Channel state information feedback method, device, and computer readable medium

Abstract

The present disclosure provides a channel state information (CSI) feedback method. In a beam management procedure, a base station device sends, to a user equipment, resource configuration information comprising mapping relationships between a CSI resource set for beam management and a CSI resource subset for beam management and a transmitting antenna group, and reporting configuration information comprising target transmitting antenna group information, and sends, to the user equipment, an RS for beam management, so as to allow for multiplexing of RS and CSI resources for beam management; and the user equipment determines first channel information of a target CSI resource subset on the basis of the RS for beam management, determines second channel information on the basis of the first channel information, and then performs CSI feedback on the basis of the second information. The present disclosure further provides a base station device, a user equipment, and a computer readable medium.

Description

Channel state information feedback method, device and computer-readable medium

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims priority to Chinese patent application No. 202311864960.8 filed with the China Patent Office on December 29, 2023, the entire contents of which are incorporated herein by reference. Technical Field

[0003] The present disclosure relates to, but is not limited to, the field of communication technology. Background Art

[0004] In 5G NR (New Radio), channel state feedback (CSF) is a crucial communication process that allows user equipment (UE) to provide information about radio channel quality and characteristics to the gNB (gNB). This feedback is critical for scheduling, beam management, and other radio resource allocation decisions. During the CSI-RS (Channel State Information-Reference Signal) feedback process, receiving CSI-RS by the UE results in system resource overhead. For example, the transmission of periodic CSI-RS can lead to significant periodic resource usage.

[0005] Related technologies have proposed event-based beam quality reporting methods, such as aperiodic beam quality reporting, to achieve more flexible CSI-RS feedback or scheduling control while avoiding the large resource overhead caused by periodic transmission. However, aperiodic CSI-RS requires more complex signal design and scheduling algorithms to determine the transmission time and location, which increases system complexity and algorithm overhead. In addition, resource overhead is still required for CSI-RS feedback during aperiodic moments or events. Summary of the Invention

[0006] The present disclosure provides a channel state information feedback method, device, and computer-readable medium.

[0007] In a first aspect, an embodiment of the present disclosure provides a channel state information feedback method, which is applied to a base station device, wherein the transmitting antenna of the base station device includes at least one transmitting antenna group, and the method includes: in a beam management process, sending resource configuration information for channel state information (CSI) feedback to a user equipment, the resource configuration information including a CSI resource set for beam management and a pre-set mapping relationship between the CSI resource subset for beam management and the transmitting antenna group, the CSI resource set for beam management including at least one CSI resource subset; sending reporting configuration information for CSI feedback to the user equipment, the reporting configuration information including target transmitting antenna group information; sending a reference signal (RS) for beam management to the user equipment; wherein the resource configuration information and the reporting configuration information are used to instruct the user equipment to determine first channel information of a target CSI resource subset based on the RS for beam management, and determine second channel information for CSI feedback based on the first channel information, and the target CSI resource subset is the CSI resource subset corresponding to the target transmitting antenna group information.

[0008] On the other hand, an embodiment of the present disclosure provides a channel state information feedback method, which is applied to a user equipment, and the method includes: receiving resource configuration information for channel state information CSI feedback and reporting configuration information for CSI feedback sent by a base station device; wherein the resource configuration information includes a CSI resource set for beam management and a mapping relationship between a pre-set CSI resource subset and a transmitting antenna group, and the CSI resource set for beam management includes at least one CSI resource subset; the reporting configuration information includes target transmitting antenna group information; determining a target CSI resource subset corresponding to the target transmitting antenna group information according to the mapping relationship, and receiving a reference signal RS for beam management sent by the base station device on the CSI resource of the target CSI resource subset; determining first channel information of the target CSI resource subset according to the RS; determining second channel information according to the first channel information, and the second channel information is full channel information for CSI feedback.

[0009] On the other hand, an embodiment of the present disclosure also provides a base station device, comprising: one or more processors; a storage device on which one or more programs are stored; when the one or more programs are executed by the one or more processors, the one or more processors implement any channel state information feedback method described herein; one or more I / O interfaces, connected between the processor and the storage device, configured to implement information interaction between the processor and the storage device.

[0010] On the other hand, an embodiment of the present disclosure also provides a user equipment, comprising: one or more processors; a storage device on which one or more programs are stored; when the one or more programs are executed by the one or more processors, the one or more processors implement any channel state information feedback method described herein; one or more I / O interfaces, connected between the processor and the storage device, configured to implement information interaction between the processor and the storage device.

[0011] On the other hand, an embodiment of the present disclosure further provides a computer-readable medium having a computer program stored thereon, wherein when the computer program is executed by a processor, any channel state information feedback method described herein is implemented. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] FIG1 is a schematic diagram of a channel state information feedback process with a base station device as the execution subject provided by an embodiment of the present disclosure;

[0013] FIG2 is a schematic diagram of the overall process of channel state information feedback provided by an embodiment of the present disclosure;

[0014] FIG3 is a schematic diagram of establishing a mapping relationship between a CSI resource subset for beam management and a transmit antenna group according to an embodiment of the present disclosure;

[0015] FIG4 is a schematic diagram of a channel state information feedback process with a user equipment as the execution subject provided by an embodiment of the present disclosure;

[0016] FIG5 is a schematic diagram of a full channel state information recovery process according to an embodiment of the present disclosure;

[0017] FIG6 is a schematic diagram of the working process of the classification model provided by an embodiment of the present disclosure;

[0018] FIG7 is a schematic diagram of the structure of a classification model provided by an embodiment of the present disclosure;

[0019] FIG8 is a schematic diagram of the classification model training and application process provided by an embodiment of the present disclosure;

[0020] FIG9a is a schematic diagram of channel recovery provided by the first exemplary embodiment of the present disclosure;

[0021] FIG9 b is a schematic diagram of channel recovery provided by the second exemplary embodiment of the present disclosure;

[0022] FIG9c is a schematic diagram of channel recovery provided by exemplary embodiment 3 of the present disclosure;

[0023] FIG10 is a schematic structural diagram of a base station device and a user equipment provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0024] Example embodiments will be described more fully hereinafter with reference to the accompanying drawings, but the example embodiments may be embodied in different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete and will fully convey the scope of this disclosure to those skilled in the art.

[0025] As used herein, the term "and / or" includes any and all combinations of one or more of the associated listed items.

[0026] The terms used herein are used only to describe specific embodiments and are not intended to limit the present disclosure. As used herein, the singular forms "a," "an," and "the" are also intended to include the plural forms, unless the context clearly indicates otherwise. It will also be understood that when the terms "comprising" and / or "made of" are used in this specification, the presence of the features, wholes, steps, operations, elements, and / or components is specified, but the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or groups thereof is not excluded.

[0027] The embodiments described herein may be described with reference to plan views and / or cross-sectional views, with the aid of idealized schematic diagrams of the present disclosure. Thus, the example illustrations may be modified based on manufacturing techniques and / or tolerances. Therefore, the embodiments are not limited to the embodiments shown in the accompanying drawings, but include modifications of the configurations formed based on the manufacturing process. Therefore, the regions illustrated in the accompanying drawings are schematic in nature, and the shapes of the regions shown in the drawings illustrate specific shapes of the regions of the elements, but are not intended to be limiting.

[0028] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art. It will also be understood that terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant art and the present disclosure, and will not be interpreted as having an idealized or overly formal meaning unless expressly defined as such herein.

[0029] CSI feedback is a channel negotiation method between the UE and the gNB, playing a crucial role in the communication system. The CSI feedback process between the UE and the gNB is as follows: The gNB configures CSI feedback parameters, indicating which information (CQI, PMI, RI, etc.) the UE requires feedback on. Based on this configuration, the gNB sends a reference signal (RS) for CSI measurement, such as a synchronization signal block (SSB) or a channel state information-reference signal (CSI-RS), to the UE. The UE measures the RS, or CSI measurement, to obtain the RS containing channel information. Based on this information, the UE performs channel estimation, CSI calculation, CSI scheduling, and ultimately reports the CSI to the gNB. After receiving the CSI reporting information, the gNB performs adaptive scheduling for each user, including beam, precoding, and MCS (Modulation and Coding Scheme) scheduling, thereby completing the entire CSI feedback process.

[0030] When performing the CSI measurement process for CSI feedback, a large amount of resources will be occupied. In communication systems, a process very similar to the CSI measurement process is the reference signal measurement in the beam management process, but the two send different signals. The reference signal used for beam management is used in complex multi-antenna systems and needs to go through a beamforming process; while the reference signal used for channel state feedback is usually sent and collected in a non-beamforming manner to better restore the channel state information. In traditional methods, the reference signal measurement for channel state feedback can be used to improve beam management performance, but it is difficult to use the reference signal used for beam management to extract channel state information and thus improve channel state feedback performance. The two similar measurement processes are independent of each other, resulting in a waste of transmission resources.

[0031] To at least address the aforementioned issues, embodiments of the present disclosure provide a channel state information feedback method. The method is applied to a base station device, wherein the base station device's transmit antennas include at least one transmit antenna group. Figure 1 illustrates a flow chart of channel state information feedback performed by a base station device according to an embodiment of the present disclosure, and Figure 2 illustrates an overall flow chart of channel state information feedback according to an embodiment of the present disclosure. As shown in Figures 1 and 2, the method may include the following steps, S11 through S13.

[0032] In step S11, in the beam management process, resource configuration information for CSI feedback is sent to the user equipment, where the resource configuration information includes a CSI resource set for beam management and a mapping relationship between a pre-set CSI resource subset for beam management and a transmitting antenna group, where the CSI resource set for beam management includes at least one CSI resource subset.

[0033] During the initialization phase, the gNB divides the CSI resources used for beam management into multiple CSI resource subsets, each of which has the same transmit antenna physical port. That is, the gNB divides the CSI resource subsets based on the transmit antenna physical port, with the same CSI resource subset corresponding to the same multiple transmit antenna physical ports. The gNB establishes transmit antenna groups containing the transmit antenna physical port numbers and establishes a mapping between each CSI resource subset used for beam management and each transmit antenna group, indicating the transmit antenna physical port to which the CSI resource subset corresponds. During the beam management process, the gNB sends resource configuration information for CSI feedback to the UE, including the CSI resource sets used for beam management and the mapping between the CSI resource subsets used for beam management and the transmit antenna groups.

[0034] In step S12, reporting configuration information for CSI feedback is sent to the user equipment, where the reporting configuration information includes target transmit antenna group information.

[0035] To complete the CSI feedback configuration, the gNB also needs to configure the reporting configuration information required for CSI feedback to the UE. The reporting configuration information includes the target transmit antenna group information, which can be the identifier of the transmit antenna group corresponding to the CSI feedback. It should be noted that the reporting configuration information can also include the number of physical ports N of the target transmit antenna group. tx .

[0036] In step S13, an RS for beam management is sent to the user equipment.

[0037] The resource configuration information and reporting configuration information are used to instruct the user equipment to determine the first channel information of the target CSI resource subset based on the RS used for beam management, and to determine the second channel information for CSI feedback based on the first channel information. The target CSI resource subset is the CSI resource subset corresponding to the target transmitting antenna group information.

[0038] After receiving the resource configuration information and reporting configuration information for CSI feedback and the RS for beam management, the UE performs channel estimation on the CSI resource subset used for CSI feedback, extracts the first channel information, and performs channel recovery on the first channel information to obtain the full channel information, i.e., the second channel information. Subsequently, CSI calculation and CSI reporting are performed based on the second channel information to realize CSI feedback.

[0039] It should be noted that, as shown in Figure 2, since this application is a CSI feedback process multiplexing beam management process, the gNB also needs to send reporting configuration information for beam management to the UE so that the UE can implement normal beam management according to the reporting configuration information for beam management.

[0040] In this disclosed embodiment, the base station's CSI feedback reuses the CSI resources and reference signals of the beam management process. CSI resource sets are grouped according to the physical ports of the transmit antennas. The reporting configuration of the CSI feedback process uses the transmit antenna group index instead of the resource index indicator in the traditional CSI reporting configuration. The user equipment obtains full channel state information through channel recovery and performs CSI feedback processing based on this full channel state information.

[0041] The channel state information feedback method provided by the embodiment of the present disclosure is as follows: the base station device sends resource configuration information including a CSI resource set for beam management and a mapping relationship between the CSI resource subset for beam management and the transmitting antenna group, as well as reporting configuration information including target transmitting antenna group information to the user equipment in the beam management process, and sends the RS for beam management to the user equipment, thereby multiplexing the RS and CSI resources for beam management; the user equipment determines the first channel information of the target CSI resource subset based on the RS for beam management, and determines the second channel information based on the first channel information, thereby performing CSI feedback based on the second information; the embodiment of the present disclosure configures resources and reports resources for CSI feedback by grouping transmitting antennas, and introduces channel recovery technology, so that CSI feedback can recover the full channel state information by borrowing the RS for beam management, thereby reducing system resources and improving system transmission efficiency.

[0042] In some embodiments, the step of establishing a mapping relationship between CSI resource subsets for beam management and transmit antenna groups includes: dividing the CSI resource set for beam management into at least one CSI resource subset according to the physical port of the transmit antenna, the physical port of the transmit antenna of each CSI resource in the same CSI resource subset is the same, and a transmit antenna group includes multiple physical ports of the transmit antenna; dividing transmit antennas with the same physical port into a transmit antenna group, and configuring group information for each transmit antenna group; establishing a mapping relationship between each CSI resource subset for beam management and each transmit antenna group, and the CSI resource subset information of each CSI resource subset corresponds one-to-one to the transmit antenna group information of each transmit antenna group.

[0043] FIG3 is a schematic diagram of establishing a mapping relationship between a CSI resource subset for beam management and a transmit antenna group according to an embodiment of the present disclosure. The process of establishing the mapping relationship is described below in conjunction with FIG3. As shown in FIG3, the transmit antenna of the gNB has 32 real physical ports, namely N tx =32, the UE's receiving antenna has 4 physical ports, that is, it has 4 receiving antennas, N rx = 4. The CSI resource set for beam management includes 64 CSI resources, namely N csi =64.

[0044] During beam management, gNB uses N p = 2 logical transmit ports send RS for beam management, UE uses N rx = 4 physical port receive antennas receive the RS. To multiplex CSI resources for beam management and RS for CSI feedback, the gNB divides the CSI resources using the same transmit antenna physical port into subsets and groups the transmit antennas by physical port. Each CSI resource subset corresponds to a transmit antenna group. Taking the embodiment shown in Figure 3 as an example, group 0 is set to correspond to transmit antenna physical port numbers 01-16, group 1 to transmit antenna physical port numbers 17-32, and group 2 to transmit antenna physical port numbers 01-32 (i.e., all transmit antennas). Thus, based on the CSI resources using the same transmit antenna physical port within the transmit antenna group, at least three CSI resource subsets are obtained, namely:

[0045] 1. CSI resource subset 1, corresponding to group 0, which contains N csi1 = 16 CSI resources. RSs received on these 16 CSI resources are transmitted using physical ports numbered 01-16 of the actual physical transmit antennas.

[0046] 2. CSI resource subset 2, corresponding to group 1, which contains N csi2= 18 CSI resources. RSs received on these 18 CSI resources are transmitted using physical ports numbered 17-32 of the actual physical transmit antennas.

[0047] 3. CSI resource subset 3, corresponding to group 2, which contains N csi3 = 24 CSI resources, and the RSs received on these 24 CSI resources are all transmitted using the physical ports numbered 01-32 of the real physical transmit antennas.

[0048] The gNB establishes a one-to-one correspondence between each CSI resource subset and the transmit antenna group, thereby obtaining the mapping relationship.

[0049] After partitioning the CSI resources for beam management into subsets, grouping the transmit antennas, and establishing a mapping between the two, the gNB sends the following information to the UE in order to obtain the complete CSI resource configuration and reporting configuration required for CSI feedback:

[0050] 1. The gNB configures the UE with the CSI resource sets for beam management, i.e., 64 CSI resources, as well as information about all resource subsets and their corresponding transmit antenna groups, i.e., CSI resource subsets 1-3 and their corresponding groups 0-2.

[0051] 2. The gNB configures the UE with the group number / number set required for CSI feedback, as well as the number of physical ports N of the transmit antenna corresponding to the group number / number set. tx ;

[0052] 3. The gNB sends RS for beam management to the UE.

[0053] After the UE receives all the above content and configuration, it can reuse the RS and resources for beam management and enter the CSI feedback process.

[0054] The principle of the embodiment of the present disclosure is described below with reference to an example.

[0055] The gNB configures the transmit antenna group required for CSI feedback to the UE as group 0. That is, the CSI resources used for CSI feedback are CSI resource subset 1 as shown in Figure 3. The corresponding transmit antenna physical ports are 01-16. At this time, the number of CSI resources in CSI resource subset 1 is N. csi1 =16. Since the beam management process requires beamforming, it is assumed that the full channel matrix required for CSI feedback is H real (ie, the second channel matrix), with a size of N rx ×N tx , N rx is the number of receiving antennas of the user equipment, Nrx =4, N tx The number of physical ports in group 0, N tx =16.

[0056] The beamforming process is equivalent to H real Based on this, the i-th CSI resource is multiplied by the beamforming matrix W i , beamforming matrix W i The size is N tx ×N p , we get the channel matrix H solved by beam-managed RS through channel estimation i , channel matrix H i The size is N rx ×N p , i is the index of the CSI resource in the CSI resource subset, (i=1,2,3,…,N csi ), in this embodiment, the UE obtains 16 H i , need to be restored from them to get H real Taking this embodiment as an example, the corresponding channel recovery process is to solve the following equations:

[0057] The above equations are a matrix equations. It is known that the 16 H on the right side of the equal sign are i , what needs to be solved is H on the left side of the equal sign real Considering that only the variables on the right side of the equation are known, the beamforming matrix Wi is constructed based on the DFT (Discrete Fourier Transform) basis, and the number of physical ports N for different transmitting antennas is tx , W i Each column selects values ​​from the same set of fixed phases. Taking this embodiment as an example, N tx =16, set N1=4, N2=2, the number of phases selectable at each transmitter is N θ = N1 × N2 × O1 × O2 = 4 × 2 × 4 × 4 = 128, N1 and N2 are single-sided antenna elements, arranged in the form of (N1, N2) = (2, 2), O1 and O2 are oversampling factors. i Each column of 16 elements is represented as Under this restriction, the solution process of the matrix equations can be split into the solution of each logical port of the transmitting antenna, and the left side of the equation H real The solution process of W i The solution process of , and then a classifier can be used to solve the system of equations, greatly improving the solvability of the system of equations.

[0058] In some embodiments, in order to ensure the convergence of the channel information calculation result of the user equipment, the target transmit antenna group information meets the following conditions: the number of CSI resources N of the target CSI resource subset corresponding to the target transmit antenna group information csi The number of logical ports N of the transmitting antenna p The product of is greater than or equal to the number of physical ports N of the target transmit antenna group tx , that is, N csi ×N p ≥N tx The target transmit antenna group is the transmit antenna group corresponding to the target transmit antenna group information, and the logical port of the transmit antenna is the port for sending the RS for beam management.

[0059] The embodiment of the present disclosure further provides a channel state information feedback method, which is applied to a user equipment. As shown in FIG4 , the method may include the following steps S21 to S24 .

[0060] In step S21, resource configuration information for channel state information (CSI) feedback and reporting configuration information for CSI feedback sent by the base station device are received; wherein the resource configuration information includes a CSI resource set for beam management and a mapping relationship between a pre-set CSI resource subset for beam management and a transmitting antenna group, and the CSI resource set for beam management includes at least one CSI resource subset; and the reporting configuration information includes target transmitting antenna group information.

[0061] During the initialization phase, the gNB divides the CSI resources used for beam management into multiple CSI resource subsets with the same transmit antenna physical port, that is, the CSI resource subsets are divided according to the physical port of the transmit antenna, and the same CSI resource subset corresponds to the same physical ports of multiple transmit antennas; one or more transmit antenna groups with the same transmit antenna physical port sequence number are established, and a mapping relationship between each CSI resource subset and each transmit antenna group is established to indicate the physical port of the transmit antenna corresponding to the CSI resource subset. In the beam management process, the gNB sends to the UE the resource configuration information for CSI feedback, including the CSI resource set for beam management and the mapping relationship between the CSI resource subset for beam management and the transmit antenna group, as well as the reporting configuration information for CSI feedback including the target transmit antenna group information. The target transmit antenna group information can be the identifier of the transmit antenna group corresponding to the CSI feedback. It should be noted that the reporting configuration information can also include the number of physical ports N of the target transmit antenna group. tx .

[0062] In step S22, the target CSI resource subset corresponding to the target transmit antenna group information is determined according to the mapping relationship, and a reference signal RS for beam management sent by the base station device is received on the CSI resources of the target CSI resource subset.

[0063] The UE queries the mapping relationship between the CSI resource subset for beam management and the transmit antenna group based on the target transmit antenna group information, determines the corresponding target CSI resource subset, and receives the reference signal RS for beam management on the CSI resources of the target CSI resource subset.

[0064] In step S23, first channel information of the target CSI resource subset is determined according to the RS.

[0065] The UE performs channel estimation based on the RS and calculates the first channel information, that is, obtains the channel matrix H corresponding to each CSI resource in the target CSI resource subset. i .

[0066] In step S24, second channel information is determined according to the first channel information. The second channel information is full channel information and is used for CSI feedback.

[0067] The UE calculates the channel matrix H corresponding to each CSI resource in the target CSI resource subset. i Recover the full channel matrix H real , so that according to the full channel matrix H real Provide CSI feedback.

[0068] The channel state information feedback method provided by the embodiment of the present disclosure is as follows: the base station device sends resource configuration information including a CSI resource set for beam management and a mapping relationship between the CSI resource subset for beam management and the transmitting antenna group, as well as reporting configuration information including target transmitting antenna group information to the user equipment in the beam management process, and sends the RS for beam management to the user equipment, thereby multiplexing the RS and CSI resources for beam management; the user equipment determines the first channel information of the target CSI resource subset based on the RS for beam management, and determines the second channel information based on the first channel information, thereby performing CSI feedback based on the second information; the embodiment of the present disclosure configures resources and reports resources for CSI feedback by grouping transmitting antennas, and introduces channel recovery technology, so that CSI feedback can recover the full channel state information by borrowing the RS for beam management, thereby reducing system resources and improving system transmission efficiency.

[0069] In some embodiments, there are multiple target transmit antenna group information, that is, CSI feedback is required for multiple CSI resource subsets. In this case, channel estimation is performed for each CSI resource subset respectively, and channel recovery is performed to obtain the full channel matrix of each CSI resource subset.

[0070] Accordingly, the determining of the first channel information of the target CSI resource subset according to the RS (i.e., step S23) includes the following steps: determining the target CSI resource subset corresponding to each target transmit antenna group information; and determining, for each target transmit antenna group information corresponding to the target CSI resource subset, determining, according to the RS, the first channel information of the target CSI resource subset corresponding to the target transmit antenna group information.

[0071] Accordingly, the determining of the second channel information based on the first channel information (ie, step S24) includes the following steps: for the first channel information of the target CSI resource subset corresponding to each target transmitting antenna group information, determining the second channel information corresponding to the first channel information.

[0072] In some embodiments, as shown in FIG5 , the step of determining the second channel information according to the first channel information (ie, step S24 ) includes the following steps S241 and S242 .

[0073] In step S241, a beamforming matrix is ​​determined according to the first channel information and a pre-trained classification model, where the classification model is a neural network model.

[0074] In some embodiments, the first channel information is a first channel matrix, which is determined based on the intermediate channel matrix of each logical port of the transmitting antenna of the base station device, and the beamforming matrix is ​​determined based on the intermediate beamforming matrix of each logical port of the transmitting antenna of the base station device.

[0075] FIG6 is a schematic diagram of the classification model working process provided by an embodiment of the present disclosure. As shown in FIG6, the beamforming matrix is ​​determined based on the first channel information and the pre-trained classification model (i.e., step S241), including the following steps: inputting the first channel information into the classification model (i.e., AI classifier) ​​to obtain each phase index; determining the value of each element in the beamforming matrix based on each phase index. In which, the first channel information of each subcarrier in the entire bandwidth is input into the classification model, that is, the N subcarriers in the entire bandwidth are input into the classification model. c The channel matrix H i Input AI classifier, AI classifier outputs beamforming matrix W i The phase index θ of each element in is multiplied by e j2π / 128 , thus we get the beamforming matrix W i The values ​​of each element of .

[0076] Taking group 0 as an example, Ntx =16, set N1=4, N2=2, the number of optional phases of each transmitting antenna physical port is N θ = N1 × N2 × O1 × O2 = 4 × 2 × 4 × 4 = 128. Therefore, the beamforming matrix W i Each column of 16 elements can be expressed as Using AI classifier to solve the beamforming matrix W i The solvability of the channel recovery equation group can be greatly improved.

[0077] FIG7 is a schematic diagram of the structure of the classification model provided by the embodiment of the present disclosure. As shown in FIG7 , the classification model is a convolutional neural network model, including a convolutional layer and three fully connected layers, namely fully connected layers 1-3. The input of the classification model is N in full bandwidth. c The channel matrix H i , the output value of the classification model is a decimal, and N θ The phase index θ of each element in the beamforming matrix is ​​obtained by cyclic rounding.

[0078] As shown in Figure 8, the training and inference process of the classification model is as follows:

[0079] 1. Build a platform for training and testing data sets, and establish a beamforming matrix W i and the channel matrix H i dataset.

[0080] Channel matrix H i It can be collected through system platform simulation or system actual scene collection, that is, obtained from the beam management process under the conventional 5G system; the beamforming matrix W i It can be collected through system platform simulation or actual system scenario collection, that is, collected by beam scanning under conventional 5G system.

[0081] 2. According to the beamforming matrix W i and the channel matrix H i size, and build a neural network model.

[0082] 3. Create a loss function. You can use the cross entropy function commonly used in AI classifiers; create an optimization function (such as the Adam function).

[0083] 4. Perform model training, including network forward propagation, loss calculation, loss backpropagation, clearing the optimizer gradient cache, and updating network parameters using the optimizer. After multiple training cycles, the model is saved after convergence.

[0084] 5. Perform model testing / inference. This step differs from the training process in that it does not require loss feedback or optimizer-related operations. The model parameters no longer need to be further updated, and the network results can be directly output to calculate the loss value.

[0085] 6. Use the trained model for model inference and use AI to achieve channel recovery in the CSI feedback process.

[0086] Step S242: Determine second channel information according to the beamforming matrix and the first channel information.

[0087] The second channel information H can be calculated according to the following formula real,k : Where k = 1, 2, ..., N c , is the first channel information, is the beamforming matrix.

[0088] As shown in FIG6 and FIG7, the size of the first channel matrix is ​​N rx ×(N p ×N csi ), N rx is the number of receiving antennas of the user equipment, N p N is the number of logical ports of the base station equipment's transmitting antenna. csi is the number of CSI resources in the target CSI resource subset.

[0089] In some embodiments, with broadband granularity, the number of phase indices is (N p ×N csi ), N p N is the number of logical ports of the base station equipment's transmitting antenna. csi is the number of CSI resources in the target CSI resource subset. That is, in the case of broadband granularity, the number of elements in the beamforming matrix is ​​(N p ×N csi In the case of narrowband granularity, the number of phase indices is (N c ×N p ×N csi ), N c is the total number of subcarriers in the entire bandwidth. That is, in the case of narrowband granularity, the number of elements in the beamforming matrix is ​​(N c ×N p ×N csi ).

[0090] To clearly illustrate the solution of the embodiment of the present disclosure, the following describes in detail the process of recovering omnidirectional antenna channel state information using a classification model, taking the three transmitting antenna groups shown in FIG3 as an example in conjunction with FIG9a-FIG9c.

[0091] As shown in Figure 9a, the target transmit antenna group required for the gNB to configure CSI feedback to the UE is group 0, that is, the target CSI resource subset is CSI resource subset 1, the corresponding physical ports of the transmit antenna are 01-16, and the number of CSI resources in CSI resource subset 1 is N. csi1 =16.

[0092] In the case of broadband beamforming, the beamforming of each subcarrier within the bandwidth transmitted by RS is consistent. Assume that there are N subcarriers within the system bandwidth. c subcarriers, we can get the following including N c A system of equations:

[0093] Where k represents the subcarrier index (k = 1, 2, ..., N c ), the beamforming matrices in the equation group are the same, that is, W1 16*2 =W2 16*2 =…=W 16 16*2 Therefore, when beamforming is performed with broadband granularity, a converged training model can be obtained more quickly during classification model training, thereby improving model training accuracy.

[0094] The embodiment of the present disclosure introduces a classification model to recover the beamforming matrix, and then finally solves H by solving the equation real To better understand the functionality implemented by the classification model, we first merge the equations under a logical port p1 of the transmitting antenna. The merged equations are as follows:

[0095] Consider a logical port and Become a one-dimensional vector, further merge it, let

[0096] Therefore, for the logic port p1, the following equation is obtained:

[0097] Taking this embodiment as an example, there are N p = 2 logical transmit ports, so the equations are written as:

[0098] The equation further adds N p To expand,

[0099] The following equation can be obtained:

[0100] Considering N within the RS receiving system bandwidth c subcarriers, a total of N c Group (k=1,2,…,N c ) as a set of inputs to the classification model, As the output of the classification model, the matrix elements are in the form of phase index θ, that is, the classification model can obtain 32 phase indexes θ, and the value range of the phase index is {0,1,2,…,N θ -1=127}. After the classification model, the full-bandwidth beamforming matrix on the left side of the equation is obtained After that, since the right side of the equation is a known quantity, by solving the equation, the full channel information on each subcarrier can be recovered, that is, H real,k 4*16 .

[0101] The solution of the embodiment of the present disclosure is also applicable to beamforming based on narrowband granularity. In the case of narrowband granularity, the beamforming on each subcarrier within the bandwidth transmitted by the RS is different. Still taking the system shown in Figure 9a as an example, the gNB configures the target transmit antenna group required for CSI feedback to the UE as group 0, that is, the target CSI resource subset is CSI resource subset 1, the corresponding physical ports of the transmit antenna are 01-16, and the number of CSI resources in CSI resource subset 1 is N. csi1 =16. Assume that there are N c subcarriers, we can get the following including N c A system of equations:

[0102] Introducing a classification model to help recover W i,k , and then solve the equation to get H real To better understand the functionality of the classification model, we first merge the equations under a logical port p1. The merged form is as follows:

[0103] Consider a port and Become a one-dimensional vector, further merge it, let

[0104] Therefore, for the logic port p1, the following equation is obtained:

[0105] Taking this embodiment as an example, there are N p= 2 logical transmit ports, so the equations are written as:

[0106] The equation further adds N p To expand,

[0107] We obtain an equation of the following form:

[0108] Considering N within the RS receiving system bandwidth c subcarriers, a total of N c Group (k=1,2,…,N c ) as a set of inputs for the classification model, a total of N c Group As the output of the classification model, the matrix elements are in the form of phase index θ, that is, the network output N c ×32 phase indexes θ, the value range of the index on each subcarrier is {0,1,2,…,N θ -1=127}. After the classification model, the full-bandwidth beamforming matrix on the left side of the equation is obtained After that, since the right side of the equation is a known quantity, by solving the equation, the full channel information on each subcarrier can be recovered, that is, H real,k 4*16 .

[0109] The following uses the system shown in Figure 9b as an example to illustrate the process of recovering omnidirectional antenna channel state information using the classification model. As shown in Figure 9b, the gNB configures the target transmit antenna group required for CSI feedback to the UE as group 1, that is, the target CSI resource subset is CSI resource subset 2, and the corresponding physical ports of the transmit antenna are 17-32. The number of CSI resources in CSI resource subset 2 is N. csi1 =18.

[0110] Assuming that the beamforming in this embodiment is broadband granularity, the omnidirectional antenna channel state information recovery process in this embodiment is basically the same as that in the embodiment shown in FIG9a, and the number of physical ports of the transmitting antenna is N. tx =16, the difference is that the number of CSI resources in the CSI resource subset is different. In this embodiment, the UE obtains 18 H i ,i=1,2,3,…,18, we need to recover H from them real , that is, the final solution equation is as follows:

[0111] Considering N within the RS receiving system bandwidth c subcarriers, a total of N c Group (k=1,2,…,N c ) as a set of inputs to the classification model, As the output of the classification model, the matrix elements are in the form of phase index θ, that is, the classification model can obtain 36 phase indexes θ. tx N of the embodiment shown in FIG9a tx The same, so the value range of the phase index θ in this embodiment is also {0, 1, 2, ..., 127}. The difference between this embodiment and the embodiment shown in FIG9a is that the dimensions of the input and output matrices are different, so the size of the classification model used will be different.

[0112] The following uses the system shown in Figure 9c as an example to illustrate the process of recovering omnidirectional antenna channel state information using the classification model. As shown in Figure 9c, the gNB configures the target transmit antenna group required for CSI feedback to the UE as group 2, that is, the target CSI resource subset is CSI resource subset 3, the corresponding physical ports of the transmit antenna are 01-32, and the number of CSI resources in CSI resource subset 3 is N. csi1 =24.

[0113] Assuming that the beamforming in this embodiment is broadband granularity, the channel state information recovery process of the omnidirectional antenna in this embodiment is basically the same as that in the embodiment shown in FIG9a. The difference lies in the number of physical ports of the transmitting antenna. In this embodiment, the number of physical ports of the transmitting antenna is N. tx =32, in the embodiment shown in FIG9a, the number of physical ports of the transmitting antenna is N tx =16. In addition, the number of resources in the CSI resource subset is also different. In this embodiment, the UE obtains 24 H i ,i=1,2,3,…,24, by the 24 H i Recover H real , that is, the final solution equation is as follows:

[0114] Considering N within the RS receiving system bandwidth c subcarriers, a total of N c Group (k=1,2,…,N c ) as a set of inputs to the classification model, As the output of the classification model, the matrix elements are in the form of phase index θ, that is, the classification model can obtain 48 phase indexes θ. tx N of the embodiment shown in FIG9a tx Differently, in this embodiment, each logical port of the transmitting antenna has N optional phases. θ= N1 × N2 × O1 × O2 = 4 × 4 × 4 × 4 = 256, so the value range of the phase index in this embodiment becomes {0, 1, 2, ..., 255}. The difference between this embodiment and the embodiment shown in Figure 9a is that the dimensions of the input and output matrices are different, so the size of the classification model used will be different.

[0115] The disclosed embodiments use a beam management reference signal for channel state information feedback. The beam management reference signal and CSI resources can be multiplexed to complete the complete CSI feedback link. That is, the CSI resources used for beam management are grouped according to the physical ports of the transmitting antenna, and resource configuration and reporting configuration are performed. The introduction of AI channel recovery technology enables CSI feedback to use the beam management RS to recover the full channel state information, thereby greatly reducing system resources and greatly improving system transmission efficiency.

[0116] The embodiments of the present disclosure can partially or completely replace traditional CSI feedback solutions to complete system CSI feedback functions. For example, when periodic CSI-RS transmission is used in traditional CSI feedback solutions, partially using the solutions of the embodiments of the present disclosure to assist CSI feedback can reduce CSI-RS transmission density, thereby reducing feedback link resource usage and reference signal transmission. When completely replacing traditional feedback solutions, the resource usage of the CSI feedback link and reference signal transmission are no longer required.

[0117] The present disclosure also provides a base station device and a user equipment, as shown in FIG10 , including:

[0118] at least one processor 1001;

[0119] a memory 1002 storing at least one program, wherein when the at least one program is executed by the at least one processor, the at least one processor implements the channel state information feedback method provided in the aforementioned embodiments;

[0120] At least one I / O interface 1003 is connected between the processor and the memory and is configured to implement information exchange between the processor and the memory.

[0121] Among them, the processor 1001 is a device with data processing capabilities, including but not limited to a central processing unit (CPU); the memory 1002 is a device with data storage capabilities, including but not limited to random access memory (RAM, more specifically SDRAM, DDR, etc.), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), and flash memory (FLASH); the I / O interface (read-write interface) 1003 is connected between the processor 1001 and the memory 1002, and can realize information interaction between the processor 1001 and the memory 1002, including but not limited to a data bus (Bus), etc.

[0122] In some embodiments, the processor 1001 , the memory 1002 , and the I / O interface 1003 are connected to each other via a bus, and further connected to other components of the computing device.

[0123] An embodiment of the present disclosure further provides a computer-readable medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the channel state information feedback method provided in the aforementioned embodiments is implemented.

[0124] It will be appreciated by those skilled in the art that all or some of the steps in the method disclosed above, and the functional modules / units in the device can be implemented as software, firmware, hardware, and appropriate combinations thereof. In a hardware implementation, the division between the functional modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed by several physical components in cooperation. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or implemented as hardware, or implemented as an integrated circuit, such as an application-specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium (or non-transitory medium) and a communication medium (or temporary medium). As is well known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer. In addition, it is well known to those skilled in the art that communication media generally embodies computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0125] Example embodiments have been disclosed herein, and although specific terms are employed, they are used and should be interpreted only in a general illustrative sense and not for purposes of limitation. In some instances, it will be apparent to those skilled in the art that, unless otherwise expressly indicated, features, characteristics, and / or elements described in conjunction with a particular embodiment may be used alone or in combination with features, characteristics, and / or elements described in conjunction with other embodiments. Therefore, it will be understood by those skilled in the art that various changes in form and detail may be made without departing from the scope of the invention as set forth in the appended claims.

Claims

1. A channel state information feedback method, applied to a base station device, wherein the transmitting antennas of the base station device include at least one transmitting antenna group, and the method includes: In a beam management process, sending resource configuration information for channel state information (CSI) feedback to a user equipment, where the resource configuration information includes a CSI resource set for beam management and a mapping relationship between a pre-set CSI resource subset for beam management and the transmitting antenna group, and the CSI resource set for beam management includes at least one CSI resource subset; Sending reporting configuration information for CSI feedback to the user equipment, where the reporting configuration information includes target transmitting antenna group information; Sending a reference signal (RS) for beam management to the user equipment; Wherein, the resource configuration information and the reporting configuration information are used to instruct the user equipment to determine first channel information of a target CSI resource subset according to the RS for beam management, and determine second channel information for CSI feedback according to the first channel information, and the target CSI resource subset is the CSI resource subset corresponding to the target transmitting antenna group information.

2. The method according to claim 1, wherein The target transmitting antenna group information satisfies the following conditions: The product of the number of CSI resources of the target CSI resource subset corresponding to the target transmitting antenna group information and the number of logical ports of the transmitting antenna is greater than or equal to the number of physical ports of the target transmitting antenna group; wherein, the target transmitting antenna group is the transmitting antenna group corresponding to the target transmitting antenna group information, and the logical port of the transmitting antenna is the port for sending the RS for beam management.

3. The method according to claim 1, wherein The steps of establishing the mapping relationship include: Dividing the CSI resource set for beam management into at least one CSI resource subset according to the physical ports of the transmitting antennas, where the physical ports of the transmitting antennas of each CSI resource in the same CSI resource subset are the same, and a transmitting antenna group includes multiple physical ports of the transmitting antennas; Dividing the transmitting antennas with the same physical ports into one transmitting antenna group, and configuring group information for each transmitting antenna group; Establishing a mapping relationship between each CSI resource subset and each transmitting antenna group, and the CSI resource subset information of each CSI resource subset corresponds one-to-one with the transmitting antenna group information of each transmitting antenna group.

4. A channel state information feedback method, applied to a user equipment, and the method includes: Receiving the resource configuration information for channel state information (CSI) feedback and the reporting configuration information for CSI feedback sent by a base station device; wherein, the resource configuration information includes a CSI resource set for beam management and a mapping relationship between a pre-set CSI resource subset for beam management and a transmitting antenna group, and the CSI resource set for beam management includes at least one CSI resource subset; the reporting configuration information includes target transmitting antenna group information; Determine a target CSI resource subset corresponding to the target transmit antenna group information according to the mapping relationship, and receive a reference signal RS for beam management sent by the base station device on the CSI resources of the target CSI resource subset; Determine first channel information of the target CSI resource subset according to the RS; Determine second channel information according to the first channel information, where the second channel information is full channel information for CSI feedback.

5. The method according to claim 4, wherein There are multiple pieces of the target transmit antenna group information. The determining first channel information of the target CSI resource subset according to the RS includes: Determine a target CSI resource subset corresponding to each piece of the target transmit antenna group information; For the target CSI resource subset corresponding to each piece of the target transmit antenna group information, determine first channel information of the target CSI resource subset corresponding to the target transmit antenna group information according to the RS; The determining second channel information according to the first channel information includes: For the first channel information of the target CSI resource subset corresponding to each piece of the target transmit antenna group information, determine second channel information corresponding to the first channel information.

6. The method according to claim 4, wherein, The determining second channel information according to the first channel information includes: Determine a beamforming matrix according to the first channel information and a pre-trained classification model, where the classification model is a neural network model; Determine the second channel information according to the beamforming matrix and the first channel information.

7. The method according to claim 6, wherein, The first channel information is a first channel matrix, which is determined according to intermediate channel matrices of each logical port of the transmit antenna of the base station device, and the beamforming matrix is determined according to intermediate beamforming matrices of each logical port of the transmit antenna of the base station device.

8. The method according to claim 6, wherein, The determining a beamforming matrix according to the first channel information and a pre-trained classification model includes: Input the first channel information into the classification model to obtain each phase index; Determine the values of each element in the beamforming matrix according to each phase index.

9. The method according to claim 8, wherein The inputting the first channel information into the classification model includes: Input the first channel information of each subcarrier in the entire bandwidth into the classification model.

10. The method according to claim 9, wherein, In the case of broadband granularity, the number of the phase indices is (N p × N csi ), where the N p is the number of logical ports of the transmit antennas of the base station device, and the N csi is the number of CSI resources of the target CSI resource subset; In the case of narrowband granularity, the number of the phase indices is (N c × N p × N csi ), where the N c is the total number of subcarriers in the entire bandwidth.

11. The method according to any one of claims 7 to 10, wherein, The size of the first channel matrix is N rx × (N p × N csi ), where the N rx is the number of receive antennas of the user equipment, and the N p is the number of logical ports of the transmit antennas of the base station equipment, and the N csi is the number of CSI resources of the target CSI resource subset.

12. The method according to claim 11, wherein the reported configuration information further includes the number of physical ports of the target transmit antenna group, and the target transmit antenna group is the transmit antenna group corresponding to the target transmit antenna group information; the second channel information is a second channel matrix of size N rx ×N tx , and the N tx is the number of physical ports of the target transmit antenna group.

13. A base station device, comprising: One or more processors; A storage device having one or more programs stored thereon; When the one or more programs are executed by the one or more processors, the one or more processors are caused to implement the channel state information feedback method according to any one of claims 1-3; One or more I / O interfaces connected between the processor and the storage device, configured to implement information interaction between the processor and the storage device.

14. A user equipment, comprising: One or more processors; A storage device having one or more programs stored thereon; When the one or more programs are executed by the one or more processors, the one or more processors are caused to implement the channel state information feedback method according to any one of claims 4-12; One or more I / O interfaces, connected between the processor and the storage device, configured to enable information interaction between the processor and the storage device.

15. A computer-readable medium having a computer program stored thereon, wherein, When the computer program is executed by the processor, it implements the channel state information feedback method according to any one of claims 1-3, or the channel state information feedback method according to any one of claims 4-12.

Images