Channel estimation method, device, electronic device and storage medium

Abstract

The present application relates to a channel estimation method, device, electronic device and storage medium, which are applied to the field of communication technology, wherein the method includes: obtaining the frequency domain channel estimation result of the pilot position channel; determining the noise power corresponding to the frequency domain channel estimation result; obtaining the frequency domain autocorrelation matrix and the time domain autocorrelation matrix corresponding to the pilot position channel; performing dimensionality reduction processing on the frequency domain autocorrelation matrix to obtain the reduced dimension matrix; determining the filter coefficient of the minimum mean square error criterion corresponding to the time domain autocorrelation matrix, the reduced dimension matrix and the noise power; and calculating the time and frequency domain channel estimation result of the pilot position channel according to the filter coefficient of the minimum mean square error criterion and the frequency domain channel estimation result. In order to solve the problem in the prior art that the channel estimation based on the minimum mean square error criterion is performed by using methods such as SVD (singular value decomposition), and after the filter coefficient is calculated, the channel at the RS position is filtered, the process has a large amount of calculation due to the high dimension.

Description

Technical Field

[0001] The present application relates to the field of communication technology, and in particular to a channel estimation method, device, electronic device and storage medium. Background Art

[0002] Currently, in OFDM (Orthogonal Frequency Division Multiplexing) systems, a carrier-assisted modulation channel estimation method is usually used, that is, a pilot signal is inserted into the transmitted data stream, the pilot is extracted at the receiving end, the channel response at the pilot position is obtained by calculation, and then the channel response at other positions without a pilot is estimated by interpolation. Summary of the invention

[0003] The inventors found that in an orthogonal frequency division multiplexing system, the LS (least squares channel) estimation result of the RS (pilot) sequence is filtered and denoised in the time and frequency domain using the MMSE (minimum mean-square error) criterion. This process will introduce a relatively large amount of computation. Therefore, it is necessary to perform channel estimation based on the minimum mean square error criterion.

[0004] In the related technology, methods such as SVD (singular value decomposition) are used to simplify the calculation process of MMSE. Since the matrix decomposition itself also has a certain complexity, and after the filter coefficients are calculated, the process of filtering the channel at the RS position has a high dimensionality and a large amount of calculation.

[0005] The present application provides a channel estimation method, device, electronic device and storage medium to solve the problem in the prior art that SVD (singular value decomposition) and other methods are used to simplify the MMSE-based channel estimation calculation process. Since the matrix decomposition itself also has a certain complexity, and after the filter coefficients are calculated, the process of filtering the channel at the RS position has a high dimensionality and a large amount of calculation.

[0006] In a first aspect, an embodiment of the present application provides a channel estimation method, including:

[0007] Obtaining a frequency domain channel estimation result of the pilot position channel;

[0008] Determining a noise power corresponding to the frequency domain channel estimation result;

[0009] Acquire a frequency domain autocorrelation matrix and a time domain autocorrelation matrix corresponding to the pilot position channel;

[0010] Performing dimensionality reduction processing on the frequency domain autocorrelation matrix to obtain a dimensionality reduction matrix;

[0011] Determine a filter coefficient of a minimum mean square error criterion corresponding to the time domain autocorrelation matrix, the dimension reduction matrix and the noise power;

[0012] The time-frequency domain channel estimation result of the pilot position channel is calculated based on the filter coefficient of the minimum mean square error criterion and the frequency-domain channel estimation result.

[0013] Optionally, performing dimensionality reduction processing on the frequency domain autocorrelation matrix to obtain a dimensionality reduction matrix includes:

[0014] Converting the frequency domain autocorrelation matrix into a frequency domain real number matrix, wherein the frequency domain real number matrix includes a real part matrix and an imaginary part matrix of the frequency domain autocorrelation matrix;

[0015] Performing a first transformation on the frequency domain real number matrix to obtain a first transformation result, wherein the matrix dimension in the first transformation result is half of the dimension of the frequency domain real number matrix;

[0016] Performing a second transformation on the first transformation result to obtain a second transformation result, wherein the second transformation result is a product of a triangular matrix, a diagonal matrix, and an orthogonal matrix of the triangular matrix;

[0017] Generating a sub-diagonal matrix satisfying a preset condition according to the diagonal matrix, wherein the dimension of the sub-diagonal matrix is ​​smaller than the dimension of the diagonal matrix;

[0018] Determine a sub-triangular matrix in the triangular matrix corresponding to the sub-diagonal matrix;

[0019] Determine an orthogonal matrix of a sub-triangular matrix corresponding to the sub-triangular matrix;

[0020] The sub-diagonal matrix, the sub-triangular matrix and the orthogonal matrix of the sub-triangular matrix are used as the dimension reduction matrix.

[0021] Optionally, performing a first transformation on the frequency domain real number matrix to obtain a first transformation result, wherein a matrix dimension in the first transformation result is half of a dimension of the frequency domain real number matrix, includes:

[0022] Multiplying both ends of the frequency domain real number matrix by a first coefficient to obtain a first intermediate matrix, where the first coefficient is a product of a first preset matrix and a transposed matrix of the first preset matrix;

[0023] The first intermediate matrix is ​​transformed to obtain a second intermediate matrix, where the second intermediate matrix is ​​the transposed matrix of the first preset matrix, the product of the third intermediate matrix and the first preset matrix, the third intermediate matrix is ​​a matrix whose main diagonal elements are the first intermediate sub-matrix and the second intermediate sub-matrix, and other elements are 0, and the dimensions of the first intermediate sub-matrix and the second intermediate sub-matrix are half the dimensions of the frequency domain real matrix.

[0024] Optionally, performing a second transformation on the first transformation result to obtain a second transformation result, where the second transformation result is the product of a triangular matrix, a diagonal matrix and an orthogonal matrix of the triangular matrix, includes:

[0025] Performing singular value decomposition on the first intermediate submatrix to obtain a first decomposition result;

[0026] Performing singular value decomposition on the second intermediate submatrix to obtain a second decomposition result;

[0027] Setting the elements in the first decomposition result at the positions of the first intermediate sub-matrix, and setting the elements in the second decomposition result at the positions of the second intermediate sub-matrix to obtain a fourth intermediate matrix;

[0028] The transposed matrix of the first preset matrix, the product of the fourth intermediate matrix and the first preset matrix are transformed to obtain the second transformation result, where the second transformation result is the product of the triangular matrix, the diagonal matrix and the orthogonal matrix of the triangular matrix.

[0029] Optionally, the first preset matrix is ​​constructed based on a first preset sub-matrix and a second preset sub-matrix, the first preset sub-matrix is ​​a unit matrix, and the second preset sub-matrix is ​​a matrix whose elements on the sub-diagonal are 1.

[0030] Optionally, determining a filter coefficient of a minimum mean square error criterion corresponding to the time domain autocorrelation matrix, the dimensionality reduction matrix, and the noise power includes:

[0031] Obtaining a Kronecker product matrix of the time domain autocorrelation matrix and the dimension reduction matrix;

[0032] According to the Kronecker product matrix and the noise power, a filter coefficient of the minimum mean square error criterion is calculated based on the minimum mean square error criterion.

[0033] Optionally, the dimension reduction matrix includes a sub-diagonal matrix, a sub-triangular matrix, and an orthogonal matrix of a sub-triangular matrix, and obtaining the Kronecker product matrix of the time domain autocorrelation matrix and the dimension reduction matrix includes:

[0034] Obtaining a sub-Kronecker product matrix of the time domain autocorrelation matrix and the sub-diagonal matrix;

[0035] The Kronecker product matrix is ​​obtained according to the sub-Kronecker product matrix, the sub-triangular matrix and an orthogonal matrix of the sub-triangular matrix.

[0036] Optionally, obtaining the Kronecker product matrix of the time domain autocorrelation matrix and the sub-diagonal matrix includes:

[0037] Extracting a set of non-zero elements in the sub-diagonal matrix;

[0038] Calculating the product of each element in the non-zero element set and the time domain autocorrelation matrix to obtain a product result;

[0039] The product result is set in the corresponding position of the sub-Kronecker product matrix, and the remaining positions in the sub-Kronecker product matrix are set to 0.

[0040] Optionally, generating a sub-diagonal matrix satisfying a preset condition according to the diagonal matrix, wherein a dimension of the sub-diagonal matrix is ​​smaller than a dimension of the diagonal matrix, includes:

[0041] Sort the non-zero elements in the diagonal matrix according to their numerical values ​​to obtain a sorting result;

[0042] The sorting results are accumulated in order according to the numerical values;

[0043] Each time a value is accumulated, it is determined whether the proportion of the accumulated result is greater than or equal to the preset value;

[0044] If not, continue to accumulate until the proportion of the accumulated result is greater than or equal to the preset value;

[0045] If so, the elements in the diagonal matrix corresponding to the accumulated results are used as elements in the sub-diagonal matrix to obtain the sub-diagonal matrix.

[0046] In a second aspect, an embodiment of the present application provides a channel estimation device, including:

[0047] A first acquisition module, used to acquire a frequency domain channel estimation result of the pilot position channel;

[0048] A first determination module, used to determine the noise power corresponding to the frequency domain channel estimation result;

[0049] A second acquisition module is used to acquire a frequency domain autocorrelation matrix and a time domain autocorrelation matrix corresponding to the pilot position channel;

[0050] A dimension reduction module, used for performing dimension reduction processing on the frequency domain autocorrelation matrix to obtain a dimension reduction matrix;

[0051] A second determination module is used to determine a filter coefficient of a minimum mean square error criterion corresponding to the time domain autocorrelation matrix, the dimension reduction matrix and the noise power;

[0052] The channel estimation module is used to calculate the time-frequency domain channel estimation result of the pilot position channel according to the filter coefficient of the minimum mean square error criterion and the frequency domain channel estimation result.

[0053] Optionally, the dimension reduction module is specifically used for:

[0054] Converting the frequency domain autocorrelation matrix into a frequency domain real number matrix, wherein the frequency domain real number matrix includes a real part matrix and an imaginary part matrix of the frequency domain autocorrelation matrix;

[0055] Performing a first transformation on the frequency domain real number matrix to obtain a first transformation result, wherein the matrix dimension in the first transformation result is half of the dimension of the frequency domain real number matrix;

[0056] Performing a second transformation on the first transformation result to obtain a second transformation result, wherein the second transformation result is a product of a triangular matrix, a diagonal matrix, and an orthogonal matrix of the triangular matrix;

[0057] Generating a sub-diagonal matrix satisfying a preset condition according to the diagonal matrix, wherein the dimension of the sub-diagonal matrix is ​​smaller than the dimension of the diagonal matrix;

[0058] Determine a sub-triangular matrix in the triangular matrix corresponding to the sub-diagonal matrix;

[0059] Determine an orthogonal matrix of a sub-triangular matrix corresponding to the sub-triangular matrix;

[0060] The sub-diagonal matrix, the sub-triangular matrix and the orthogonal matrix of the sub-triangular matrix are used as the dimension reduction matrix.

[0061] Optionally, the dimension reduction module is specifically used for:

[0062] Multiplying both ends of the frequency domain real number matrix by a first coefficient to obtain a first intermediate matrix, where the first coefficient is a product of a first preset matrix and a transposed matrix of the first preset matrix;

[0063] The first intermediate matrix is ​​transformed to obtain a second intermediate matrix, where the second intermediate matrix is ​​the transposed matrix of the first preset matrix, the product of the third intermediate matrix and the first preset matrix, the third intermediate matrix is ​​a matrix whose main diagonal elements are the first intermediate sub-matrix and the second intermediate sub-matrix, and other elements are 0, and the dimensions of the first intermediate sub-matrix and the second intermediate sub-matrix are half the dimensions of the frequency domain real matrix.

[0064] Optionally, the dimension reduction module is specifically used for:

[0065] Performing singular value decomposition on the first intermediate submatrix to obtain a first decomposition result;

[0066] Performing singular value decomposition on the second intermediate submatrix to obtain a second decomposition result;

[0067] Setting the elements in the first decomposition result at the positions of the first intermediate sub-matrix, and setting the elements in the second decomposition result at the positions of the second intermediate sub-matrix to obtain a fourth intermediate matrix;

[0068] The transposed matrix of the first preset matrix, the product of the fourth intermediate matrix and the first preset matrix are transformed to obtain the second transformation result, where the second transformation result is the product of the triangular matrix, the diagonal matrix and the orthogonal matrix of the triangular matrix.

[0069] Optionally, the first preset matrix is ​​constructed based on a first preset sub-matrix and a second preset sub-matrix, the first preset sub-matrix is ​​a unit matrix, and the second preset sub-matrix is ​​a matrix whose elements on the sub-diagonal are 1.

[0070] Optionally, the second determining module is specifically configured to:

[0071] Obtaining a Kronecker product matrix of the time domain autocorrelation matrix and the dimension reduction matrix;

[0072] According to the Kronecker product matrix and the noise power, a filter coefficient of the minimum mean square error criterion is calculated based on the minimum mean square error criterion.

[0073] Optionally, the dimension reduction matrix includes a sub-diagonal matrix, a sub-triangular matrix and an orthogonal matrix of a sub-triangular matrix, and the second determining module is specifically used to:

[0074] Obtaining a sub-Kronecker product matrix of the time domain autocorrelation matrix and the sub-diagonal matrix;

[0075] The Kronecker product matrix is ​​obtained according to the sub-Kronecker product matrix, the sub-triangular matrix and an orthogonal matrix of the sub-triangular matrix.

[0076] Optionally, the second determining module is specifically configured to:

[0077] Extracting a set of non-zero elements in the sub-diagonal matrix;

[0078] Calculating the product of each element in the non-zero element set and the time domain autocorrelation matrix to obtain a product result;

[0079] The product result is set in the corresponding position of the sub-Kronecker product matrix, and the remaining positions in the sub-Kronecker product matrix are set to 0.

[0080] Optionally, the dimension reduction module is specifically used to:

[0081] Sort the non-zero elements in the diagonal matrix according to their numerical values ​​to obtain a sorting result;

[0082] The sorting results are accumulated in order according to the numerical values;

[0083] Each time a value is accumulated, it is determined whether the proportion of the accumulated result is greater than or equal to the preset value;

[0084] If not, continue to accumulate until the proportion of the accumulated result is greater than or equal to the preset value;

[0085] If so, the elements in the diagonal matrix corresponding to the accumulated results are used as elements in the sub-diagonal matrix to obtain the sub-diagonal matrix.

[0086] In a third aspect, an embodiment of the present application provides an electronic device, comprising: a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other via the communication bus;

[0087] The memory is used to store computer programs;

[0088] The processor is used to execute the program stored in the memory to implement the channel estimation method described in the first aspect.

[0089] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the channel estimation method described in the first aspect.

[0090] The above technical solution provided by the embodiment of the present application has the following advantages over the prior art: the method provided by the embodiment of the present application obtains the frequency domain channel estimation result of the pilot position channel; determines the noise power corresponding to the frequency domain channel estimation result; obtains the frequency domain autocorrelation matrix and the time domain autocorrelation matrix corresponding to the pilot position channel; performs dimensionality reduction processing on the frequency domain autocorrelation matrix to obtain a reduced dimension matrix; determines the filter coefficient of the minimum mean square error criterion corresponding to the time domain autocorrelation matrix, the reduced dimension matrix and the noise power; calculates the time-frequency domain channel estimation result of the pilot position channel based on the filter coefficient of the minimum mean square error criterion and the frequency domain channel estimation result. In this way, the frequency domain autocorrelation matrix is ​​reduced in dimension, which reduces the amount of calculation. BRIEF DESCRIPTION OF THE DRAWINGS

[0091] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0092] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0093] Figure 1 A flow chart of a channel estimation method provided in one embodiment of the present application;

[0094] Figure 2 A flow chart of a channel estimation method provided by another embodiment of the present application;

[0095] Figure 3 A flow chart of a channel estimation method provided by another embodiment of the present application;

[0096] Figure 4 A flow chart of a channel estimation method provided by another embodiment of the present application;

[0097] Figure 5 A flow chart of a channel estimation method provided by another embodiment of the present application;

[0098] Figure 6 A flow chart of a channel estimation method provided by another embodiment of the present application;

[0099] Figure 7 A flow chart of a channel estimation method provided by another embodiment of the present application;

[0100] Figure 8 A flow chart of a channel estimation method provided by another embodiment of the present application;

[0101] Fig. 9 A structural diagram of a channel estimation device provided in one embodiment of the present application;

[0102] Fig.10 A structural diagram of an electronic device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0103] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0104] In OFDM systems, effective channel estimation is required to improve transmission rate and quality. Currently, considering the independence of channels in the time domain and frequency domain in OFDM systems, channel estimation can be decomposed into two parts: channel estimation in the frequency domain and channel estimation in the time domain.

[0105] Regardless of whether it is channel estimation in the frequency domain or channel estimation in the time domain, it includes two parts: one is the estimation of the reference signal channel, and the other is the interpolation of the data channel.

[0106] When estimating a reference signal channel, the following channel model may be used:

[0107] Y P =X P H+W (1)

[0108] Where H is the channel response, X P is the signal sent by the transmitter at the pilot position, Y P is the pilot signal extracted by the receiving end from the received data stream, and W is the observation noise.

[0109] The LS (least square channel) estimation result of the RS (pilot) sequence is filtered and denoised in the time-frequency domain using the MMSE (minimum mean square error) criterion. This process will introduce a relatively large amount of computation, so it is necessary to simplify the MMSE criterion calculation process. The SVD (singular value decomposition) and other methods are used to simplify the MMSE calculation process. Since the matrix decomposition itself also has a certain complexity, and after the filter coefficients are calculated, the process of filtering the channel at the RS position has a high dimensionality and a large amount of computation.

[0110] To solve the above problems, an embodiment of the present application provides a channel estimation method, which can be applied to any form of electronic equipment, such as a terminal and a server. Figure 1 As shown, the channel estimation method comprises:

[0111] Step 101: Obtain a frequency domain channel estimation result of a pilot position channel.

[0112] In some embodiments, the frequency domain channel estimation result of the pilot position channel obtained may be a least squares (LS, Least-Square) channel estimation result.

[0113] Step 102: Determine the noise power corresponding to the frequency domain channel estimation result.

[0114] In this embodiment, the noise power can be estimated by a power delay profile (PDP).

[0115] Step 103: Obtain a frequency domain autocorrelation matrix and a time domain autocorrelation matrix corresponding to the pilot position channel.

[0116] In some embodiments, the autocorrelation matrix is ​​a matrix in which the original matrix and the correlation matrix are the same matrix.

[0117] Specifically, for the autocorrelation matrix, the element in the i-th row and j-th column of the correlation matrix is ​​the correlation coefficient between the i-th column and the j-th column of the original matrix.

[0118] Moreover, the autocorrelation matrix is ​​a conjugate symmetric positive definite Toeplitz matrix.

[0119] In this embodiment, the frequency domain autocorrelation matrix is ​​represented as R F , the time domain autocorrelation matrix is ​​expressed as R T .

[0120] Step 104: Perform dimensionality reduction processing on the frequency domain autocorrelation matrix to obtain a dimensionality reduction matrix.

[0121] In some embodiments, by performing dimensionality reduction processing on the frequency domain autocorrelation matrix, the amount of calculation is reduced when subsequently calculating the channel estimation value.

[0122] Specifically, the frequency domain autocorrelation matrix is ​​subjected to dimensionality reduction processing to obtain a dimensionality reduction matrix, including:

[0123] Step 201: Convert the frequency domain autocorrelation matrix into a frequency domain real number matrix, where the frequency domain real number matrix includes a real part matrix and an imaginary part matrix of the frequency domain autocorrelation matrix.

[0124] In some embodiments, the frequency domain autocorrelation matrix is ​​a complex matrix, and therefore, it can be represented by real and imaginary parts:

[0125]

[0126] in, Represents R F The real part of the matrix and Represents R F The imaginary part of composes the matrix.

[0127] In this embodiment, the frequency domain real number matrix is ​​defined as

[0128] Furthermore, due to R F The eigenvalues ​​of are the same as those of the frequency domain real matrix, and R F The eigenvector of can also be transformed from the eigenvector of the frequency domain real matrix, so R F The SVD decomposition of is transformed into the problem of solving the eigenvalues ​​and eigenvectors of the real matrix in the frequency domain.

[0129] Step 202: Perform a first transformation on the frequency domain real number matrix to obtain a first transformation result, in which the matrix dimension is half of the dimension of the frequency domain real number matrix.

[0130] Specifically, step 202 includes:

[0131] Step 301: multiply both ends of a frequency domain real number matrix by a first coefficient to obtain a first intermediate matrix, where the first coefficient is the product of a first preset matrix and a transposed matrix of the first preset matrix.

[0132] In some embodiments, the first coefficient is defined as K·K T , after multiplying both ends of the frequency domain real matrix by the first coefficient, the following first intermediate matrix is ​​obtained:

[0133]

[0134] The first preset matrix is ​​constructed based on the first preset sub-matrix and the second preset sub-matrix, the first preset sub-matrix is ​​a unit matrix, and the second preset sub-matrix is ​​a matrix whose elements on the secondary diagonal are 1.

[0135] In this embodiment, a first preset matrix is ​​defined Wherein, I is the first preset sub-matrix, and J is the second preset sub-matrix.

[0136] Step 302: transform the first intermediate matrix to obtain a second intermediate matrix, which is the transposed matrix of the first preset matrix, the product of the third intermediate matrix and the first preset matrix, the third intermediate matrix is ​​a matrix whose main diagonal elements are the first intermediate sub-matrix and the second intermediate sub-matrix, and other elements are 0, and the dimensions of the first intermediate sub-matrix and the second intermediate sub-matrix are half of the dimensions of the frequency domain real matrix.

[0137] In some embodiments, the first intermediate matrix is ​​transformed, and the obtained second intermediate matrix is ​​the product of the transposed matrix of the first preset matrix, the third intermediate matrix and the first preset matrix. Therefore, when transforming the first intermediate matrix, it is only necessary to calculate The middle part is fine.

[0138] Substituting the first preset matrix K into the above middle part, we can get:

[0139]

[0140] Because, R F is a conjugate symmetric matrix, so we have And according to R F is a characteristic of the Toeplitz matrix. Obviously and

[0141] therefore:

[0142]

[0143] in, This is the third intermediate matrix. is the first intermediate submatrix, is the second intermediate submatrix; the dimensions of the first intermediate submatrix and the second intermediate submatrix are half of the dimensions of the frequency domain real number matrix.

[0144] Step 203: Perform a second transformation on the first transformation result to obtain a second transformation result, where the second transformation result is the product of an orthogonal matrix of a triangular matrix, a diagonal matrix, and a triangular matrix.

[0145] Based on the above-mentioned related embodiments, the first transformation result is the product of the transposed matrix of the first preset matrix, the third intermediate matrix and the first preset matrix, that is:

[0146]

[0147] After obtaining the first transformation result, a second transformation needs to be performed to obtain a second transformation result.

[0148] Specifically, step 203 includes:

[0149] Step 401: Perform singular value decomposition on a first intermediate submatrix to obtain a first decomposition result.

[0150] In some embodiments, The SVD decomposition is the first decomposition result The process of performing SVD decomposition on the matrix may refer to related technologies and will not be described in detail here.

[0151] Step 402: Perform singular value decomposition on the second intermediate submatrix to obtain a second decomposition result.

[0152] In some embodiments, The SVD decomposition of the second decomposition result is

[0153] Step 403: Set the elements in the first decomposition result at the positions of the first intermediate sub-matrix, and set the elements in the second decomposition result at the positions of the second intermediate sub-matrix to obtain a fourth intermediate matrix.

[0154] Based on the above-mentioned related embodiments, after obtaining the first decomposition matrix, substituting it back into the second intermediate matrix, the fourth matrix is ​​obtained.

[0155] Step 404: transform the transposed matrix of the first preset matrix, the product of the fourth intermediate matrix and the first preset matrix to obtain a second transformation result, where the second transformation result is the product of a triangular matrix, a diagonal matrix and an orthogonal matrix of a triangular matrix.

[0156] Based on the above-mentioned related embodiments, after substituting the obtained fourth matrix back into the first matrix, the following is obtained:

[0157]

[0158] After obtaining the above transformation result, due to R F The eigenvectors of have undergone a series of transformations, and the obtained eigenvectors cannot guarantee orthogonality. Therefore, the eigenvectors need to be further orthogonalized.

[0159] In the above results, due to and is obtained by SVD decomposition of the third sub-intermediate matrix, so and are all diagonal matrices, so we get is also a diagonal matrix, so, That is R F The diagonal array of .

[0160] Step 204: Generate a sub-diagonal matrix that meets preset conditions based on the diagonal matrix, where the dimension of the sub-diagonal matrix is ​​smaller than the dimension of the diagonal matrix.

[0161] In some embodiments, after transforming the frequency domain real number matrix to obtain a first transformation result, the problem of finding eigenvalues ​​and eigenvectors of the 2n×2n dimensional frequency domain real number matrix can be converted into two n×n third sub-intermediate matrices to solve the eigenvalue and eigenvector problems, thereby reducing the amount of calculation.

[0162] Specifically, generating a sub-diagonal matrix that meets a preset condition according to the diagonal matrix, where the dimension of the sub-diagonal matrix is ​​smaller than the dimension of the diagonal matrix, includes:

[0163] Step 501: sort the non-zero elements in the diagonal matrix according to their numerical values ​​to obtain a sorting result.

[0164] Step 502: Add up the sorting results in order of numerical values.

[0165] Step 503: Every time a value is accumulated, determine whether the proportion of the accumulated result is greater than or equal to a preset value.

[0166] The proportion of the cumulative result is the proportion of the cumulative result to the cumulative results of all non-zero elements.

[0167] Step 504: If not, continue to accumulate until the proportion of the accumulated result is greater than or equal to the preset value.

[0168] Step 505: If yes, use the elements in the diagonal matrix corresponding to the accumulation result as the elements in the sub-diagonal matrix to obtain the sub-diagonal matrix.

[0169] Based on the above-mentioned related embodiments, the calculated eigenvalues ​​are the diagonal matrix VF. Since most of the energy is concentrated on the largest eigenvalues, according to the idea of ​​PCA dimensionality reduction, only the largest eigenvalues ​​and their corresponding eigenvectors need to be retained to preserve the basically complete RF matrix, thereby reducing the dimensionality of the diagonal matrix.

[0170] Therefore, in this embodiment, the preset value can be set according to actual conditions and is not limited here. For example, the preset value can be 99.99%. The elements in the diagonal matrix are accumulated from large to small. Each time an eigenvalue is accumulated, the proportion of the accumulated result is compared with the preset value until the proportion of the accumulated result is greater than the preset value. The remaining eigenvalues ​​are discarded, thereby reducing the dimension of the diagonal matrix and greatly reducing the amount of calculation.

[0171] Step 205: Determine the sub-triangular matrix in the triangular matrix corresponding to the sub-diagonal matrix.

[0172] Based on the above-mentioned related embodiments, since the diagonal matrix is ​​the eigenvalue of the frequency domain real number matrix, and the triangular matrix is ​​the eigenvector of the frequency domain real number matrix, after determining the sub-diagonal matrix, only the eigenvectors corresponding to the eigenvalues ​​can be retained in the triangular matrix, thereby reducing the dimension of the triangular matrix and obtaining a sub-triangular matrix.

[0173] Step 206: Determine the orthogonal matrix of the sub-triangular matrix corresponding to the sub-triangular matrix.

[0174] Based on the above-mentioned related embodiments, the dimension of the triangular matrix is ​​reduced to obtain a sub-triangular matrix. Since the orthogonal matrix of the triangular matrix corresponds to the triangular matrix, after the sub-triangular matrix is ​​determined, the orthogonal matrix of the corresponding sub-triangular matrix can also be determined.

[0175] Step 207: Use the sub-diagonal matrix, the sub-triangular matrix and the orthogonal matrix of the sub-triangular matrix as a dimension reduction matrix.

[0176] Based on the above-mentioned related embodiments, after obtaining the sub-diagonal matrix, the sub-triangular matrix and the orthogonal matrix of the sub-triangular matrix, the product of the three is used as a dimension reduction matrix to calculate the filter coefficients of the minimum mean square error criterion.

[0177] Step 105: Determine the filter coefficient of the minimum mean square error criterion corresponding to the time domain autocorrelation matrix, the dimension reduction matrix and the noise power.

[0178] Specifically, determining the filtering coefficient of the minimum mean square error criterion corresponding to the time domain autocorrelation matrix, the dimension reduction matrix and the noise power includes:

[0179] Step 601: Obtain a Kronecker product matrix of a time domain autocorrelation matrix and a dimension reduction matrix.

[0180] Step 602: According to the Kronecker product matrix and the noise power, based on the minimum mean square error criterion, a filter coefficient of the minimum mean square error criterion is calculated.

[0181] In some embodiments, the filter coefficient of the minimum mean square error criterion may be determined according to the minimum mean square error (MMSE) criterion, and the determined filter coefficient of the minimum mean square error criterion may be:

[0182] W=R(R+σ 2 I) -1 ;

[0183] Among them, W is the filter coefficient of the minimum mean error estimation criterion, R is the time-frequency domain autocorrelation matrix, σ 2 is the noise power and I is the unit matrix.

[0184] Among them, the time-frequency domain autocorrelation matrix is ​​the Kronecker product of the time domain autocorrelation matrix and the frequency domain autocorrelation matrix After reducing the dimension of the frequency domain autocorrelation matrix to obtain the reduced dimension matrix, the Kronecker product matrix of the reduced dimension matrix and the time domain autocorrelation matrix can be calculated, and then the filtering coefficient of the minimum mean square error criterion can be calculated.

[0185] Based on the above-mentioned related embodiments, the dimension reduction matrix includes a sub-diagonal matrix, a sub-triangular matrix and an orthogonal matrix of a sub-triangular matrix. The dimension reduction matrix is ​​substituted into the filter coefficient formula of the above-mentioned minimum mean square error estimation criterion, and further simplified to obtain the filter coefficient of the minimum mean square error criterion:

[0186]

[0187] Specifically, the Kronecker product matrix of the time domain autocorrelation matrix and the dimension reduction matrix is ​​obtained, including:

[0188] Step 701: Obtain a sub-Kronecker product matrix of a time domain autocorrelation matrix and a sub-diagonal matrix.

[0189] Step 702: Obtain a Kronecker product matrix according to the sub-Kronecker product matrix, the sub-triangular matrix and the orthogonal matrix of the sub-triangular matrix.

[0190] In some embodiments, according to the above formula, it can be known that when calculating the filter coefficients of the minimum mean square error criterion, it is only necessary to obtain the sub-Kronecker product matrix of the sub-diagonal matrix in the dimensionality reduction matrix and the sub-diagonal matrix of the time domain autocorrelation matrix, the first sub-Kronecker product matrix of the unit matrix I and the triangular matrix, and the second sub-Kronecker product matrix of the unit matrix I and the orthogonal matrix of the triangular matrix, so as to calculate the filter coefficients of the minimum mean square error criterion.

[0191] Specifically, the sub-Kronecker product matrix of the time domain autocorrelation matrix and the sub-diagonal matrix is ​​obtained, including:

[0192] Step 801: Extract a set of non-zero elements in a sub-diagonal matrix.

[0193] Step 802: Calculate the product of each element in the non-zero element set and the time domain autocorrelation matrix to obtain a product result.

[0194] Step 803: Set the product result at the corresponding position of the sub-Kronecker product matrix, and set the remaining positions in the sub-Kronecker product matrix to 0.

[0195] In some embodiments, the filter coefficients of the minimum mean square error criterion are Let matrix RV eff . Since V F is a diagonal matrix, Most of the elements of are 0, so here we can Decompose and invert to reduce the amount of calculation.

[0196] If V F It is an N-dimensional diagonal matrix, and N matrices RV are calculated. i ,0≤i≤N-1, the calculation method is as follows:

[0197]

[0198] The Cholesky algorithm can be used to find the inverse. Then, the RV is recovered from these N matrices according to the following rules: eff :

[0199]

[0200] RV eff =0, else

[0201] In this embodiment, since only R T With V F The non-zero elements in , therefore, reduce the amount of computation required to calculate the filter coefficients for the minimum mean square error criterion.

[0202] Step 106: Calculate the time-frequency domain channel estimation result of the pilot position channel according to the filter coefficient of the minimum mean square error criterion and the frequency domain channel estimation result.

[0203] The formula that can be used to calculate the time-frequency domain channel estimation result based on the minimum average error estimation coefficient and the frequency domain channel estimation result is:

[0204]

[0205] in, is the minimum mean square error estimation result.

[0206] Based on the above related embodiments, the minimum average error estimation coefficient is substituted into the above formula to obtain:

[0207]

[0208] In the above embodiment, U in the above formula has been F , and V F Dimensionality reduction was performed and The amount of calculation is reduced, so that in the present application, when calculating the time-frequency domain channel estimation results, the amount of filtering calculation is greatly reduced.

[0209] Assume that the original R F It is M F ×M F The dimensional autocorrelation matrix is ​​taken as N after dimensionality reduction.

[0210] In the estimation formula, V F From M F ×M F Reduced to an N×N diagonal matrix. F From M F ×M F Downgrade to M F ×N. If the dimension of the time domain autocorrelation matrix is ​​M T ×M T , then the dimension of the original time-frequency domain filter coefficient W is M T ·M F ×M T ·M F , now it can be decomposed into two filters, W1 and W2.

[0211]

[0212] W1 is M T N×M T ·M F dimensional coefficients, that is, for each time domain OFDM symbol, do M T ·M F Order filtering, but only needs to calculate the values ​​of N points.

[0213] W2 is M T ·M F ×M T N-dimensional coefficients, that is, for each time-domain OFDM symbol, do M T N-order filtering, W F The value of each point needs to be calculated.

[0214] The output result is the accurate result of channel estimation under the MMSE rule.

[0215] Based on the same concept, a channel estimation device is provided in the embodiment of the present application. The specific implementation of the device can refer to the description of the method embodiment part, and the repeated parts will not be repeated. Fig. 9 As shown, the device mainly includes:

[0216] A first acquisition module 901 is used to acquire a frequency domain channel estimation result of a pilot position channel;

[0217] A first determination module 902, configured to determine a noise power corresponding to a frequency domain channel estimation result;

[0218] The second acquisition module 903 is used to acquire a frequency domain autocorrelation matrix and a time domain autocorrelation matrix corresponding to the pilot position channel;

[0219] A dimension reduction module 904 is used to perform dimension reduction processing on the frequency domain autocorrelation matrix to obtain a dimension reduction matrix;

[0220] A second determination module 905 is used to determine a filter coefficient of a minimum mean square error criterion corresponding to the time domain autocorrelation matrix, the dimension reduction matrix and the noise power;

[0221] The channel estimation module 906 is used to calculate the time-frequency domain channel estimation result of the pilot position channel according to the filter coefficient of the minimum mean square error criterion and the frequency domain channel estimation result.

[0222] Based on the same concept, an electronic device is provided in an embodiment of the present application, such as Fig.10 As shown, the electronic device mainly includes: a processor 1001, a communication interface 1002, a memory 1003 and a communication bus 1004, wherein the processor 1001, the communication interface 1002 and the memory 1003 communicate with each other through the communication bus 1004. The memory 1003 stores a program that can be executed by the processor 1001, and the processor 1001 executes the program stored in the memory 1003 to implement the following steps:

[0223] Obtaining a frequency domain channel estimation result of a pilot position channel;

[0224] Determining a noise power corresponding to a frequency domain channel estimation result;

[0225] Obtaining a frequency domain autocorrelation matrix and a time domain autocorrelation matrix corresponding to the pilot position channel;

[0226] Perform dimensionality reduction processing on the frequency domain autocorrelation matrix to obtain a dimensionality reduction matrix;

[0227] Determine the filter coefficients of the minimum mean square error criterion corresponding to the time domain autocorrelation matrix, the dimension reduction matrix and the noise power;

[0228] According to the filter coefficient of the minimum mean square error criterion and the frequency domain channel estimation result, the time-frequency domain channel estimation result of the pilot position channel is calculated.

[0229] The communication bus 1004 mentioned in the above electronic device can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. The communication bus 1004 can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Fig.10 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0230] The communication interface 1002 is used for communication between the above electronic device and other devices.

[0231] The memory 1003 may include a random access memory (RAM) or a non-volatile memory, such as at least one disk memory. Alternatively, the memory may also be at least one storage device located away from the processor 1001.

[0232] The above-mentioned processor 1001 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc., and can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components.

[0233] In another embodiment of the present application, a computer-readable storage medium is provided, in which a computer program is stored. When the computer program is executed on a computer, the computer executes the channel estimation method described in the above embodiment.

[0234] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer instruction is loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network or other programmable device. The computer instruction can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instruction is transmitted from a website site, a computer, a server or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, microwave, etc.) mode to another website site, computer, server or data center. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server, a data center, etc. that contains one or more available media integration. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape, etc.), an optical medium (e.g., a DVD) or a semiconductor medium (e.g., a solid-state hard disk), etc.

[0235] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0236] The foregoing is merely a specific embodiment of the present invention, which enables those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A channel estimation method, characterized in that: include: Obtaining a frequency domain channel estimation result of a pilot position channel; Determining a noise power corresponding to the frequency domain channel estimation result; Acquire a frequency domain autocorrelation matrix and a time domain autocorrelation matrix corresponding to the pilot position channel; Performing dimensionality reduction processing on the frequency domain autocorrelation matrix to obtain a dimensionality reduction matrix; Determine a filter coefficient of a minimum mean square error criterion corresponding to the time domain autocorrelation matrix, the dimension reduction matrix and the noise power; Calculate the time-frequency domain channel estimation result of the pilot position channel according to the filter coefficient of the minimum mean square error criterion and the frequency domain channel estimation result; The step of performing dimensionality reduction processing on the frequency domain autocorrelation matrix to obtain a dimensionality reduction matrix comprises: Converting the frequency domain autocorrelation matrix into a frequency domain real number matrix, wherein the frequency domain real number matrix includes a real part matrix and an imaginary part matrix of the frequency domain autocorrelation matrix; Performing a first transformation on the frequency domain real number matrix to obtain a first transformation result, wherein the matrix dimension in the first transformation result is half of the dimension of the frequency domain real number matrix; Performing a second transformation on the first transformation result to obtain a second transformation result, wherein the second transformation result is a product of a triangular matrix, a diagonal matrix, and an orthogonal matrix of the triangular matrix; Generating a sub-diagonal matrix satisfying a preset condition according to the diagonal matrix, wherein the dimension of the sub-diagonal matrix is ​​smaller than the dimension of the diagonal matrix; Determine a sub-triangular matrix in the triangular matrix corresponding to the sub-diagonal matrix; Determine an orthogonal matrix of a sub-triangular matrix corresponding to the sub-triangular matrix; The sub-diagonal matrix, the sub-triangular matrix and the orthogonal matrix of the sub-triangular matrix are used as the dimension reduction matrix.

2. The channel estimation method according to claim 1, characterized in that The performing a first transformation on the frequency domain real number matrix to obtain a first transformation result, wherein the matrix dimension in the first transformation result is half of the dimension of the frequency domain real number matrix, comprises: Multiplying both ends of the frequency domain real number matrix by a first coefficient to obtain a first intermediate matrix, where the first coefficient is a product of a first preset matrix and a transposed matrix of the first preset matrix; The first intermediate matrix is ​​transformed to obtain a second intermediate matrix, where the second intermediate matrix is ​​the transposed matrix of the first preset matrix, the product of the third intermediate matrix and the first preset matrix, the third intermediate matrix is ​​a matrix whose main diagonal elements are the first intermediate sub-matrix and the second intermediate sub-matrix, and other elements are 0, and the dimensions of the first intermediate sub-matrix and the second intermediate sub-matrix are half the dimensions of the frequency domain real matrix.

3. The channel estimation method according to claim 2, characterized in that: The performing a second transformation on the first transformation result to obtain a second transformation result, wherein the second transformation result is the product of a triangular matrix, a diagonal matrix and an orthogonal matrix of the triangular matrix, comprises: Performing singular value decomposition on the first intermediate submatrix to obtain a first decomposition result; Performing singular value decomposition on the second intermediate submatrix to obtain a second decomposition result; Setting the elements in the first decomposition result at the positions of the first intermediate sub-matrix, and setting the elements in the second decomposition result at the positions of the second intermediate sub-matrix to obtain a fourth intermediate matrix; The transposed matrix of the first preset matrix, the product of the fourth intermediate matrix and the first preset matrix are transformed to obtain the second transformation result, where the second transformation result is the product of the triangular matrix, the diagonal matrix and the orthogonal matrix of the triangular matrix.

4. The channel estimation method according to claim 2, characterized in that: The first preset matrix is ​​constructed according to a first preset sub-matrix and a second preset sub-matrix, the first preset sub-matrix is ​​a unit matrix, and the second preset sub-matrix is ​​a matrix whose elements on the sub-diagonal are 1.

5. The channel estimation method according to claim 1, characterized in that: The determining of the filtering coefficient of the minimum mean square error criterion corresponding to the time domain autocorrelation matrix, the dimension reduction matrix and the noise power comprises: Obtaining a Kronecker product matrix of the time domain autocorrelation matrix and the dimension reduction matrix; According to the Kronecker product matrix and the noise power, a filter coefficient of the minimum mean square error criterion is calculated based on the minimum mean square error criterion.

6. The channel estimation method according to claim 5, characterized in that: The dimension reduction matrix includes a sub-diagonal matrix, a sub-triangular matrix and an orthogonal matrix of a sub-triangular matrix, and the step of obtaining the Kronecker product matrix of the time domain autocorrelation matrix and the dimension reduction matrix includes: Obtaining a sub-Kronecker product matrix of the time domain autocorrelation matrix and the sub-diagonal matrix; The Kronecker product matrix is ​​obtained according to the sub-Kronecker product matrix, the sub-triangular matrix and an orthogonal matrix of the sub-triangular matrix.

7. The channel estimation method according to claim 6, characterized in that: The step of obtaining a sub-Kronecker product matrix of the time domain autocorrelation matrix and the sub-diagonal matrix comprises: Extracting a set of non-zero elements in the sub-diagonal matrix; Calculating the product of each element in the non-zero element set and the time domain autocorrelation matrix to obtain a product result; The product result is set in the corresponding position of the sub-Kronecker product matrix, and the remaining positions in the sub-Kronecker product matrix are set to 0.

8. The channel estimation method according to claim 1, characterized in that: generating a sub-diagonal matrix satisfying a preset condition according to the diagonal matrix, wherein the dimension of the sub-diagonal matrix is ​​smaller than the dimension of the diagonal matrix, include: Sort the non-zero elements in the diagonal matrix according to their numerical values ​​to obtain a sorting result; The sorting results are accumulated in order according to the numerical values; Each time a value is accumulated, it is determined whether the proportion of the accumulated result is greater than or equal to the preset value; If not, continue to accumulate until the proportion of the accumulated result is greater than or equal to the preset value; If so, the elements in the diagonal matrix corresponding to the accumulated results are used as elements in the sub-diagonal matrix to obtain the sub-diagonal matrix.

9. A channel estimation device, characterized in that: include: A first acquisition module, used to acquire a frequency domain channel estimation result of a pilot position channel; A first determination module, used to determine the noise power corresponding to the frequency domain channel estimation result; A second acquisition module is used to acquire a frequency domain autocorrelation matrix and a time domain autocorrelation matrix corresponding to the pilot position channel; A dimension reduction module, used for performing dimension reduction processing on the frequency domain autocorrelation matrix to obtain a dimension reduction matrix; The step of performing dimensionality reduction processing on the frequency domain autocorrelation matrix to obtain a dimensionality reduction matrix comprises: Converting the frequency domain autocorrelation matrix into a frequency domain real number matrix, wherein the frequency domain real number matrix includes a real part matrix and an imaginary part matrix of the frequency domain autocorrelation matrix; Performing a first transformation on the frequency domain real number matrix to obtain a first transformation result, wherein the matrix dimension in the first transformation result is half of the dimension of the frequency domain real number matrix; Performing a second transformation on the first transformation result to obtain a second transformation result, wherein the second transformation result is a product of a triangular matrix, a diagonal matrix, and an orthogonal matrix of the triangular matrix; Generating a sub-diagonal matrix satisfying a preset condition according to the diagonal matrix, wherein the dimension of the sub-diagonal matrix is ​​smaller than the dimension of the diagonal matrix; Determine a sub-triangular matrix in the triangular matrix corresponding to the sub-diagonal matrix; Determine an orthogonal matrix of a sub-triangular matrix corresponding to the sub-triangular matrix; Using the sub-diagonal matrix, the sub-triangular matrix and the orthogonal matrix of the sub-triangular matrix as the dimension reduction matrix; A second determination module is used to determine a filter coefficient of a minimum mean square error criterion corresponding to the time domain autocorrelation matrix, the dimension reduction matrix and the noise power; The channel estimation module is used to calculate the time-frequency domain channel estimation result of the pilot position channel according to the filter coefficient of the minimum mean square error criterion and the frequency domain channel estimation result.

10. An electronic device, characterized in that: include: A processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus; The memory is used to store computer programs; The processor is used to execute the program stored in the memory to implement the channel estimation method according to any one of claims 1 to 8.

11. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the channel estimation method according to any one of claims 1 to 8 is implemented.

Images