种实施方式中,所述装置还包括:
第一发送模块,用于如果所述相关性计算结果小于所述相关性阈值,则向网络设备发送所述第二CSI。
本申请实施例提供的装置能够实现图4所示的方法实施例实现的各个过程,并达到相同的技术效果,为避免重复,这里不再赘述。
参见图10,本申请实施例提供一种CSI反馈的装置,应用于网络设备,装置1000包括:
第二发送模块1001,用于发送第一参考信号;
第一接收模块1002,用于接收第一CSI,所述第一CSI是终端根据所述第一参考信号测量处理得到的;
第三发送模块1003,用于发送第二参考信号;
第二接收模块1004,用于接收第一回复或者第二CSI,所述第一回复或者第二CSI是终端根据第一CSI和第二CSI进行比较确定反馈的,所述第二CSI是 终端根据所述第二参考信号测量处理得到的,所述第一回复用于表示所述终端已接收到第二参考信号,且无需反馈所述第二CSI。
在本申请的一种实施方式中,第三发送模块1003进一步用于:
根据所述第一CSI和AI模型,进行CSI预测,得到预测的第五CSI;
根据所述第五CSI,生成相应的预编码矩阵;
根据所述预编码矩阵,确定第一波束赋形;
通过所述第一波束赋形发送第二参考信号。
在本申请的一种实施方式中,所述装置还包括:第一处理模块,用于:
根据所述第一CSI和AI模型,进行CSI预测,得到预测的第六CSI;
根据所述第六CSI,生成相应的预编码矩阵;
根据所述预编码矩阵,确定第二波束赋形;
通过所述第二波束赋形发送第三参考信号。
在本申请的一种实施方式中,所述装置还包括:第二处理模块,用于:
根据所述第二CSI和AI模型,进行CSI预测,得到预测的第七CSI;
根据所述第七CSI,生成相应的预编码矩阵;
根据所述预编码矩阵,确定第三波束赋形;
通过所述第三波束赋形发送第四参考信号。
在本申请的一种实施方式中,所述装置还包括:
配置模块,用于配置所述相关性阈值,所述相关性阈值表示所述网络设备根据服务质量需求得到所能接受的最小CSI时延多普勒谱相关性。
本申请实施例提供的装置能够实现图5所示的方法实施例实现的各个过程,并达到相同的技术效果,为避免重复,这里不再赘述。
如图11所示,本申请实施例还提供一种通信设备1100,包括处理器1101,存储器1102,存储在存储器1102上并可在所述处理器1101上运行的程序或指令,该程序或指令被处理器1101执行时实现上述图4或图5方法实施例的各个过程,且能达到相同的技术效果。为避免重复,这里不再赘述。
本申请实施例还提供一种可读存储介质,所述可读存储介质上存储有程序或指令,该程序或指令被处理器执行时实现上述图4或图5所示方法实施例的各个过程,且能达到相同的技术效果,为避免重复,这里不再赘述。
其中,所述处理器为上述实施例中所述的终端中的处理器。所述可读存储介质,包括计算机可读存储介质,如计算机只读存储器(Read-Only Memory,ROM)、随机存取存储器(Random Access Memory,RAM)、磁碟或者光盘等。
结合本申请公开内容所描述的方法或者算法的步骤可以硬件的方式来实现,也可以由在处理器执行软件指令的方式来实现。软件指令可以由相应的软件模块组成,软件模块可以被存放于RAM、闪存、ROM、可擦除可编程只读存储器(Erasable Programmable Read-Only Memory,EPROM)、电可擦除可编程只读存储器(Electrically Erasable Programmable Read-Only Memory,EEPROM)、寄存器、硬盘、移动硬盘、只读光盘或者本领域熟知的任何其它形式的存储介质中。一种示例性的存储介质耦合至处理器,从而使处理器能够从该存储介质读取信息,且可向该存储介质写入信息。当然,存储介质也可以是处理器的组成部分。处理器和存储介质可以携带在应用专用集成电路(Application Specific Integrated Circuit,ASIC)中。另外,该ASIC可以携带在核心网接口设备中。当然,处理器和存储介质也可以作为分立组件存在于核心网接口设备中。
本领域技术人员应该可以意识到,在上述一个或多个示例中,本申请所描述的功能可以用硬件、软件、固件或它们的任意组合来实现。当使用软件实现时,可以将这些功能存储在计算机可读介质中或者作为计算机可读介质上的一个或多个指令或代码进行传输。计算机可读介质包括计算机存储介质和通信介质,其中通信介质包括便于从一个地方向另一个地方传送计算机程序的任何介质。存储介质可以是通用或专用计算机能够存取的任何可用介质。
以上所述的具体实施方式,对本申请的目的、技术方案和有益效果进行了进一步详细说明,所应理解的是,以上所述仅为本申请的具体实施方式而已,并不用于限定本申请的保护范围,凡在本申请的技术方案的基础之上,所做的任何修改、等同替换、改进等,均应包括在本申请的保护范围之内。
本领域内的技术人员应明白,本申请实施例可提供为方法、系统、或计算机程序产品。因此,本申请实施例可采用完全硬件实施例、完全软件实施例、或结合软件和硬件方面的实施例的形式。而且,本申请实施例可采用在一个或多个其中包含有计算机可用程序代码的计算机可用存储介质(包括但 不限于磁盘存储器、只读光盘(Compact Disc Read-Only Memory,CD-ROM)、光学存储器等)上实施的计算机程序产品的形式。
本申请实施例是参照根据本申请实施例的方法、设备(系统)、和计算机程序产品的流程图和/或方框图来描述的。应理解可由计算机程序指令实现流程图和/或方框图中的每一流程和/或方框、以及流程图和/或方框图中的流程和/或方框的结合。可提供这些计算机程序指令到通用计算机、专用计算机、嵌入式处理机或其他可编程数据处理设备的处理器以产生一个机器,使得通过计算机或其他可编程数据处理设备的处理器执行的指令产生用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的装置。
这些计算机程序指令也可存储在能引导计算机或其他可编程数据处理设备以特定方式工作的计算机可读存储器中,使得存储在该计算机可读存储器中的指令产生包括指令装置的制造品,该指令装置实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能。
这些计算机程序指令也可装载到计算机或其他可编程数据处理设备上,使得在计算机或其他可编程设备上执行一系列操作步骤以产生计算机实现的处理,从而在计算机或其他可编程设备上执行的指令提供用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的步骤。
显然,本领域的技术人员可以对本申请实施例进行各种改动和变型而不脱离本申请的精神和范围。这样,倘若本申请实施例的这些修改和变型属于本申请权利要求及其等同技术的范围之内,则本申请也意图包含这些改动和变型在内。

Claims (17)

  1. 一种信道状态信息CSI反馈的方法,应用于终端,所述方法包括:
    获取第一CSI和第二CSI,所述第一CSI和第二CSI是所述终端分别基于第一参考信号和第二参考信号进行测量处理得到的;
    根据所述第一CSI和第二CSI进行信道比较,确定反馈所述第二CSI或第一回复,所述第一回复用于表示所述终端已接收到所述第二参考信号,且无需反馈所述第二CSI。
  2. 根据权利要求1所述的方法,其中,获取第一CSI和第二CSI,包括:
    对所述第一参考信号进行测量,根据所述第一参考信号的测量结果得到多个第三CSI;
    根据所述多个第三CSI得到所述第一CSI,所述第一CSI为时延多普勒域上的CSI;
    对所述第二参考信号进行测量,根据所述第二参考信号的测量结果得到多个第四CSI;
    根据所述多个第四CSI得到所述第二CSI,所述第二CSI为时延多普勒域上的CSI。
  3. 根据权利要求2所述的方法,其中,所述第一参考信号包括第一信道状态信息参考信号CSI-RS组,所述第二参考信号包括第二CSI-RS组。
  4. 根据权利要求2所述的方法,其中,所述第二参考信号是网络设备根据第一波束赋形发送的,所述第一波束赋形是所述网络设备根据预编码矩阵确定的,所述预编码矩阵是所述网络设备根据第五CSI生成的,所述第五CSI是所述网络设备根据接收到的所述第一CSI和人工智能AI模型预测得到的。
  5. 根据权利要求2所述的方法,其中,根据所述多个第三CSI得到所述第一CSI,包括:
    对所述多个第三CSI进行辛傅里叶变换,获得所述第一CSI;
    根据所述多个第四CSI得到所述第二CSI,包括:
    对所述多个第四CSI进行辛傅里叶变换,获得所述第二CSI。
  6. 根据权利要求1所述的方法,其中,根据所述第一CSI和第二CSI进行 信道比较,确定反馈所述第二CSI或第一回复,包括:
    根据所述第一CSI和第二CSI进行信道相关性计算,并根据相关性计算结果,确定反馈所述第二CSI或第一回复。
  7. 根据权利要求6所述的方法,其中,根据所述第一CSI和第二CSI进行信道相关性计算,并根据相关性计算结果,确定反馈所述第二CSI或第一回复,包括:
    根据所述第一CSI和第二CSI进行信道相关性计算,得到相关性计算结果;
    将所述相关性计算结果与相关性阈值进行比较;
    如果所述相关性计算结果大于或等于所述相关性阈值,则向网络设备发送第一回复;
    如果所述相关性计算结果小于所述相关性阈值,则向网络设备发送所述第二CSI。
  8. 根据权利要求7所述的方法,所述方法还包括:
    获取网络设备配置的所述相关性阈值,所述相关性阈值表示所述网络设备根据服务质量需求得到所能接受的最小CSI时延多普勒谱相关性。
  9. 一种CSI反馈的方法,应用于网络设备,所述方法包括:
    发送第一参考信号;
    接收第一CSI,所述第一CSI是终端根据所述第一参考信号测量处理得到的;
    发送第二参考信号;
    接收第一回复或者第二CSI,所述第一回复或者第二CSI是终端根据所述第一CSI和所述第二CSI进行比较确定反馈的,所述第二CSI是终端根据所述第二参考信号测量处理得到的,所述第一回复用于表示所述终端已接收到所述第二参考信号,且无需反馈所述第二CSI。
  10. 根据权利要求9所述的方法,其中,发送第二参考信号,包括:
    根据所述第一CSI和AI模型,进行CSI预测,得到预测的第五CSI;
    根据所述第五CSI,生成相应的预编码矩阵;
    根据所述预编码矩阵,确定第一波束赋形;
    通过所述第一波束赋形发送第二参考信号。
  11. 根据权利要求9所述的方法,其中,在接收第一回复之后,所述方法还包括:
    根据所述第一CSI和AI模型,进行CSI预测,得到预测的第六CSI;
    根据所述第六CSI,生成相应的预编码矩阵;
    根据所述预编码矩阵,确定第二波束赋形;
    通过所述第二波束赋形发送第三参考信号。
  12. 根据权利要求9所述的方法,其中,在接收第二CSI之后,所述方法还包括:
    根据所述第二CSI和AI模型,进行CSI预测,得到预测的第七CSI;
    根据所述第七CSI,生成相应的预编码矩阵;
    根据所述预编码矩阵,确定第三波束赋形;
    通过所述第三波束赋形发送第四参考信号。
  13. 根据权利要求9所述的方法,所述方法还包括:
    配置相关性阈值,所述相关性阈值表示所述网络设备根据服务质量需求得到所能接受的最小CSI时延多普勒谱相关性。
  14. 一种CSI反馈的装置,应用于终端,所述装置包括:
    第一获取模块,用于获取第一CSI和第二CSI,所述第一CSI和第二CSI是所述终端分别基于第一参考信号和第二参考信号进行测量处理得到的;
    第一确定模块,用于根据所述第一CSI和第二CSI进行信道比较,确定反馈所述第二CSI或第一回复,所述第一回复用于表示所述终端已接收到所述第二参考信号,且无需反馈所述第二CSI。
  15. 一种CSI反馈的装置,应用于网络设备,所述装置包括:
    第二发送模块,用于发送第一参考信号;
    第一接收模块,用于接收第一CSI,所述第一CSI是终端根据所述第一参考信号测量处理得到的;
    第三发送模块,用于发送第二参考信号;
    第二接收模块,用于接收第一回复或者第二CSI,所述第一回复或者第二CSI是终端根据第一CSI和第二CSI进行比较确定反馈的,所述第二CSI是终端根据所述第二参考信号测量处理得到的,所述第一回复用于表示所述终端已 接收到所述第二参考信号,且无需反馈所述第二CSI。
  16. 一种通信设备,包括处理器,存储器及存储在所述存储器上并可在所述处理器上运行的程序或指令,所述程序或指令被所述处理器执行时实现如权利要求1至13中任一项所述的方法的步骤。
  17. 一种可读存储介质,所述可读存储介质上存储程序或指令,所述程序或指令被处理器执行时实现如权利要求1至13中任一项所述的方法的步骤。
PCT/CN2023/103970 2022-07-05 2023-06-29 Csi反馈的方法、装置、设备及可读存储介质 WO2024007957A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202210846696.4 2022-07-05
CN202210846696.4A CN117394891A (zh) 2022-07-05 2022-07-05 Csi反馈的方法、装置、设备及可读存储介质

Publications (1)

Publication Number Publication Date
WO2024007957A1 true WO2024007957A1 (zh) 2024-01-11

Family

ID=89439783

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/103970 WO2024007957A1 (zh) 2022-07-05 2023-06-29 Csi反馈的方法、装置、设备及可读存储介质

Country Status (2)

Country Link
CN (1) CN117394891A (zh)
WO (1) WO2024007957A1 (zh)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111769859A (zh) * 2019-04-02 2020-10-13 电信科学技术研究院有限公司 一种信道状态信息csi上报的方法和设备
WO2021253205A1 (en) * 2020-06-16 2021-12-23 Qualcomm Incorporated Frequency domain basis adjustments for channel state information reporting
CN114079493A (zh) * 2020-08-13 2022-02-22 华为技术有限公司 一种信道状态信息测量反馈方法及相关装置
KR20220029039A (ko) * 2020-09-01 2022-03-08 주식회사 엘지유플러스 기지국 장치 및 이의 빔포밍 방법

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111769859A (zh) * 2019-04-02 2020-10-13 电信科学技术研究院有限公司 一种信道状态信息csi上报的方法和设备
WO2021253205A1 (en) * 2020-06-16 2021-12-23 Qualcomm Incorporated Frequency domain basis adjustments for channel state information reporting
CN114079493A (zh) * 2020-08-13 2022-02-22 华为技术有限公司 一种信道状态信息测量反馈方法及相关装置
KR20220029039A (ko) * 2020-09-01 2022-03-08 주식회사 엘지유플러스 기지국 장치 및 이의 빔포밍 방법

Also Published As

Publication number Publication date
CN117394891A (zh) 2024-01-12

Similar Documents

Publication Publication Date Title
CN115037608B (zh) 量化的方法、装置、设备及可读存储介质
US20220103211A1 (en) Method and device for switching transmission methods in massive mimo system
WO2023246618A1 (zh) 信道矩阵处理方法、装置、终端及网络侧设备
CN115441914A (zh) 一种通信方法及装置
JP2024512358A (ja) 情報報告方法、装置、第1機器及び第2機器
WO2023071683A1 (zh) 用于反馈信道状态的方法和装置
WO2023185978A1 (zh) 信道特征信息上报及恢复方法、终端和网络侧设备
WO2024007957A1 (zh) Csi反馈的方法、装置、设备及可读存储介质
US12021665B2 (en) Methods and wireless network for selecting pilot pattern for optimal channel estimation
WO2022236785A1 (zh) 信道信息的反馈方法、收端设备和发端设备
KR20230170975A (ko) 채널 상태 피드백을 위한 신경망을 사용하는 무선 네트워크
WO2023179476A1 (zh) 信道特征信息上报及恢复方法、终端和网络侧设备
US20230412230A1 (en) Systems, methods, and apparatus for artificial intelligence and machine learning based reporting of communication channel information
WO2023185980A1 (zh) 信道特征信息传输方法、装置、终端及网络侧设备
WO2023179473A1 (zh) 信道特征信息上报及恢复方法、终端和网络侧设备
WO2023179540A1 (zh) 信道预测方法、装置及无线通信设备
WO2023185995A1 (zh) 信道特征信息传输方法、装置、终端及网络侧设备
WO2024037380A1 (zh) 信道信息处理方法、装置、通信设备及存储介质
US20240129004A1 (en) Method and apparatus for transmitting channel state information
WO2024037321A1 (zh) Ai模型训练方法、装置、设备及可读存储介质
WO2024007949A1 (zh) Ai模型处理方法、装置、终端及网络侧设备
WO2024051594A1 (zh) 信息传输方法、ai网络模型训练方法、装置和通信设备
WO2022236788A1 (zh) 通信方法、设备及存储介质
WO2023179570A1 (zh) 信道特征信息传输方法、装置、终端及网络侧设备
US20240063852A1 (en) Method and apparatus for learning-based channel matrix prediction

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 23834724

Country of ref document: EP

Kind code of ref document: A1