Channel estimation method and apparatus, electronic device, storage medium, and program product

Abstract

The application relates to a channel estimation method, device, electronic equipment, storage medium and program product, and relates to the technical field of communication. The channel estimation method comprises the following steps: determining a pilot signal matrix according to the number of transmitting antennas and the pilot length corresponding to the communication signal of a target channel; constructing a to-be-processed matrix of the target channel according to the pilot signal matrix, the noise power of the communication signal and the number of receiving antennas; determining time domain weight and frequency domain weight based on a correlation influence parameter and a weight decision model; generating an iteration initial value according to the time domain weight, a first reference matrix corresponding to a second time slot and a first subcarrier, the frequency domain weight and a second reference matrix corresponding to a first time slot and a second subcarrier; performing iteration operation on the to-be-processed matrix according to the iteration initial value and a preset inverse iteration model; and obtaining the inverse matrix of the to-be-processed matrix to determine the channel matrix of the target channel when the preset iteration termination condition is met. The method can improve the communication performance.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of communication, and in particular to a channel estimation method and device, electronic equipment, storage medium and program product. BACKGROUND

[0002] Matrix operation is an important mathematical tool in the field of communication technology, and matrix inversion operation is a commonly used operation in matrix operation and a key step in channel estimation scenarios. In the channel estimation scenario, the matrix inversion operation usually adopts an approximate iterative inversion manner to gradually approach the approximate value of the inversion result. However, the approximate iterative inversion manner has a large number of iterations and a large amount of calculation, resulting in low channel estimation efficiency, and thus the problem of mismatch between the channel estimation result and the actual channel state occurs, affecting the communication performance. SUMMARY

[0003] To overcome the problems in the related art, the present application provides a channel estimation method and device, electronic equipment, storage medium and program product.

[0004] According to a first aspect of an embodiment of the present application, a channel estimation method is provided, the method comprising:

[0005] determining a pilot signal matrix corresponding to the communication signal according to the number of transmit antennas and the pilot length corresponding to the communication signal transmitted by the target channel, the target channel being a channel corresponding to a first time slot and a first subcarrier;

[0006] constructing a to-be-processed matrix of the target channel according to the pilot signal matrix, the noise power of the communication signal and the number of receive antennas;

[0007] determining a time domain weight and a frequency domain weight based on a correlation influence parameter and a weight decision model;

[0008] generating an iteration initial value according to the time domain weight, a first reference matrix corresponding to a second time slot and the first subcarrier, the frequency domain weight and a second reference matrix corresponding to the first time slot and a second subcarrier, the second time slot being used to indicate a previous time slot of the first time slot in a continuously distributed time domain resource, and the second subcarrier being used to indicate a previous subcarrier of the first subcarrier in a continuously distributed frequency resource;

[0009] performing iterative operation on the to-be-processed matrix according to the iteration initial value and a preset inversion iteration model, and in the iterative operation process, determining an iteration step length according to a step length decision model, and updating the preset inversion iteration model for the next iteration operation according to the iteration step length;

[0010] obtaining an inverse matrix of the to-be-processed matrix when a preset iteration termination condition is met;

[0011] determining a channel matrix of the target channel according to an inverse matrix of the to-be-processed matrix.

[0012] In this embodiment, the iteration initial value is generated according to the reference matrices of the channel that have a correlation with the time-frequency resources of the target channel, i.e., the first reference matrix and the second reference matrix, so that the iteration result can more quickly approximate the optimal solution, the number of iterations can be reduced, the amount of calculation can be reduced, the iteration step can be updated in time in each iteration operation, the oscillation can be avoided, the iteration result can more quickly approximate the optimal solution, the number of iterations can be further reduced, the amount of calculation can be further reduced, the channel estimation efficiency can be improved, the problem that the channel estimation result does not match the actual channel state can be avoided, and the communication performance can be improved.

[0013] In an example embodiment, the method further includes:

[0014] obtaining a channel matrix of a first channel and a channel matrix of a second channel, the first channel being a channel corresponding to the second time slot and the first subcarrier, and the second channel being a channel corresponding to the first time slot and the second subcarrier;

[0015] taking the channel matrix of the first channel as the first reference matrix and taking the channel matrix of the second channel as the second reference matrix.

[0016] In this embodiment, the channel matrix of the first channel and the channel matrix of the second channel are used to generate the iteration initial value, so that the iteration initial value is closer to the channel state of the target channel, thereby reducing the number of iterations and reducing the amount of calculation.

[0017] In an example embodiment, the determining of the time domain weight and the frequency domain weight based on the correlation influence parameter and the weight decision model includes:

[0018] inputting the moving speed of the electronic device, the correlation between the second time slot and the first time slot, and the correlation between the first subcarrier and the second subcarrier into the weight decision model to obtain the time domain weight and the frequency domain weight;

[0019] The weight decision model is trained according to a reference moving speed of the electronic device, a correlation between reference time domains, a correlation between reference frequency domains, and reference time domain weight and reference frequency domain weight.

[0020] In this embodiment, the time domain weight and the frequency domain weight are determined by the pre-trained weight decision model, so that the obtained weight is more consistent with the correlation characteristics between the time domains and the correlation characteristics between the frequency domains, to ensure the reliability of the iteration initial value generated according to the weight.

[0021] In an example embodiment, the generating an initial value of iteration according to the time domain weight, a first reference matrix corresponding to the first time slot and the first subcarrier, the frequency domain weight, and a second reference matrix corresponding to the first time slot and the second subcarrier comprises:

[0022] encoding the first reference matrix, extracting features of the first reference matrix, and decoding the features of the first reference matrix to obtain a first matrix after reconstruction of the first reference matrix;

[0023] encoding the second reference matrix, extracting features of the second reference matrix, and decoding the features of the second reference matrix to obtain a second matrix after reconstruction of the second reference matrix;

[0024] generating the initial value of iteration according to the time domain weight and the first matrix, and the frequency domain weight and the second matrix.

[0025] In the embodiment, the hidden relationship in the reference matrix can be mined by reconstructing the first reference matrix and the second reference matrix, and the reconstructed matrix can better reflect the real channel state, so that the generated initial value of iteration is closer to the optimal solution.

[0026] In an example embodiment, the preset inverse iteration model is used to indicate the association between the iteration result of the current iteration and the iteration result of the next iteration, and the step decision model is used to determine the iteration step, and the preset inverse iteration model of the next iteration is updated according to the iteration step.

[0027] inputting the condition number of the to-be-processed matrix, the iteration residual, and the number of times of the current iteration into the step decision model to obtain the iteration step;

[0028] determining the preset inverse iteration model corresponding to the next iteration according to the iteration step, the iteration result of the current iteration, and the to-be-processed matrix.

[0029] In the embodiment, the iteration step is determined according to the iteration residual and the step decision model, the iteration step can be adjusted in time according to the iteration result obtained each time, the iteration step is more matched to the current iteration condition, and the oscillation is effectively avoided.

[0030] In an example embodiment, the preset iteration termination condition is that the iteration residual is less than a first threshold value, and the convergence probability is greater than a second threshold value.

[0031] In the embodiment, the iteration is terminated when the iteration residual is less than the first threshold value and the convergence probability is greater than the second threshold value, the optimal solution can be obtained in time, and the inverse efficiency is improved.

[0032] In an example embodiment, the determining the channel matrix of the target channel according to the inverse matrix of the to-be-processed matrix comprises:

[0033] inputting the inverse matrix of the to-be-processed matrix into a preset channel estimation model to obtain the channel matrix of the target channel, wherein the preset channel estimation model is one of a linear minimum mean square error channel estimation model and a least square channel estimation model.

[0034] In the embodiment, the channel estimation is performed through the channel estimation model according to the inverse matrix of the to-be-processed matrix, and an accurate channel matrix can be obtained.

[0035] In an example embodiment, the method further comprises:

[0036] obtaining scene information, and determining the corresponding preset channel estimation model according to the scene information.

[0037] In the embodiment, the preset channel estimation model is selected according to the application scene, so that the preset channel estimation model can be more matched with the current application scene, and the accuracy of the channel matrix is further ensured.

[0038] According to a second aspect of an embodiment of the present application, a channel estimation device is provided, and the device comprises:

[0039] a first determining module configured to determine a pilot signal matrix corresponding to a communication signal according to a number of transmit antennas and a pilot length corresponding to the target channel, the target channel being a channel corresponding to a first time slot and a first subcarrier;

[0040] a constructing module configured to construct a to-be-processed matrix of the target channel according to the pilot signal matrix, a noise power of the communication signal and a number of receive antennas;

[0041] a second determining module configured to determine a time domain weight and a frequency domain weight based on a correlation influence parameter and a weight decision model;

[0042] a generating module configured to generate an iteration initial value according to the time domain weight, a first reference matrix corresponding to the first time slot and the first subcarrier, the frequency domain weight and a second reference matrix corresponding to the first time slot and a second subcarrier, the second time slot being used to indicate a previous time slot of the first time slot in a continuously distributed time domain resource, and the second subcarrier being used to indicate a previous subcarrier of the first subcarrier in a continuously distributed frequency resource;

[0043] an iteration module configured to perform iteration operation on the to-be-processed matrix according to the iteration initial value and a preset inverse iteration model, wherein during the iteration operation, an iteration step is determined according to a step decision model, and the preset inverse iteration model for next iteration operation is updated according to the iteration step;

[0044] a third determination module configured to obtain an inverse matrix of the to-be-processed matrix when a preset iteration termination condition is met;

[0045] a fourth determination module configured to determine a channel matrix of the target channel according to the inverse matrix of the to-be-processed matrix.

[0046] According to a third aspect of an embodiment of the present application, an electronic device is provided, comprising:

[0047] a processor;

[0048] a memory for storing processor-executable instructions;

[0049] The processor is configured to perform the method as described in the first aspect of the embodiment of the present application.

[0050] According to a fourth aspect of an embodiment of the present application, a non-transitory computer-readable storage medium is provided, when instructions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to perform the method as described in the first aspect of the embodiment of the present application.

[0051] According to a fifth aspect of an embodiment of the present application, a computer program product is provided, comprising a computer program, when the computer program is executed by a processor, the method as described in the first aspect of the embodiment of the present application is implemented. BRIEF DESCRIPTION OF DRAWINGS

[0052] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and serve to explain the principles of the present application.

[0053] Figure 1 is a flow chart of a channel estimation method according to an exemplary embodiment;

[0054] Figure 2 is a flow chart of generating an iteration initial value according to an exemplary embodiment;

[0055] Figure 3 is a flow chart of determining a preset inverse iteration model according to an exemplary embodiment;

[0056] Figure 4 is a block diagram of a channel estimation device according to an exemplary embodiment;

[0057] Figure 5 is a block diagram of an electronic device according to an exemplary embodiment. DETAILED DESCRIPTION

[0058] The exemplary embodiments will be described in detail with reference to the accompanying drawings. In the following description, same drawing reference numerals are used to denote elements having the same or similar functions and / or compositions. The following exemplary embodiments are not representative of all embodiments consistent with the present disclosure.

[0059] In an exemplary embodiment of the present disclosure, a channel estimation method is provided, Figure 1 is a flowchart of a channel estimation method according to an exemplary embodiment, as shown in Figure 1 includes the following steps S101 to S107.

[0060] In step S101, a pilot signal matrix corresponding to a communication signal is determined according to the number of transmit antennas and the length of a pilot corresponding to the communication signal transmitted by a target channel. The target channel is a channel corresponding to a first time slot and a first subcarrier.

[0061] In step S102, a matrix to be processed of the target channel is constructed according to the pilot signal matrix, the noise power of the communication signal, and the number of receive antennas.

[0062] In step S103, a time domain weight and a frequency domain weight are determined based on a correlation influence parameter and a weight decision model.

[0063] In step S104, an initial value of iteration is generated according to the time domain weight, a first reference matrix corresponding to a second time slot and the first subcarrier, the frequency domain weight, and a second reference matrix corresponding to the first time slot and a second subcarrier. The second time slot is used to indicate a previous time slot of the first time slot in a continuously distributed time domain resource, and the second subcarrier is used to indicate a previous subcarrier of the first subcarrier in a continuously distributed frequency resource.

[0064] In step S105, an iteration operation is performed on the matrix to be processed according to the initial value of iteration and a preset inverse iteration model. In the iteration operation, an iteration step is determined according to a step decision model, and the preset inverse iteration model for the next iteration operation is updated according to the iteration step.

[0065] In step S106, an inverse matrix of the matrix to be processed is obtained when a preset iteration termination condition is satisfied.

[0066] In step S107, a channel matrix of the target channel is determined according to the inverse matrix of the matrix to be processed.

[0067] The method in the embodiment of the application is applied to an electronic device, including a mobile phone, a tablet, a smart wearable device, a router and the like having a wireless communication function. Optionally, the electronic device includes a Multiple-Input Multiple-Output (MIMO) system, in which a plurality of antennas are simultaneously used by a transmitting end and a receiving end, and the number of transmitting antennas is the same as the number of receiving antennas.

[0068] In step S101, the target channel is a channel corresponding to a first time slot and a first subcarrier, the first time slot is a time domain resource used for transmitting and receiving a communication signal by the target channel, and the first subcarrier is a frequency domain resource used for transmitting and receiving the communication signal by the target channel, wherein the first time slot can be any time slot in the time domain resource, and the first subcarrier can be any subcarrier in the frequency domain resource. The pilot length corresponding to the communication signal represents the length of the time domain resource occupied by the transmitting pilot signal, which can be denoted as , and the number of transmitting antennas in the electronic device can be denoted as According to the pilot length and the number of transmitting antennas, a pilot signal matrix corresponding to the communication signal is constructed, and the dimension of the pilot signal matrix is .

[0069] In step S102, the noise power of the communication signal represents the power of the noise generated when the communication signal is received by the receiving antenna, and according to the pilot signal matrix, the noise power of the communication signal and the number of receiving antennas, a to-be-processed matrix of the target channel is constructed by the following formula:

[0070]

[0071] wherein represents the to-be-processed matrix of the target channel, represents the position of the first time slot in the time domain resource, and the first time slot can also be denoted as the th time slot, represents the number of the first subcarrier in the frequency domain resource, and the first subcarrier can also be denoted as the th subcarrier, represents the pilot signal matrix with the dimension of , represents the conjugate matrix of the pilot signal matrix, with the dimension of , represents the number of transmitting antennas, represents the pilot length, represents the noise power of the communication signal, represents the unit matrix with the dimension of , represents the number of receiving antennas, and thus, The dimension of the matrix is Or .

[0072] In step S103, the correlation includes time domain correlation and frequency domain correlation, the time domain correlation represents the correlation between different time slots in the time domain resource, and the frequency domain correlation represents the correlation between different subcarriers in the frequency domain resource. The correlation influence parameter represents a parameter influencing the time domain correlation and / or the frequency domain correlation, for example, the parameter influencing the time domain correlation includes the moving speed of the electronic device and the communication environment, the faster the moving speed, the more complex the communication environment, and the lower the time domain correlation; the parameter influencing the frequency domain correlation includes the subcarrier spacing and the communication environment, the larger the subcarrier spacing, the more complex the communication environment, and the lower the frequency domain correlation. The weight decision model represents a model for determining the time domain weight and the frequency domain weight according to the correlation influence parameter, the time domain weight represents the weight of the reference matrix corresponding to the channel different from the time domain resource of the target channel, and the frequency domain weight represents the weight of the reference matrix corresponding to the channel different from the frequency domain resource of the target channel.

[0073] In step S104, the second time slot is used to indicate the previous time slot of the first time slot in the continuously distributed time domain resource, for example, the continuously distributed time domain resource includes four time slots numbered t1, t2, t3, and t4, when the first time slot is t3, the second time slot is t2. The second subcarrier is used to indicate the previous subcarrier of the first subcarrier in the continuously distributed frequency resource, for example, the continuously distributed frequency domain resource includes four subcarriers numbered k1, k2, k3, and k4, when the first subcarrier is k2, the second subcarrier is k1.

[0074] The first reference matrix can be a channel matrix corresponding to the second time slot and the first subcarrier, or an inverse matrix of a to-be-processed matrix of a channel corresponding to the second time slot and the first subcarrier; the second reference matrix can be a channel matrix corresponding to the first time slot and the second subcarrier, or an inverse matrix of a to-be-processed matrix of a channel corresponding to the first time slot and the second subcarrier. The first reference matrix and the second reference matrix are of the same type, both being a channel matrix or both being an inverse matrix of a to-be-processed matrix.

[0075] The first reference matrix is a reference matrix corresponding to a channel different from the time domain resource of the target channel, and the time domain weight is the weight of the first reference matrix; the second reference matrix is a reference matrix corresponding to a channel different from the frequency domain resource of the target channel, and the frequency domain weight is the weight of the second reference matrix. The iteration initial value is generated according to the product of the time domain weight and the first reference matrix, and the product of the frequency domain weight and the second reference matrix, for example, the iteration initial value can be represented as:

[0076]

[0077] wherein, Indicates the target channel (the first) Time slot and the The initial value for iteration corresponding to the subcarrier. Represents frequency domain weights, Indicates the first Time slot and the The second reference matrix corresponding to the subcarrier, Represents time-domain weights, Indicates the first Time slot and the The first reference matrix corresponding to the subcarrier.

[0078] In step S105, a preset inversion iterative model is used to perform iterative operations on the matrix to be processed to obtain the inverse matrix. When performing iterative operations on the matrix to be processed using the preset inversion iterative model, the initial iteration value and iteration step size are input into the preset inversion iterative model, and the result of the next iteration is output. Then, the result of the next iteration and the iteration step size are input into the preset inversion iterative model, and the result of the next iteration is output. The initial iteration value uses the initial iteration value calculated in the previous steps. The initial iteration value can be understood as the initial value of the inverse matrix of the matrix to be processed. The result of each iteration can be understood as gradually approximating the value of the inverse matrix of the matrix to be processed. The iteration step size is determined according to the step size decision model and is used to update the preset inversion iterative model for the next iteration. The iteration step size for each iteration is adjusted according to the actual iteration result to ensure the adaptability of the iteration step size and avoid oscillations.

[0079] In step S106, the preset iteration termination condition is determined based on the iteration residual and the convergence probability. The iteration residual represents the difference between the iteration results obtained from two adjacent iterations. The smaller the iteration residual, the closer the iteration result is to the optimal solution, i.e., the closer the iteration result is to the inverse matrix of the matrix to be processed. The convergence probability represents the probability that the iteration result obtained from the current iteration operation converges to the optimal solution. The larger the convergence probability, the closer the iteration result is to the optimal solution. When the iteration operation meets the preset iteration termination condition, the iteration result obtained at the time of termination is taken as the inverse matrix of the matrix to be processed.

[0080] In step S107, after obtaining the inverse matrix of the matrix to be processed, channel estimation is performed based on the inverse matrix of the matrix to be processed and the preset channel estimation model to obtain the channel matrix of the target channel.

[0081] In an exemplary embodiment of the present invention, after constructing the processing matrix of the target channel, an initial iteration value is generated based on the time-domain weight, the first reference matrix corresponding to the second time slot and the first subcarrier, the frequency-domain weight, and the second reference matrix corresponding to the first time slot and the second subcarrier. Compared with related technologies that use fixed initial iteration values ​​or random numbers as initial iteration values, this method generates initial iteration values ​​based on the reference matrix of the channel that is correlated with the time-frequency resources of the target channel. This allows the iteration result to approach the optimal solution faster, reduces the number of iterations, and reduces the amount of computation. Then, based on the initial iteration value and the preset inversion iteration model, iterative calculations are performed on the processing matrix, and during the iterative calculation process, the step size is determined. The model determines the iteration step size and updates the preset inverse iteration model for the next iteration operation based on the iteration step size. Compared with related technologies that use a fixed iteration step size or use random numbers as the iteration step size, this method updates the iteration step size in a timely manner in each iteration operation, which can avoid oscillations and make the iteration results approach the optimal solution faster, reducing the number of iterations and the amount of computation. When the preset iteration termination condition is met, the inverse matrix of the matrix to be processed is obtained. Channel estimation is performed based on the inverse matrix of the matrix to be processed to obtain the channel matrix of the target channel, which can improve the efficiency of channel estimation and avoid the problem of mismatch between the channel estimation results and the actual channel state, thereby improving communication performance.

[0082] In some embodiments, the first reference matrix and the second reference matrix are obtained through the following steps:

[0083] Obtain the channel matrix of the first channel and the channel matrix of the second channel. The first channel is the channel corresponding to the second time slot and the first subcarrier, and the second channel is the channel corresponding to the first time slot and the second subcarrier. Use the channel matrix of the first channel as the first reference matrix and the channel matrix of the second channel as the second reference matrix.

[0084] The first time slot is denoted as the [number]. Time slot, the first subcarrier is denoted as the th Subcarrier, then the second time slot is the first Time slot, the second subcarrier is denoted as the 1st time slot. Subcarriers, therefore, the channel matrix of the first channel can be denoted as... The channel matrix of the second channel can be denoted as: Using the channel matrix of the first channel as the first reference matrix and the channel matrix of the second channel as the second reference matrix, the initial value for iteration can be expressed as:

[0085]

[0086] in, Indicates the target channel (the first) Time slot and the The initial value for iteration corresponding to the subcarrier. denotes a frequency domain weight, denotes a channel matrix of a first channel corresponding to the first time slot and the first subcarrier, denotes a time domain weight, denotes a channel matrix of a second channel corresponding to the second time slot and the second subcarrier. denotes a frequency domain weight, denotes a channel matrix of a first channel corresponding to the first time slot and the first subcarrier, denotes a time domain weight, denotes a channel matrix of a second channel corresponding to the second time slot and the second subcarrier.

[0087] Since the channel matrix of the first channel can reflect the channel state of the first channel, the channel matrix of the second channel can reflect the channel state of the second channel, and the first channel and the second channel both have correlation with the target channel in time-frequency resources, therefore, taking the channel matrix of the first channel as the first reference matrix and the channel matrix of the second channel as the second reference matrix for generating the iteration initial value can make the iteration initial value closer to the channel state of the target channel, thereby reducing the iteration times.

[0088] In some embodiments, the step S103 of determining the time domain weight and the frequency domain weight based on the correlation influence parameter and the weight decision model comprises:

[0089] The moving speed of the electronic device, the correlation between the second time slot and the first time slot, and the correlation between the first subcarrier and the second subcarrier are input into the weight decision model to obtain the time domain weight and the frequency domain weight; wherein the weight decision model is trained according to the reference moving speed of the electronic device, the correlation between the reference time domain, the correlation between the reference frequency domain, and the reference time domain weight and the reference frequency domain weight.

[0090] The moving speed of the electronic device is obtained by a speed sensor in the electronic device, and the correlation between the second time slot and the first time slot, and the correlation between the first subcarrier and the second subcarrier can be obtained according to any correlation estimation algorithm, for example, obtaining the statistical values such as mean, variance, power, etc. of the communication signals received in different time slots or different frequency domains, and taking the ratio of the difference between the statistical values and the maximum value as the correlation. The smaller the difference between the statistical values, the higher the correlation. For example, the average value of the communication signal received in the first time slot is , and the average value of the communication signal received in the second time slot is , then the correlation between the second time slot and the first time slot is . The moving speed of the electronic device, the correlation between the second time slot and the first time slot, and the correlation between the first subcarrier and the second subcarrier are input into the weight decision model to output the time domain weight and the frequency domain weight.

[0091] The weight decision model is a model obtained by pre-training according to a plurality of sets of reference data, the reference data can also be understood as experimental data, each set of reference data includes a reference moving speed of the electronic device, a correlation between reference time domains, a correlation between reference frequency domains, a reference time domain weight, and a reference frequency domain weight, the correlation between the reference time domains represents a correlation between a reference first time domain and a reference second time domain, and the correlation between the reference frequency domains represents a correlation between a reference first subcarrier and a reference second subcarrier. It should be noted that the correlation between the reference time domains is calculated in the same manner as the correlation between the second time slot and the first time slot, and the correlation between the reference frequency domains is calculated in the same manner as the correlation between the first subcarrier and the second subcarrier.

[0092] In this embodiment, the time domain weight and the frequency domain weight are determined by the pre-trained weight decision model, which can make the obtained weight more consistent with the correlation characteristics between the time domains and the correlation characteristics between the frequency domains, so as to ensure the reliability of the iteration initial value generated according to the weight.

[0093] In some embodiments, the step S104 of generating the iteration initial value according to the time domain weight, the first reference matrix corresponding to the second time slot and the first subcarrier, the frequency domain weight, and the second reference matrix corresponding to the first time slot and the second subcarrier includes the following steps S1041 to S1043 as shown. Figure 2

[0094] The step S1041 encodes the first reference matrix, extracts the features of the first reference matrix, and then decodes the features of the first reference matrix to obtain a first matrix after reconstruction of the first reference matrix.

[0095] The step S1042 encodes the second reference matrix, extracts the features of the second reference matrix, and then decodes the features of the second reference matrix to obtain a second matrix after reconstruction of the second reference matrix.

[0096] The step S1043 generates the iteration initial value according to the time domain weight and the first matrix, and the frequency domain weight and the second matrix.

[0097] The first reference matrix is encoded by the first matrix encoder to extract the features of the first reference matrix, the first matrix encoder is composed of a plurality of layers of convolutional neural networks (CNN), for example, composed of 3 layers of CNN. Then the features of the first reference matrix are decoded by the first matrix decoder to realize feature mapping, the first matrix encoder is composed of a plurality of layers of transposed CNN, for example, composed of 3 layers of transposed CNN, and the reconstruction of the first reference matrix is realized through the feature mapping, and the matrix after reconstruction of the first reference matrix is taken as the first matrix.

[0098] ​The second reference matrix is ​​encoded by a second matrix encoder to extract its features. The second matrix encoder consists of a multi-layer convolutional neural network (CNN), such as a 3-layer CNN. The second matrix encoder can be the same as or different from the first matrix encoder. Then, the features of the second reference matrix are decoded by a second matrix decoder to achieve feature mapping. The second matrix encoder consists of a multi-layer transposed CNN, such as a 3-layer transposed CNN. The second matrix decoder can be the same as or different from the first matrix decoder. The second reference matrix is ​​reconstructed through feature mapping, and the reconstructed matrix is ​​used as the second matrix.

[0099] For example, the second matrix encoder is the same as the first matrix encoder, and the second matrix decoder is the same as the first matrix decoder. The matrix encoder is denoted as... The matrix decoder is denoted as , No. Time slot and the The first reference matrix corresponding to the subcarrier is denoted as , No. Time slot and the The second reference matrix corresponding to the subcarrier is denoted as Then the first matrix Represented as: The second matrix Represented as: .

[0100] After obtaining the first and second matrices, the initial values ​​for iteration are generated based on the time-domain weights and the first matrix, as well as the frequency-domain weights and the second matrix, using the following formula:

[0101]

[0102] in, Indicates the target channel (the first) Time slot and the The initial value for the iteration corresponding to the subcarrier. Indicates a matrix encoder. Represents a matrix decoder. Represents the second matrix, Describes the first matrix. Represents the frequency domain weights. This represents the time-domain weight.

[0103] Optionally, when the channel matrix of the first channel is used as the first reference matrix and the channel matrix of the second channel is used as the second reference matrix, the formula for generating the initial values ​​of the iteration above... It can be replaced with the first Time slot and the Channel matrix of the second channel corresponding to the subcarrier , the first channel corresponding to the first time slot and the first subcarrier the first channel corresponding to the first time slot and the first subcarrier the channel matrix of the first channel corresponding to the first time slot and the first subcarrier .

[0104] Optionally, when the inverse matrix of the to-be-processed matrix of the first channel is taken as the first reference matrix and the inverse matrix of the to-be-processed matrix of the second channel is taken as the second reference matrix, the in the formula for generating the iteration initial value can be replaced by the inverse matrix of the to-be-processed matrix of the second channel corresponding to the first time slot and the first subcarrier the inverse matrix of the to-be-processed matrix of the second channel corresponding to the first time slot and the first subcarrier the inverse matrix of the to-be-processed matrix of the first channel corresponding to the first time slot and the first subcarrier , the inverse matrix of the to-be-processed matrix of the first channel corresponding to the first time slot and the first subcarrier the inverse matrix of the to-be-processed matrix of the first channel corresponding to the first time slot and the first subcarrier the inverse matrix of the to-be-processed matrix of the first channel corresponding to the first time slot and the first subcarrier .

[0105] In this embodiment, by reconstructing the first reference matrix and the second reference matrix, the hidden relationship in the reference matrix can be mined, and the reconstructed matrix can better reflect the real channel state, so that the generated iteration initial value is closer to the optimal solution.

[0106] In some embodiments, the preset inverse iteration model in the step S105 is used to indicate the correlation between the iteration result of the current iteration and the iteration result of the next iteration, the iteration step is determined according to the step decision model, and the preset inverse iteration model for the next iteration operation is updated according to the iteration step, including the following steps S1051 to S1052 as shown in the figure. Figure 3

[0107] In step S1051, the condition number of the to-be-processed matrix, the iteration residual and the number of times of the current iteration are input into the step decision model to obtain the iteration step.

[0108] In step S1052, the preset inverse iteration model corresponding to the next iteration is determined according to the iteration step, the iteration result of the current iteration and the to-be-processed matrix.

[0109] ​The condition number of the matrix to be processed characterizes the singularity of the matrix. The larger the condition number, the closer the matrix is ​​to singularity, meaning a higher probability that the matrix is ​​not invertible or that the error of the inverse matrix is ​​larger, and the more prone the iterative operation is to oscillation. The iteration residual represents the difference between the result of the current iteration and the result of the previous iteration, or the Frobenius norm of the difference, used to characterize the magnitude of parameter change between the two iterations. The larger the iteration residual, the more severe the oscillation of the previous iteration. The current iteration number indicates which iteration the current operation is in. The step size decision model is a pre-trained machine learning model that determines the iteration step size based on the above three parameters, such as a Multi-Layer Perceptron (MLP) model. The condition number of the matrix to be processed, the iteration residual, and the current iteration number are input into the step size decision model, and the iteration step size is output. The iteration step size can be expressed as:

[0110]

[0111] in, Indicates the current number The iteration step size of the next iteration Indicates the first The iteration result of the next iteration. Indicates the first The iteration result of the next iteration. The matrix to be processed representing the target channel The condition number of Represents the iterative residual, i.e. and The Frobenius norm of the difference between them This represents the step-size decision model.

[0112] The pre-defined inverse iteration model is used to indicate the relationship between the iteration result of the current iteration and the iteration result of the next iteration. After obtaining the iteration result and iteration step size of the current iteration, the pre-defined inverse iteration model corresponding to the next iteration is expressed by the following formula:

[0113]

[0114] in, Indicates the first The iteration result of the next iteration. Indicates the first The iteration result of the next iteration. Indicates the first The iteration step size of the next iteration This represents the matrix to be processed in the target channel.

[0115] It should be noted that the first iteration iterates over the initial value, and the iteration step size for the first iteration is 1. ,in, As the initial value for iteration, Therefore, the iteration step size of the second iteration is The result of the second iteration is .

[0116] In this embodiment, the iteration step size is determined based on the iteration residual and step size decision model. This allows for timely adjustment of the iteration step size according to the iteration results obtained in each iteration, making the iteration step size more suitable for the current iteration situation and effectively avoiding oscillations.

[0117] In some embodiments, the preset iteration termination condition in step S106 is that the iteration residual is less than a first threshold and the convergence probability is greater than a second threshold.

[0118] For the In the next iteration, the iterative residual is denoted as The convergence probability is denoted as The first and second thresholds are empirical values; for example, the first threshold is... The second threshold is 0.99. and Then the iteration terminates, and the first iteration is... The result of the iteration is used as the inverse of the matrix to be processed.

[0119] In this embodiment, the iteration is terminated when the iterative residual is less than the first threshold and the convergence probability is greater than the second threshold, which can obtain the optimal solution in a timely manner and improve the efficiency of inversion.

[0120] In some embodiments, determining the channel matrix of the target channel based on the inverse matrix of the matrix to be processed in step S107 above includes:

[0121] The inverse of the matrix to be processed is input into the preset channel estimation model to obtain the channel matrix of the target channel. The preset channel estimation model is one of the linear minimum mean square error channel estimation model and the least squares channel estimation model.

[0122] When the preset channel estimation model is the least squares channel estimation model, the least squares channel estimation model is represented by the following formula. The inverse matrix of the matrix to be processed is input into this formula, and the channel matrix of the target channel is output:

[0123]

[0124] in, Indicates the target channel (the first) Time slot and the Channel matrix of subcarriers, This represents the inverse of the matrix to be processed in the target channel. denotes a conjugate matrix of a pilot signal matrix when the pilot signal is sent, and has a dimension of , denotes a pilot signal matrix when the pilot signal is received, and has a dimension of , denotes a number of receiving antennas, and , then has a dimension of , according to the step S102 mentioned above, therefore, has a dimension of or .

[0125] When the preset channel estimation model is a linear minimum mean square error channel estimation model, the linear minimum mean square error channel estimation model is represented by the following formula, and the inverse matrix of the to-be-processed matrix is input into the formula to output the channel matrix of the target channel:

[0126]

[0127] wherein, denotes a channel matrix of a target channel (a first time slot and a first subcarrier), denotes a covariance matrix of the target channel, and has a dimension of , denotes a conjugate matrix of a pilot signal matrix when the pilot signal is sent, and has a dimension of , denotes an inverse matrix of a to-be-processed matrix of the target channel, according to the step S102 mentioned above, , denotes a pilot signal matrix when the pilot signal is received, and has a dimension of , denotes a number of receiving antennas, and therefore, has a dimension of . In this embodiment, the channel estimation is performed through the channel estimation model according to the inverse matrix of the to-be-processed matrix, and an accurate channel matrix can be obtained.

[0128] In some embodiments, the step of determining the channel matrix of the target channel further comprises:

[0129] acquiring scene information, and determining a corresponding preset channel estimation model according to the scene information.

[0130] acquiring scene information, and determining a corresponding preset channel estimation model according to the scene information.

[0131] ​The scene information is used to represent a wireless communication scene of the electronic device, for example, including a 5G communication scene provided by a 5G communication system, a 6G communication scene provided by a 6G communication system, a WiFi communication scene, and the like. Different preset channel estimation models are selected in different communication scenes, and a mapping relationship between the scene information and the preset channel estimation model is pre-stored, for example, for the 5G communication scene and the 6G communication scene, the electronic device corresponds to using a least square channel estimation model when initially accessing a network, and corresponds to switching to using a linear minimum mean square error channel estimation model when performing data transmission after successfully accessing the network; for the WiFi communication scene, if the electronic device is a low-complexity device such as a mobile phone, a tablet, and the like, the least square channel estimation model is correspondingly used, and if the electronic device is a high-complexity device such as a high-performance router, a wireless network card, and the like, the linear minimum mean square error channel estimation model is correspondingly used. After the current scene information is obtained, the preset channel estimation model corresponding to the current scene information is found from the pre-stored mapping relationship.

[0132] In the embodiment, the preset channel estimation model is selected according to the application scene, so that the preset channel estimation model can be more matched with the current application scene, and the accuracy of the channel matrix is further ensured.

[0133] In an exemplary embodiment of the application, a channel estimation device is provided, Figure 4 is a block diagram of a channel estimation device according to an exemplary embodiment, as Figure 4 shown, the channel estimation device comprises:

[0134] The first determination module 401 is configured to determine a pilot signal matrix corresponding to a communication signal according to the number of transmit antennas and the pilot length corresponding to the communication signal transmitted by a target channel, and the target channel is a channel corresponding to a first time slot and a first subcarrier.

[0135] The construction module 402 is configured to construct a to-be-processed matrix of the target channel according to the pilot signal matrix, the noise power of the communication signal, and the number of receive antennas.

[0136] The second determination module 403 is configured to determine a time domain weight and a frequency domain weight based on a correlation influence parameter and a weight decision model.

[0137] The generation module 404 is configured to generate an iteration initial value according to the time domain weight, a first reference matrix corresponding to a second time slot and the first subcarrier, the frequency domain weight, and a second reference matrix corresponding to the first time slot and a second subcarrier, the second time slot is used to indicate a previous time slot of the first time slot in a continuously distributed time domain resource, and the second subcarrier is used to indicate a previous subcarrier of the first subcarrier in a continuously distributed frequency resource.

[0138] The iteration module 405 is configured to perform iteration operation on the to-be-processed matrix according to the preset inverse iteration model and the iteration initial value, and in the iteration operation process, determine an iteration step length according to the step length decision model, and update the preset inverse iteration model for the next iteration operation according to the iteration step length.

[0139] The third determination module 406 is configured to obtain the inverse matrix of the to-be-processed matrix when the preset iteration termination condition is met.

[0140] The fourth determination module 407 is configured to determine the channel matrix of the target channel according to the inverse matrix of the to-be-processed matrix.

[0141] In an example embodiment, the generation module 404 is further configured to:

[0142] obtain the channel matrix of the first channel and the channel matrix of the second channel, the first channel being a channel corresponding to the second time slot and the first subcarrier, and the second channel being a channel corresponding to the first time slot and the second subcarrier, take the channel matrix of the first channel as a first reference matrix, and take the channel matrix of the second channel as a second reference matrix.

[0143] In an example embodiment, the second determination module 403 is further configured to:

[0144] input the moving speed of the electronic device, the correlation between the second time slot and the first time slot, the correlation between the first subcarrier and the second subcarrier, into the weight decision model, and obtain the time domain weight and the frequency domain weight.

[0145] The weight decision model is obtained by training according to the reference moving speed of the electronic device, the correlation between the reference time domain, the correlation between the reference frequency domain, and the reference time domain weight and the reference frequency domain weight.

[0146] In an example embodiment, the generation module 404 is further configured to:

[0147] encode the first reference matrix, extract the features of the first reference matrix, and then decode the features of the first reference matrix to obtain a first matrix after reconstruction of the first reference matrix; encode the second reference matrix, extract the features of the second reference matrix, and then decode the features of the second reference matrix to obtain a second matrix after reconstruction of the second reference matrix; and generate the iteration initial value according to the time domain weight and the first matrix, and the frequency domain weight and the second matrix.

[0148] In an example embodiment, the preset inverse iteration model is used to indicate the association relationship between the iteration result of the current iteration and the iteration result of the next iteration, and the iteration module 405 is further configured to:

[0149] The condition number of the to-be-processed matrix, the iteration residual, and the number of the current iteration are input into the step length decision model to obtain an iteration step length; and the iteration result of the current iteration and the to-be-processed matrix are used to determine a preset inversion iteration model corresponding to the next iteration.

[0150] In an example embodiment, the preset iteration termination condition is that the iteration residual is less than a first threshold value and the convergence probability is greater than a second threshold value.

[0151] In an example embodiment, the fourth determination module 407 is further configured to:

[0152] The inverse matrix of the to-be-processed matrix is input into the preset channel estimation model to obtain a channel matrix of the target channel, wherein the preset channel estimation model is one of a linear minimum mean square error channel estimation model and a least square channel estimation model.

[0153] In an example embodiment, the fourth determination module 407 is further configured to:

[0154] The scene information is acquired, and the corresponding preset channel estimation model is determined according to the scene information.

[0155] As to the apparatus in the above-described embodiments, the specific manners in which various modules perform operations have been described in detail in the embodiments of the method, and thus will not be described here in detail.

[0156] Figure 5 is a block diagram of an electronic device 500 according to an example embodiment.

[0157] Referring to Figure 5 , the electronic device 500 can include one or more of the following components: a processing component 502, a memory 504, a power supply component 506, a multimedia component 508, an audio component 510, an input / output (I / O) interface 512, a sensor component 514, and a communication component 516.

[0158] The processing component 502 usually controls overall operations of the electronic device 500, such as operations associated with displaying, making phone calls, data communications, camera operations, and recording operations. The processing component 502 can include one or more processors 520 to execute instructions to complete all or part of steps of the methods described above. Further, the processing component 502 can include one or more modules to facilitate the interaction between the processing component 502 and other components. For example, the processing component 502 can include a multimedia module to facilitate the interaction between the multimedia component 508 and the processing component 502.

[0159] The memory 504 is configured to store various types of data to support the operation of the electronic device 500. Examples of such data include instructions for any application or method operating on the electronic device 500, contact data, phonebook data, messages, pictures, videos, and the like. The memory 504 can be implemented by any type of volatile or nonvolatile storage devices or a combination thereof, such as static random access memory, electrically erasable programmable read-only memory, erasable programmable read-only memory, programmable read-only memory, read-only memory, magnetic storage, flash memory, magnetic or optical disks.

[0160] The power supply component 506 supplies power for various components of the electronic device 500. The power supply component 506 can include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power for the electronic device 500.

[0161] The multimedia component 508 includes a screen providing an output interface between the electronic device 500 and a user. In some embodiments, the screen can include a liquid crystal display and a touch panel. If the screen includes the touch panel, the screen can be implemented as a touch screen to receive an input signal from a user. The touch panel includes one or more touch sensors to sense a touch, a slide, and a gesture on the touch panel. The touch sensor can not only sense a boundary of a touching or a sliding action, but also detect duration and pressure related to the touching or sliding action. In some embodiments, the multimedia component 508 includes a front camera and / or a back camera. The front camera and / or the back camera can receive external multimedia data when the electronic device 500 is in an operation mode, such as a photographing mode or a video mode. Each of the front camera and the back camera can be a fixed optical lens system or have a focal length and optical zoom capability.

[0162] The audio component 510 is configured to output and / or input an audio signal. For example, the audio component 510 includes a microphone configured to receive an external audio signal when the electronic device 500 is in an operation mode, such as a call mode, a recording mode, and a voice recognition mode. The received audio signal can be further stored in the memory 504 or transmitted via the communication component 516. In some embodiments, the audio component 510 also includes a speaker for outputting an audio signal.

[0163] The I / O interface 512 provides an interface between the processing component 502 and peripheral interface modules, which can be a keyboard, a click wheel, a button, and the like. The buttons can include, but are not limited to, a home button, a volume button, a start button, and a lock button.

[0164] The sensor component 514 includes one or more sensors for providing status assessments for various aspects of the electronic device 500. For example, the sensor component 514 can detect an open / closed position of the electronic device 500, relative positioning of components, such as a display and keypad of the electronic device 500, a change in position of the electronic device 500 or a component of the electronic device 500, the presence or absence of user contact with the electronic device 500, the orientation or acceleration / deceleration of the electronic device 500, and a temperature change of the electronic device 500. The sensor component 514 can include an optical sensor for detecting ambient light, a proximity sensor configured to detect the presence of nearby objects without any physical touch, a light sensor, such as a CMOS (Complementary Metal Oxide Semiconductor) or CCD (Charge Coupled Device) image sensor, for use in imaging applications, and / or a motion sensor, such as an accelerometer, a gyroscope, a magnetometer, a pressure sensor, or a temperature sensor.

[0165] The communication component 516 is configured to facilitate wired or wireless communication between the electronic device 500 and other devices. The electronic device 500 can access a wireless network based on a communication standard, such as WiFi, 2G, or 3G, or a combination thereof. In an example embodiment, the communication component 516 receives broadcast signals or broadcast-related information from an external broadcast management system via a broadcast channel. In an example embodiment, the communication component 516 can further include a Near Field Communication (NFC) module to facilitate short-range communication. For example, the NFC module can be implemented based on Radio Frequency Identification (RFID) technology, Infrared Data Association (IrDA) technology, Ultra-WideBand (UWB) technology, Bluetooth (BT) technology and other technologies.

[0166] In exemplary embodiments, the electronic device 500 can be implemented by one or more Application-Specific Integrated Circuits (ASICs), Digital Signal Processors (DSPs), Digital Signal Processing Devices (DSPDs), Programmable Logic Devices (PLDs), Field-Programmable Gate Arrays (FPGAs), controllers, micro-controllers, microprocessors, or other electronic elements for performing the above-described methods.

[0167] In exemplary embodiments, a non-transitory computer-readable storage medium including instructions, such as the memory 504 including instructions, is also provided, which can be executed by the processor 520 of the electronic device 500 to complete the above-described methods. For example, the non-transitory computer-readable storage medium can be a Read-Only Memory (ROM), a Random Access Memory (RAM), a Compact Disc Read-Only Memory (CD-ROM), a magnetic tape, a floppy disk, and an optical data storage device, etc.

[0168] A non-transitory computer-readable storage medium, when instructions in the storage medium are executed by a processor of an electronic device, enables the electronic device to perform a channel estimation method, the channel estimation method including any one of the above-described methods.

[0169] A computer program product including a computer program, which, when executed by a processor, implements a channel estimation method, the channel estimation method including any one of the above-described methods.

[0170] Other embodiments of the present application will be apparent to those skilled in the art from consideration of the specification and practice of the application disclosed herein. It is intended that the specification and examples be considered as exemplary only, with the true scope and spirit of the application being indicated by the following claims.

Claims

1. A channel estimation method, characterized in that, The method includes: The pilot signal matrix corresponding to the communication signal is determined based on the number of transmitting antennas and the pilot length corresponding to the target channel. The target channel is the channel corresponding to the first time slot and the first subcarrier. Based on the pilot signal matrix, the noise power of the communication signal, and the number of receiving antennas, construct the processing matrix of the target channel; Based on the correlation influence parameters and weight decision model, the time domain weight and frequency domain weight are determined. An initial value for iteration is generated based on the time domain weight, the second time slot and the first reference matrix corresponding to the first subcarrier, the frequency domain weight, the second reference matrix corresponding to the first time slot and the second subcarrier. The second time slot is used to indicate the previous time slot of the first time slot in the continuously distributed time domain resources, and the second subcarrier is used to indicate the previous subcarrier of the first subcarrier in the continuously distributed frequency resources. Based on the initial iteration value and the preset inversion iteration model, the matrix to be processed is subjected to iterative operation. During the iterative operation, the iteration step size is determined according to the step size decision model, and the preset inversion iteration model for the next iteration operation is updated according to the iteration step size. The step size decision model is a machine learning model that is pre-trained to determine the iteration step size based on the condition number of the matrix to be processed, the iteration residual, and the current iteration number. When the preset iteration termination condition is met, the inverse matrix of the matrix to be processed is obtained; The channel matrix of the target channel is determined based on the inverse of the matrix to be processed. The method further includes: Obtain the channel matrix of the first channel and the channel matrix of the second channel, wherein the first channel is the channel corresponding to the second time slot and the first subcarrier, and the second channel is the channel corresponding to the first time slot and the second subcarrier; The channel matrix of the first channel is used as the first reference matrix, and the channel matrix of the second channel is used as the second reference matrix.

2. The method according to claim 1, characterized in that, The determination of time-domain weights and frequency-domain weights based on the correlation influence parameters and weight decision model includes: The moving speed of the electronic device, the correlation between the second time slot and the first time slot, and the correlation between the first subcarrier and the second subcarrier are input into the weight decision model to obtain the time domain weight and the frequency domain weight. The weighted decision model is trained based on the reference moving speed of the electronic device, the correlation between reference time domains, the correlation between reference frequency domains, and the reference time domain weights and reference frequency domain weights.

3. The method according to claim 1, characterized in that, The step of generating initial values ​​for iteration based on the time-domain weights, the first reference matrix corresponding to the second time slot and the first subcarrier, the frequency-domain weights, and the second reference matrix corresponding to the first time slot and the second subcarrier includes: The first reference matrix is ​​encoded, its features are extracted, and then its features are decoded to obtain the first matrix reconstructed from the first reference matrix. The second reference matrix is ​​encoded, its features are extracted, and then its features are decoded to obtain the second matrix reconstructed from the second reference matrix. The initial values ​​for the iteration are generated based on the time-domain weights and the first matrix, and the frequency-domain weights and the second matrix.

4. The method according to claim 1, characterized in that, The preset inversion iterative model is used to indicate the correlation between the iteration result of the current iteration and the iteration result of the next iteration. The step of determining the iteration step size based on the step size decision model and updating the preset inversion iterative model for the next iteration operation based on the iteration step size includes: The condition number, iteration residual, and current iteration number of the matrix to be processed are input into the step size decision model to obtain the iteration step size; Based on the iteration step size, the iteration result of the current iteration, and the matrix to be processed, determine the preset inversion iteration model corresponding to the next iteration.

5. The method according to claim 1, characterized in that, The preset iteration termination condition is that the iteration residual is less than a first threshold and the convergence probability is greater than a second threshold.

6. The method according to claim 1, characterized in that, Determining the channel matrix of the target channel based on the inverse matrix of the matrix to be processed includes: The inverse of the matrix to be processed is input into a preset channel estimation model to obtain the channel matrix of the target channel. The preset channel estimation model is one of a linear minimum mean square error channel estimation model and a least squares channel estimation model.

7. The method according to claim 6, characterized in that, The method further includes: Obtain scene information, and determine the corresponding preset channel estimation model based on the scene information.

8. A channel estimation device, characterized in that, The device includes: The first determining module is configured to determine the pilot signal matrix corresponding to the communication signal based on the number of transmitting antennas and the pilot length corresponding to the communication signal transmitted through the target channel, wherein the target channel is the channel corresponding to the first time slot and the first subcarrier. The construction module is configured to construct the target channel's unprocessed matrix based on the pilot signal matrix, the noise power of the communication signal, and the number of receiving antennas; The second determining module is configured to determine the time domain weights and frequency domain weights based on the correlation influence parameters and the weight decision model. The generation module is configured to generate an initial value for iteration based on the time domain weight, the second time slot and the first reference matrix corresponding to the first subcarrier, the frequency domain weight, the second reference matrix corresponding to the first time slot and the second subcarrier, wherein the second time slot is used to indicate the previous time slot of the first time slot in the continuously distributed time domain resources, and the second subcarrier is used to indicate the previous subcarrier of the first subcarrier in the continuously distributed frequency resources. The iteration module is configured to perform iterative operations on the matrix to be processed based on the initial iteration value and a preset inversion iteration model. During the iterative operation, the iteration step size is determined according to the step size decision model, and the preset inversion iteration model for the next iteration operation is updated according to the iteration step size. The step size decision model is a machine learning model that is pre-trained to determine the iteration step size based on the condition number of the matrix to be processed, the iteration residual, and the current iteration number. The third determining module is configured to obtain the inverse matrix of the matrix to be processed when a preset iteration termination condition is met. The fourth determining module is configured to determine the channel matrix of the target channel based on the inverse matrix of the matrix to be processed; The device is also used for: Obtain the channel matrix of the first channel and the channel matrix of the second channel, wherein the first channel is the channel corresponding to the second time slot and the first subcarrier, and the second channel is the channel corresponding to the first time slot and the second subcarrier; The channel matrix of the first channel is used as the first reference matrix, and the channel matrix of the second channel is used as the second reference matrix.

9. An electronic device, characterized in that, include: processor; Memory used to store processor-executable instructions; The processor is configured to perform the method as described in any one of claims 1-7.

10. A non-transitory computer-readable storage medium, characterized in that, When the instructions in the storage medium are executed by the processor of the electronic device, the electronic device is able to perform the method as described in any one of claims 1-7.

11. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the method as described in any one of claims 1-7.

Images