Channel state information feedback method, device, equipment, storage medium and computer program product

Abstract

The invention discloses a channel state information feedback method, a channel state information feedback device, channel state information feedback equipment, a storage medium and a computer program product, optimization is carried out aiming at the problem that overhead is large when a codebook is generated and transmitted, a feedback codebook is generated by utilizing a moving average method, and compared with an existing codebook generation mode, the channel state information feedback method has the advantages that the cost is low. Configuration parameters and channel state information test data volume required by the moving average method configured by the network side are unified, the number of actual measurement channel state information samples required for generating the same number of codebooks is smaller, and the overhead in the codebook generation and transmission process is effectively reduced.

Description

Technical Field

[0001] The present invention relates to the field of wireless communications, and in particular to a channel state information feedback method, apparatus, device, storage medium, and computer program product. Background Art

[0002] Channel state information (CSI) is a metric reported by the terminal to the network. Prior art typically uses a codebook-based feedback scheme. The terminal divides the measured CSI into k groups and then averages the CSI in each group to obtain k codebook vectors. Each codebook vector corresponds to a label, and during feedback, only the label corresponding to the codebook is fed back. However, to avoid performance losses caused by codebooks, the number of codebooks needs to be increased as much as possible to improve feedback accuracy. For example, if feedback is performed on the order of 1,000 to 10,000 codebooks, the downlink feedback required for each terminal to transmit the codebook alone would be hundreds of megabits to several gigabits. Furthermore, the uplink overhead required to transmit the full CSI to generate the codebook on the base station side would be even higher. Therefore, traditional codebook-based CSI feedback schemes suffer from high uplink and downlink transmission overhead. Summary of the Invention

[0003] The purpose of the embodiments of the present invention is to provide a channel state information feedback method, apparatus, device, storage medium and computer program product, which can reduce the uplink and downlink transmission overhead in the process of configuring a codebook.

[0004] To achieve the above objectives, an embodiment of the present invention provides a channel state information feedback method, which is applied to a terminal. The method includes:

[0005] Perform downlink channel measurement on the downlink reference signal sent by the network to obtain the original channel state information of the downlink channel;

[0006] Performing a sliding average calculation on the original channel state information according to preset configuration parameters to obtain a feedback codebook of the original channel state information;

[0007] Send the feedback codebook to the network.

[0008] As an improvement to the above solution, before performing downlink channel measurement on the reference signal sent by the network, the method further includes:

[0009] Receive configuration parameters sent by the network; wherein the configuration parameters include a sliding window length and a sliding distance.

[0010] As an improvement to the above solution, performing a sliding average calculation on the original channel state information according to preset configuration parameters to obtain a feedback codebook of the original channel state information includes:

[0011] Intercept local channel state information that meets the sliding window length in the original channel state information according to the sliding distance;

[0012] Perform an average calculation on the local channel state information to obtain a local codebook corresponding to the local channel state information;

[0013] Summarize the local codebooks of all local channel state information to obtain the feedback codebook of the original channel state information.

[0014] As an improvement to the above solution, the averaging of the local channel state information includes:

[0015] Perform an average calculation on the local channel state information in a parallel computing mode.

[0016] As an improvement to the above solution, the performing an average calculation on the local channel state information in a parallel computing mode includes:

[0017] Store the repeated channel state information in two consecutive sliding windows into a shared cache;

[0018] When the average calculation of the previous local channel state information has not ended, the channel state information in the shared cache participates in the average calculation of the subsequent local channel state information.

[0019] To achieve the above object, an embodiment of the present invention further provides a method for feedback of channel state information, which is applied to a network, and the method includes:

[0020] Receive the initial channel state information of the downlink channel sent by the terminal;

[0021] Input the initial channel state information and the uplink channel information into a channel information estimation model, so that the channel information estimation model outputs a full-channel estimation result.

[0022] As an improvement to the above solution, after obtaining the full-channel estimation result, the method further includes:

[0023] Perform a sliding average calculation on the full-channel estimation result according to preset configuration parameters to obtain a feedback codebook of the full-channel estimation result;

[0024] Send the feedback codebook to the terminal.

[0025] As an improvement to the above solution, the configuration parameters include a sliding window length and a sliding distance.

[0026] As an improvement to the above solution, the performing a sliding average calculation on the full-channel estimation result according to preset configuration parameters to obtain the codebook data of the full-channel estimation result includes:

[0027] intercepting local channel state information that satisfies a sliding window length from the full channel estimation result according to the sliding distance;

[0028] Performing an average calculation on the local channel state information to obtain a local codebook corresponding to the local channel state information;

[0029] All local codebooks of local channel state information are aggregated to obtain a feedback codebook of the original channel state information.

[0030] As an improvement to the above solution, before receiving the initial channel state information of the downlink channel sent by the terminal, the method further includes:

[0031] A downlink reference signal is sent to the terminal, so that the terminal obtains full channel state information of the downlink channel according to the downlink reference signal, and obtains partial channel state information from the full channel state information as the initial channel state information.

[0032] As an improvement to the above solution, before performing sliding average calculation on the full channel estimation result according to preset configuration parameters, the method further includes:

[0033] receiving an uplink reference signal sent by a terminal;

[0034] Perform uplink channel measurement on the uplink reference signal to obtain the uplink channel information.

[0035] As an improvement to the above solution, the terminal performs a sliding average calculation on the full amount of channel state information according to preset configuration parameters, and obtains a feedback codebook of the full amount of channel state information.

[0036] As an improvement of the above solution, the channel information estimation model includes a forget gate, an input gate and an output gate; wherein,

[0037] After receiving the initial channel state information and the uplink channel information, the channel information estimation model determines the uplink channel information to be retained through the forget gate according to the initial channel state information, fuses the initial channel state information passing through the input gate with the uplink channel information to be retained, and outputs the information through the output gate to obtain the full channel estimation result.

[0038] To achieve the above objectives, an embodiment of the present invention further provides a channel state information feedback device, comprising:

[0039] A downlink channel measurement module is used to perform downlink channel measurement on the downlink reference signal sent by the network to obtain the original channel state information of the downlink channel;

[0040] a feedback codebook generation module, configured to perform a sliding average calculation on the original channel state information according to preset configuration parameters to obtain a feedback codebook of the original channel state information;

[0041] A feedback codebook synchronization module is configured to send the feedback codebook to the network.

[0042] To achieve the above objectives, an embodiment of the present invention further provides a channel state information feedback device, comprising:

[0043] An original channel state information receiving module, configured to receive the initial channel state information of the downlink channel sent by the terminal;

[0044] The full channel estimation result generating module is configured to input the initial channel state information and uplink channel information into a channel information estimation model, so that the channel information estimation model outputs a full channel estimation result.

[0045] To achieve the above-mentioned objectives, an embodiment of the present invention further provides a channel state information feedback device, comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, the channel state information feedback method as described in any of the above-mentioned embodiments is implemented.

[0046] To achieve the above-mentioned purpose, an embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the channel state information feedback method as described in any of the above-mentioned embodiments.

[0047] To achieve the above objectives, an embodiment of the present invention further provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the channel state information feedback method as described in any of the above embodiments.

[0048] Compared with the prior art, the channel state information feedback method, apparatus, device, storage medium, and computer program product disclosed in the present invention optimize the problem of high overhead during codebook generation and transmission, and utilize a sliding average method to generate a feedback codebook. Compared with the existing codebook generation method, the configuration parameters and the amount of channel state information test data required by the sliding average method are uniformly configured by the network side, so that the number of measured channel state information samples required to generate the same number of codebooks is reduced, effectively reducing the overhead during codebook generation and transmission. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 This is a flow chart of a first channel state information feedback method provided by an embodiment of the present invention;

[0050] Figure 2 is another flowchart of the first channel state information feedback method provided by an embodiment of the present invention;

[0051] Figure 3 is a schematic diagram of generating a codebook by using a moving average calculation method provided by an embodiment of the present invention;

[0052] Figure 4 is a schematic diagram of generating a codebook by using a parallel moving average calculation method provided by an embodiment of the present invention;

[0053] Figure 5 is a flowchart of the second channel state information feedback method provided by an embodiment of the present invention;

[0054] Figure 6 is another flowchart of the second channel state information feedback method provided by an embodiment of the present invention;

[0055] Figure 7 is a schematic diagram of generating a codebook synchronously at both ends provided by an embodiment of the present invention;

[0056] Figure 8 is a working flowchart of a channel information estimation model provided by an embodiment of the present invention;

[0057] Figure 9 is a flowchart of the third channel state information feedback method provided by an embodiment of the present invention;

[0058] Figure 10 is another flowchart of the third channel state information feedback method provided by an embodiment of the present invention;

[0059] Figure 11 is a schematic diagram of generating a codebook asynchronously at both ends provided by an embodiment of the present invention;

[0060] Figure 12 is a structural block diagram of the first channel state information feedback device provided by an embodiment of the present invention;

[0061] Figure 13 is a structural block diagram of the second channel state information feedback device provided by an embodiment of the present invention;

[0062] Figure 14 is a structural block diagram of a channel state information feedback device provided by an embodiment of the present invention. Detailed implementation manners

[0063] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without making creative efforts belong to the scope of protection of the present invention.

[0064] In a wireless communication system, to achieve sufficient performance gain, the network requires accurate downlink channel state information (CSI) for beamforming, subcarrier allocation, and power control. Mathematically, CSI is usually represented as a matrix, where the row and column indices respectively correspond to antenna subcarrier index information. Due to the similarity of electromagnetic wave transmission between different antennas and subcarriers, there is an implicit correlation between the elements in the CSI matrix, which makes the entire CSI matrix compressible. Therefore, how to effectively compress CSI to reduce the feedback cost has become an important issue in the field of wireless communication. To solve the problem of large uplink and downlink transmission overheads caused by using the traditional codebook feedback CSI method in the prior art, the embodiments of the present invention provide three channel state information feedback methods. The first is that the terminal generates a feedback codebook and synchronizes it to the network; the second is that the network generates a feedback codebook and synchronizes it to the terminal; the third is that the terminal and the network generate codebooks asynchronously respectively. Among them, in the second and third methods, it is necessary to use a channel information estimation model on the network side to output the full-channel estimation result.

[0065] The terminal may be a device that includes a wireless transceiver function and can cooperate with the network to provide communication services for users. Specifically, the terminal may refer to a user equipment (UE), an access terminal, a user unit, a user station, a mobile station, a mobile device, a remote station, a remote terminal, a mobile device, a user terminal, a wireless communication device, a user agent, or a user device, etc. For example, the terminal device may be a cellular phone, a cordless phone, a Session Initiation Protocol (SIP) phone, a Wireless Local Loop (WLL) station, a Personal Digital Assistant (PDA), a handheld device with wireless communication function, a computing device, or other processing devices connected to a wireless modem, a vehicle-mounted device, a wearable device, etc.

[0066] The network is a device used to communicate with the terminal device, which can be a base station (Base Transceiver Station, BTS) in the Global System for Mobile Communication (Global System for Mobile Communication, GSM) or Code Division Multiple Access (Code Division Multiple Access, CDMA) communication system, or a base station (Node B, NB) in the Wideband Code Division Multiple Access (WCDMA) system, or an evolved Node B (eNB or eNode B) in the LTE system.

[0067] See also Figure 1 , Figure 1 This is a flowchart of a first channel state information feedback method provided by an embodiment of the present invention. The channel state information feedback method according to an embodiment of the present invention is applied to a terminal and adopts the first method described above, i.e., the terminal generates a feedback codebook and synchronizes it to the network. The method includes:

[0068] S11. Perform downlink channel measurement on a downlink reference signal sent by the network to obtain original channel state information of the downlink channel;

[0069] S12. Performing a sliding average calculation on the original channel state information according to preset configuration parameters to obtain a feedback codebook of the original channel state information;

[0070] S13: Send the feedback codebook to the network.

[0071] Specifically, before executing step S11, the method further includes: receiving configuration parameters sent by the network; wherein the configuration parameters include a sliding window length and a sliding distance.

[0072] For example, see Figure 2 , Figure 2 is another flow chart of the first channel state information feedback method provided by an embodiment of the present invention, according to Figure 2 The process describes the above steps S11 to S13 in detail:

[0073] 1. The network sends configuration parameters and CSI test data volume; the CSI test data volume is used to inform the terminal of the amount of CSI data that can be measured (i.e., the original channel state information).

[0074] 2. The network sends a corresponding number of downlink reference signals to the terminal according to the CSI test data volume. If the downlink reference signal is CSI-RS (Channel State Information - Reference Signal), for channel sounding purposes, CSI-RS is often used for measuring the characteristics of the wireless channel so that the network and the terminal can perform correct modulation, coding, beamforming, etc. It should be noted that in order to avoid measurement signaling storms, the network can adopt a batch sending method when sending downlink reference signals to the terminal.

[0075] 3. According to the CSI test data volume, the terminal can know whether the downlink reference signals sent in batches by the network have been sent completely. The terminal performs downlink channel measurement on the downlink reference signals sent by the network each time to obtain the original channel state information of the downlink channel.

[0076] 4. The terminal performs a sliding average calculation on the original channel state information according to the sliding distance and the sliding window length to obtain a feedback codebook.

[0077] 5. The terminal sends the feedback codebook to the network to complete the synchronization of the feedback codebook.

[0078] Specifically, in order to reduce the amount of original channel state information data required for generating the codebook on the network side and improve the codebook generation speed, a sliding average calculation method is used to generate the codebook. Then step S12 specifically includes:

[0079] S121. Intercept local channel state information that meets the sliding window length in the original channel state information according to the sliding distance;

[0080] S122. Perform an average calculation on the local channel state information to obtain a local codebook corresponding to the local channel state information;

[0081] S123. Aggregate the local codebooks of all local channel state information to obtain the feedback codebook of the original channel state information.

[0082] Exemplarily, see Figure 3 , Figure 3 which is a schematic diagram of generating a codebook by using a sliding average calculation method provided by an embodiment of the present invention. Figure 3Among them, w1, w2, w3, w4, w5, and w6 are channel state information measured based on the downlink reference signals sent in batches by the network. After summarization, the original channel state information is obtained. The network side configures the sliding window length and the sliding distance of the window when generating different codebooks. The number of generated codebooks can be calculated from the actual measured data volume of CSI. Assume that the sliding distance is 2 and the sliding window length is 4. At this time, every two channel state information are taken to obtain channel state information with a length of 4. Taking this channel state information with a length of 4 as the local channel state information, for example, taking w1, w2, w3, w4 as the local channel state information, averaging w1, w2, w3, w4 to obtain the corresponding local codebook A1. Similarly, taking w3, w4, w5, w6 as the local channel state information, averaging w3, w4, w5, w6 to obtain the corresponding local codebook A2.

[0083] It should be noted that the method of averaging the matrix can refer to the prior art. In the embodiments of the present invention, a method of averaging a single channel state information is provided: Assume that the matrix of channel state information w1 contains n*m elements, n is the number of rows, and m is the number of columns. Taking each row of elements as the calculation basis, calculating the average value of m elements in each row can obtain n average values. Finally, summarizing the n average values of w1, w2, w3, w4 to obtain the local codebook A1.

[0084] In the embodiments of the present invention, compared with the existing codebook generation methods, the method of generating codebooks by using the sliding average calculation method can repeatedly intercept the channel state information of the front and back windows due to the sliding window. For the same actual measured data volume of CSI, the number of generated codebooks in the traditional codebook generation method is much less than the number of generated codebooks in the embodiments of the present invention using the sliding average calculation method. The embodiments of the present invention can generate more codebooks. In this way, the network side does not need to send a large number of downlink reference signals to the terminal for measurement, thereby reducing the uplink and downlink transmission overhead when the network and the terminal perform data interaction.

[0085] Specifically, the averaging of the local channel state information includes: performing average calculation on the local channel state information in a parallel computing mode.

[0086] Exemplarily, the performing average calculation on the local channel state information in a parallel computing mode includes: storing the repeated channel state information in two consecutive sliding windows into a shared cache; when the average calculation of the previous local channel state information has not ended, the channel state information in the shared cache participates in the average calculation of the subsequent local channel state information. See Figure 4 , Figure 4 is a schematic diagram of generating a codebook by using the parallel sliding average calculation method provided by the embodiments of the present invention. Figure 4Among them, t1, t2, and t3 are time periods for parallel computing. The three time periods can perform calculations synchronously. The repeated channel state information in two consecutive sliding windows can share the cache. When the preamble codebook calculation has not yet ended, some CSI samples in the cache can directly participate in the calculation of subsequent codebook generation. The longer the sliding window and the shorter the sliding distance, the higher the utilization rate of the measured CSI samples.

[0087] In the embodiments of the present invention, a parallel computing mode is adopted to perform average calculation on the local channel state information in parallel, which can improve the generation speed and efficiency of the codebook.

[0088] Furthermore, for the setting of the sliding window length and the sliding distance, the network side can pre-judge according to the received feedback channel modeling parameters. The sliding window length is inversely proportional to the Doppler shift parameter of the terminal. For example, on a high-speed moving train, when the Doppler shift is high at this time, when the network side receives the Doppler shift parameter fed back by the terminal, the codebook collected at the current moment is classified as the codebook in the high-speed moving scenario, and the sliding window length can be considered to be reduced to adapt to a shorter channel stationary interval; when the terminal is in a relatively open scenario where the distribution of backscatterers is not complex, the network side receives smaller parameters such as delay and angular spread. At this time, the network side can increase the sliding window length. When the channel environment is not complex, only a small number of codebooks can reflect the current channel characteristics.

[0089] Compared with the prior art, the channel state information feedback method disclosed in the present invention optimizes the problem of large overhead during the generation and transmission of the codebook. The sliding average method is used to generate the feedback codebook. Compared with the existing codebook generation methods, the network side uniformly configures the configuration parameters and the test data volume of the channel state information required by the sliding average method, so that the number of samples of the measured channel state information required to generate the same number of codebooks is less, effectively reducing the overhead during the codebook generation and transmission process.

[0090] For the second and third channel state information feedback methods, it is necessary to use a channel information estimation model on the network side to output the full-channel estimation result. At this time, the channel state information feedback method includes:

[0091] Receiving the initial channel state information of the downlink channel sent by the terminal;

[0092] Inputting the initial channel state information and the uplink channel information into the channel information estimation model so that the channel information estimation model outputs the full-channel estimation result.

[0093] See Figure 5 , Figure 5It is a flowchart of the second channel state information feedback method provided by an embodiment of the present invention. The channel state information feedback method described in the embodiment of the present invention is applied to a network and adopts the above-mentioned second method, that is, the network generates a feedback codebook and synchronizes it to the terminal. The method includes:

[0094] S21. Receive the initial channel state information of the downlink channel sent by the terminal;

[0095] S22. Input the initial channel state information and the uplink channel information into a channel information estimation model so that the channel information estimation model outputs a full-channel estimation result;

[0096] S23. Perform a moving average calculation on the full-channel estimation result according to preset configuration parameters to obtain a feedback codebook of the full-channel estimation result;

[0097] S24. Send the feedback codebook to the terminal.

[0098] Specifically, before performing step S21, the method further includes: sending a downlink reference signal to the terminal so that the terminal obtains the full-channel state information of the downlink channel according to the downlink reference signal, and obtaining partial channel state information from the full-channel state information as the initial channel state information.

[0099] Exemplarily, refer to Figure 6 , Figure 6 It is another flowchart of the second channel state information feedback method provided by an embodiment of the present invention. The above steps S21 to S24 are described in detail according to Figure 6 :

[0100] 1. The network configures the uplink channel data volume and the CSI feedback data volume, and sends the uplink channel data volume and the CSI feedback data volume to the terminal to synchronize the configuration with the terminal. The uplink channel data volume is used to indicate that the terminal needs to send uplink reference signals satisfying this data volume, and the CSI feedback data volume is used to indicate the CSI data volume that the terminal needs to feedback (that is, the initial channel state information). In addition, the configuration parameters are pre-stored in the network, and the configuration parameters include the sliding window length and the sliding distance.

[0101] 2. The terminal sends an uplink reference signal (ARS) of a corresponding amount to the network based on the amount of uplink channel data. If the ARS is an SRS (Sounding Reference Signal), the network sends a corresponding number of downlink reference signals (CSI-RS) to the terminal. It is worth noting that to avoid measurement signaling storms, the network sends downlink reference signals to the terminal in batches. Similarly, the terminal also sends uplink reference signals to the network in batches.

[0102] 3. The terminal performs downlink channel measurement on the downlink reference signal sent by the network each time to obtain full channel state information of the downlink channel.

[0103] 4. The terminal obtains partial channel state information from the full channel state information as the initial channel state information based on the amount of CSI feedback data. For example, the terminal obtains a target channel state information at intervals of k (e.g., k=1) channel state information from the full channel state information, and finally aggregates all the obtained target channel state information to obtain initial channel state information. The terminal sends the initial channel state information to the network. Alternatively, the terminal may use other methods (e.g., averaging) to obtain partial channel state information from the full channel state information as the initial channel state information.

[0104] 5. After receiving the uplink reference signal sent by the terminal, the network performs uplink channel measurement on the uplink reference signal to obtain the uplink channel information. Because the terminal sends partial CSI, not full CSI, to the network, incremental processing is required for this partial CSI. In this case, the network uses the uplink channel information and the initial channel state information as inputs to a channel information estimation model, so that the channel information estimation model outputs a full channel estimation result.

[0105] 6. After obtaining the full channel estimation result, the network performs a sliding average calculation on the full channel estimation result according to the sliding distance and the sliding window length to obtain a feedback codebook.

[0106] 7. The network sends the feedback codebook to the terminal to complete synchronization of the feedback codebook.

[0107] Specifically, in order to reduce the amount of original channel state information data required by the network side to generate the codebook and improve the codebook generation speed, a sliding average calculation method is used to generate the codebook. Then the step S23 specifically includes:

[0108] S231, intercepting local channel state information that meets the sliding window length from the full channel estimation result according to the sliding distance;

[0109] S232. Calculate the average of the local channel state information to obtain a local codebook corresponding to the local channel state information.

[0110] S233. Aggregate the local codebooks of all local channel state information to obtain the feedback codebook of the original channel state information.

[0111] It should be noted that the specific working processes of steps S231 to S233 can refer to the processes of steps S121 to S123 in the above first channel state information feedback method, and will not be elaborated here.

[0112] See Figure 7 , Figure 7 is a schematic diagram of the dual - end synchronous generation of a codebook provided by an embodiment of the present invention. When the terminal measures the full amount of channel state information, but according to the CSI feedback data volume configured by the network, the terminal only feeds back part of the channel state information (i.e., the initial channel state information) to the network. The channel information estimation model trained on the network side uses the measured uplink channel information and the initial channel state information fed back by the terminal as model inputs, and the output data is the full - amount channel estimation result obtained after incrementing based on the uplink channel information. Then, based on this, a full - amount CSI feedback codebook is further generated and the full - amount CSI feedback codebook is sent to the terminal for synchronization.

[0113] Specifically, the channel information estimation model includes a Forgotten Gate, an Input Gate, and an Output Gate. Among them, after receiving the initial channel state information and the uplink channel information, the channel information estimation model determines the uplink channel information to be retained through the Forgotten Gate according to the initial channel state information, fuses the initial channel state information passing through the Input Gate with the uplink channel information to be retained, and outputs through the Output Gate to obtain the full - amount channel estimation result.

[0114] Exemplarily, see Figure 8 , Figure 8 is a working flowchart of the channel information estimation model provided by an embodiment of the present invention. The initial channel state information is respectively input into the Forgotten Gate, the Input Gate, and the Output Gate. The outputs of the Forgotten Gate and the Input Gate are added and averaged with the uplink channel information to determine the uplink channel information to be retained. The initial channel state information and the uplink channel information to be retained are used as the inputs of the Output Gate, and the Output Gate fuses these two data. The output of the Output Gate is the output of the channel information estimation model.

[0115] In the embodiment of the present invention, the channel information estimation model is used to increment the initial channel state information to restore the full - amount CSI for subsequent moving average calculation.

[0116] Compared with the prior art, the channel state information feedback method disclosed in the present invention optimizes the problem of large overhead during the generation and transmission of codebooks. The sliding average method is used to generate the feedback codebook. Compared with the existing codebook generation method, the network side uniformly configures the configuration parameters and the amount of channel state information test data required by the sliding average method, so that the number of measured channel state information samples required to generate the same number of codebooks is less, effectively reducing the overhead during the codebook generation and transmission. In addition, when the terminal feeds back CSI to the network, in order to reduce the uplink transmission overhead, it does not feed back the full amount of CSI, but feeds back part of the CSI, effectively reducing the CSI feedback overhead required during the generation of the CSI feedback codebook; at the same time, by incrementing part of the CSI with the measured uplink channel information, the full amount of CSI can be effectively restored.

[0117] See Figure 9 , Figure 9 FIG.

[0118] S31. Receive the initial channel state information of the downlink channel sent by the terminal;

[0119] S32. Input the initial channel state information and the uplink channel information into the channel information estimation model, so that the channel information estimation model outputs the full-channel estimation result;

[0120] S33. Perform a sliding average calculation on the full-channel estimation result according to the preset configuration parameters to obtain the feedback codebook of the full-channel estimation result.

[0121] Specifically, the terminal performs a sliding average calculation on the full-channel state information according to the preset configuration parameters and obtains the feedback codebook of the full-channel state information, and the initial channel state information is part of the channel state information of the full-channel state information.

[0122] Specifically, before performing step S31, the method further includes: sending a downlink reference signal to the terminal, so that the terminal obtains the full-channel state information of the downlink channel according to the downlink reference signal, and obtains part of the channel state information from the full-channel state information as the initial channel state information.

[0123] Exemplarily, see Figure 10 , Figure 10 FIG. Figure 10 The above steps S31 to S33 are described in detail as follows:

[0124] 1. The network configures the uplink channel data volume, downlink channel data volume, and CSI feedback data volume, and synchronizes the uplink channel data volume, downlink channel data volume, and CSI feedback data volume with the terminal. The uplink channel data volume is used to indicate that the terminal needs to send an uplink reference signal that meets this data volume, the downlink channel data volume is used to inform the terminal of the amount of CSI data that can be measured (i.e., the full amount of channel state information), and the CSI feedback data volume is used to indicate to the terminal the amount of CSI data that it needs to feedback (i.e., the initial channel state information). In addition, the configuration parameters are pre-stored in the network, and the configuration parameters include the sliding window length and the sliding distance. The network synchronously sends the configuration parameters to the terminal.

[0125] 2. The terminal sends an uplink reference signal (SRS) corresponding to the amount of uplink channel data to the network. The network then sends a corresponding number of downlink reference signals (CSI-RS) to the terminal. It is worth noting that to avoid measurement signaling storms, the network sends downlink reference signals to the terminal in batches. Similarly, the terminal also sends uplink reference signals to the network in batches.

[0126] 3. Based on the downlink channel data volume, the terminal can know whether the downlink reference signals sent by the network in batches have been sent. The terminal performs downlink channel measurement on the downlink reference signals sent by the network each time to obtain full channel state information of the downlink channel.

[0127] 4. The terminal obtains partial channel state information from the full channel state information as the initial channel state information based on the amount of CSI feedback data. For example, the terminal obtains a target channel state information at intervals of k (e.g., k=1) channel state information from the full channel state information, and finally aggregates all the obtained target channel state information to obtain initial channel state information. The terminal sends the initial channel state information to the network. Alternatively, the terminal may use other methods (e.g., averaging) to obtain partial channel state information from the full channel state information as the initial channel state information.

[0128] 5. The terminal performs a sliding average calculation on the original channel state information according to the sliding distance and the sliding window length to obtain a feedback codebook. This process can be referred to the process of steps S121 to S123 above and will not be repeated here.

[0129] 6. After the network receives the uplink reference signal sent by the terminal, it performs uplink channel measurement on the uplink reference signal to obtain the uplink channel information. Since the terminal does not send all CSI to the network, but only part of the CSI, incremental processing needs to be performed on this part of the CSI. At this time, the network uses the uplink channel information and the initial channel state information as the input of the channel information estimation model, so that the channel information estimation model outputs the full-channel estimation result.

[0130] 7. After the network obtains the full-channel estimation result, it performs a moving average calculation on the full-channel estimation result according to the sliding distance and the sliding window length to obtain a feedback codebook.

[0131] Specifically, to reduce the amount of original channel state information data required for the network to generate a codebook and improve the codebook generation speed, a moving average calculation method is used to generate the codebook. Then step S33 specifically includes:

[0132] S331. Intercept the local channel state information that meets the sliding window length in the full-channel estimation result according to the sliding distance;

[0133] S332. Perform an average calculation on the local channel state information to obtain a local codebook corresponding to the local channel state information;

[0134] S333. Aggregate the local codebooks of all local channel state information to obtain the feedback codebook of the original channel state information.

[0135] It should be noted that the specific working processes of steps S331 to S333 can refer to the processes of steps S121 to S123 in the above first channel state information feedback method, and will not be elaborated here.

[0136] See Figure 11 , Figure 11 is a schematic diagram of double-ended asynchronous codebook generation provided by an embodiment of the present invention. When the terminal measures the full-channel state information, but according to the CSI feedback data volume configured by the network, the terminal only feeds back part of the channel state information (i.e., the initial channel state information) to the network. The channel information estimation model trained on the network side uses the measured uplink channel information and the initial channel state information fed back by the terminal as the model input, and the output data is the full-channel estimation result obtained after incrementing based on the uplink channel information, and then further generates a full CSI feedback codebook corresponding to the full-channel estimation result. In addition, the terminal itself uses a moving average method to calculate the full CSI measured, and obtains the full-channel state information feedback codebook.

[0137] Specifically, the channel information estimation model includes a forget gate, an input gate, and an output gate. After receiving the initial channel state information and the uplink channel information, the channel information estimation model determines the uplink channel information to be retained through the forget gate based on the initial channel state information, fuses the initial channel state information passing through the input gate with the uplink channel information to be retained, and outputs the information through the output gate to obtain the full channel estimation result.

[0138] It is worth noting that the specific structure and working process of the channel information estimation model can refer to the channel information estimation model in the second channel state information feedback method mentioned above, and will not be repeated here.

[0139] Compared with the prior art, the channel state information feedback method disclosed in the present invention optimizes the problem of high overhead during codebook generation and transmission, and uses a sliding average method to generate a feedback codebook. Compared with the existing codebook generation method, the configuration parameters and the amount of channel state information test data required by the sliding average method are uniformly configured by the network side, so that the number of samples of measured channel state information required to generate the same number of codebooks is reduced, effectively reducing the overhead during codebook generation and transmission. In addition, when the terminal feeds back CSI to the network, in order to reduce the uplink transmission overhead, it does not feed back the full amount of CSI, but feeds back partial CSI, effectively reducing the CSI feedback overhead required in the CSI feedback codebook generation process; at the same time, by using the measured uplink channel information to increment the partial CSI, the full amount of CSI can be effectively restored. Moreover, from the perspective of optimizing the AI model, it is no longer necessary to force the synchronization of the codebooks between the network and the terminal. Instead, the ability of the AI model is used to enable the network to output a result that is as similar as possible to the terminal codebook, thereby further saving the network's overhead of sending the codebook to the terminal.

[0140] See also Figure 12 , Figure 12 1 is a structural block diagram of a first channel state information feedback apparatus 100 provided in an embodiment of the present invention. The channel state information feedback apparatus 100 includes:

[0141] Downlink channel measurement module 11 is used to perform downlink channel measurement on the downlink reference signal sent by the network to obtain the original channel state information of the downlink channel;

[0142] A feedback codebook generating module 12 is configured to perform a sliding average calculation on the original channel state information according to preset configuration parameters to obtain a feedback codebook of the original channel state information;

[0143] The feedback codebook synchronization module 13 is configured to send the feedback codebook to the network.

[0144] Specifically, the channel state information feedback device 100 further includes:

[0145] A configuration parameter receiving module, configured to receive configuration parameters sent by the network; wherein, the configuration parameters include a sliding window length and a sliding distance.

[0146] Specifically, the feedback codebook generation module 12 is specifically configured to: intercept local channel state information satisfying the sliding window length in the original channel state information according to the sliding distance; perform an average calculation on the local channel state information to obtain a local codebook corresponding to the local channel state information; and summarize the local codebooks of all local channel state information to obtain a feedback codebook of the original channel state information.

[0147] Specifically, the averaging of the local channel state information includes: performing an average calculation on the local channel state information in a parallel computing mode.

[0148] Specifically, the performing an average calculation on the local channel state information in a parallel computing mode includes: storing the repeated channel state information in two consecutive sliding windows into a shared cache; and when the average calculation of the previous local channel state information has not ended, the channel state information in the shared cache participates in the average calculation of the subsequent local channel state information.

[0149] It should be noted that the working processes of the various modules in the channel state information feedback device 100 according to the embodiments of the present invention may refer to the working process of the first channel state information feedback method described in the above embodiments, and will not be elaborated herein.

[0150] See Figure 13 , Figure 13 FIG. is a structural block diagram of a second channel state information feedback device 200 provided by an embodiment of the present invention. The channel state information feedback device 200 includes:

[0151] An original channel state information receiving module 21, configured to receive initial channel state information of a downlink channel sent by a terminal;

[0152] A full-channel estimation result generation module 22, configured to input the initial channel state information and uplink channel information into a channel information estimation model, so that the channel information estimation model outputs a full-channel estimation result.

[0153] Specifically, the channel state information feedback device 200 further includes:

[0154] A feedback codebook generation module 23, configured to perform a sliding average calculation on the full-channel estimation result according to preset configuration parameters to obtain a feedback codebook of the full-channel estimation result;

[0155] The feedback codebook synchronization module 24 is configured to send the feedback codebook to the terminal.

[0156] Specifically, the configuration parameters include sliding window length and sliding distance.

[0157] Specifically, the feedback codebook generation module 23 is specifically used to: intercept local channel state information that meets the sliding window length in the full channel estimation result according to the sliding distance; average the local channel state information to obtain a local codebook corresponding to the local channel state information; and summarize the local codebooks of all local channel state information to obtain the feedback codebook of the original channel state information.

[0158] Specifically, the channel state information feedback device 200 further includes:

[0159] The downlink reference signal sending module is used to send a downlink reference signal to the terminal, so that the terminal obtains the full channel state information of the downlink channel according to the downlink reference signal, and obtains partial channel state information from the full channel state information as the initial channel state information.

[0160] Specifically, the channel state information feedback device 200 further includes:

[0161] The uplink channel information acquisition module is configured to receive an uplink reference signal sent by a terminal; perform uplink channel measurement on the uplink reference signal to obtain the uplink channel information.

[0162] Specifically, the channel information estimation model includes a forget gate, an input gate, and an output gate; wherein, after receiving the initial channel state information and the uplink channel information, the channel information estimation model determines the uplink channel information to be retained through the forget gate based on the initial channel state information, fuses the initial channel state information passing through the input gate with the uplink channel information to be retained, and outputs the result through the output gate to obtain the full channel estimation result.

[0163] It is worth noting that the working process of each module in the channel state information feedback apparatus 200 described in the embodiment of the present invention can refer to the working process of the third channel state information feedback method described in the above embodiment, and will not be repeated here.

[0164] See also Figure 14 , Figure 14It is a structural block diagram of a channel state information feedback device 300 provided by an embodiment of the present invention. The channel state information feedback device 300 includes a processor 31, a memory 32, and a computer program stored in the memory 32 and executable on the processor 31. When the processor 31 executes the computer program, the steps in the above-described embodiments of various channel state information feedback methods are implemented, such as steps S11 - S13, S21 - S24, S31 - S33.

[0165] Exemplarily, the computer program can be divided into one or more modules / units, and the one or more modules / units are stored in the memory 32 and executed by the processor 31 to complete the present invention. The one or more modules / units can be a series of computer program instruction segments capable of performing specific functions, and these instruction segments are used to describe the execution process of the computer program in the channel state information feedback device 300.

[0166] The channel state information feedback device 300 may include, but is not limited to, a processor 31 and a memory 32. Those skilled in the art can understand that the schematic diagram is only an example of the channel state information feedback device 300, and does not constitute a limitation on the channel state information feedback device 300. It may include more or fewer components than shown in the figure, or combine certain components, or have different components. For example, the channel state information feedback device 300 may further include input / output devices, network access devices, buses, etc.

[0167] The processor 31 can be a central processing unit (CPU), or can also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The processor 31 is the control center of the channel state information feedback device 300, and connects various parts of the entire channel state information feedback device 300 through various interfaces and lines.

[0168] The memory 32 can be used to store the computer programs and / or modules. By running or executing the computer programs and / or modules stored in the memory 32 and calling the data stored in the memory 32, the processor 31 realizes various functions of the channel state information feedback device 300. The memory 32 mainly includes a program storage area and a data storage area. Among them, the program storage area can store an operating system, application programs required for at least one function (such as a sound playback function, an image playback function, etc.); the data storage area can store data created according to the use of the mobile phone (such as audio data, phone book, etc.). In addition, the memory 32 can include high-speed random access memory, and can also include non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, at least one magnetic disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0169] Among them, if the modules / units integrated in the channel state information feedback device 300 are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, to implement all or part of the processes in the above-described embodiment methods of the present invention, it can also be completed by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor 31, the steps of the above-described method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, an executable file, or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a mobile hard disk, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0170] The above is the preferred implementation manner of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements are also regarded as the protection scope of the present invention.

Claims

1. A method for channel state information feedback, characterized in that, Applied to a terminal, the method includes: Performing downlink channel measurement on the downlink reference signal sent by the network to obtain the original channel state information of the downlink channel; Performing a moving average calculation on the original channel state information according to preset configuration parameters to obtain the feedback codebook of the original channel state information; Sending the feedback codebook to the network.

2. The channel state information feedback method according to claim 1, characterized in that, Before performing downlink channel measurement on the reference signal sent by the network, the method further includes: Receiving the configuration parameters sent by the network; wherein the configuration parameters include the moving window length and the moving distance.

3. The channel state information feedback method according to claim 2, wherein The performing a moving average calculation on the original channel state information according to preset configuration parameters to obtain the feedback codebook of the original channel state information includes: Intercepting the local channel state information that meets the moving window length in the original channel state information according to the moving distance; Performing an average calculation on the local channel state information to obtain the local codebook corresponding to the local channel state information; Summarizing the local codebooks of all local channel state information to obtain the feedback codebook of the original channel state information.

4. The channel state information feedback method according to claim 3, wherein The averaging the local channel state information includes: Performing an average calculation on the local channel state information in a parallel computing mode.

5. The channel state information feedback method according to claim 4, wherein The performing an average calculation on the local channel state information in a parallel computing mode includes: Storing the repeated channel state information in two consecutive moving windows into a shared cache; When the average calculation of the previous local channel state information has not ended, the channel state information in the shared cache participates in the average calculation of the subsequent local channel state information.

6. A channel state information feedback method, characterized in that, Applied to a network, the method includes: Receiving the initial channel state information of the downlink channel sent by the terminal; Inputting the initial channel state information and the uplink channel information into a channel information estimation model so that the channel information estimation model outputs a full-channel estimation result.

7. The channel state information feedback method according to claim 6, wherein After obtaining the full-channel estimation result, the method further includes: Performing a moving average calculation on the full-channel estimation result according to preset configuration parameters to obtain the feedback codebook of the full-channel estimation result; Sending the feedback codebook to the terminal.

8. The channel state information feedback method according to claim 7, wherein The configuration parameters include the moving window length and the moving distance.

9. The channel state information feedback method according to claim 7, wherein The performing a moving average calculation on the full-channel estimation result according to preset configuration parameters to obtain the codebook data of the full-channel estimation result includes: Intercepting the local channel state information that meets the moving window length in the full-channel estimation result according to the moving distance; Performing an average calculation on the local channel state information to obtain the local codebook corresponding to the local channel state information; Summarizing the local codebooks of all local channel state information to obtain the feedback codebook of the original channel state information.

10. The channel state information feedback method according to claim 6, wherein Before receiving the initial channel state information of the downlink channel sent by the terminal, the method further includes: Sending a downlink reference signal to the terminal so that the terminal obtains the full-channel state information of the downlink channel according to the downlink reference signal and obtains partial channel state information from the full-channel state information as the initial channel state information.

11. The channel state information feedback method according to claim 7, characterized in that Before performing a moving average calculation on the full-channel estimation result according to preset configuration parameters, the method further includes: Receiving an uplink reference signal sent by a terminal; Performing uplink channel measurement on the uplink reference signal to obtain the uplink channel information.

12. The channel state information feedback method according to claim 6, characterized in that, The terminal performs a moving average calculation on the full-channel state information according to preset configuration parameters and obtains a feedback codebook of the full-channel state information.

13. The channel state information feedback method according to any one of claims 6 to 12, characterized in that The channel information estimation model includes a forget gate, an input gate, and an output gate; wherein, After receiving the initial channel state information and the uplink channel information, the channel information estimation model determines the uplink channel information to be retained through the forget gate according to the initial channel state information, fuses the initial channel state information passing through the input gate with the uplink channel information to be retained, and outputs through the output gate to obtain the full-channel estimation result.

14. A channel state information feedback device, characterized in that, Including: A downlink channel measurement module, configured to perform downlink channel measurement on a downlink reference signal sent by a network to obtain the original channel state information of the downlink channel; A feedback codebook generation module, configured to perform a moving average calculation on the original channel state information according to preset configuration parameters to obtain a feedback codebook of the original channel state information; A feedback codebook synchronization module, configured to send the feedback codebook to the network.

15. A channel state information feedback device, characterized in that, Including: An original channel state information receiving module, configured to receive the initial channel state information of the downlink channel sent by a terminal; A full-channel estimation result generation module, configured to input the initial channel state information and the uplink channel information into a channel information estimation model, so that the channel information estimation model outputs a full-channel estimation result.

16. A channel state information feedback device, characterized in that, Including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, the method for feedback of channel state information according to any one of claims 1 to 13 is implemented.

17. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored computer program, wherein when the computer program runs, it controls the device where the computer-readable storage medium is located to execute the method for feedback of channel state information according to any one of claims 1 to 13.

18. A computer program product, comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, the method for feedback of channel state information according to any one of claims 1 to 13 is implemented.

Images