Space-frequency anti-interference method based on adaptive null broadening

Abstract

The invention belongs to the field of satellite navigation, and particularly relates to a space-frequency anti-interference method based on self-adaptive null broadening, which comprises the following steps: converting a digital intermediate frequency signal into a frequency domain signal through windowing and discrete Fourier transform, and calculating a covariance matrix in a conventional operation scene; the Doppler change rate is adopted to identify a current motion scene, so that whether matrix reconstruction is carried out or not is judged, and an adaptive null broadening function is realized; and performing spatial filtering on each narrow sub-band, and transforming a frequency domain signal back to a time domain signal by using inverse discrete Fourier transform and inverse window to obtain anti-interference data. According to the method, the Doppler change rate is innovatively introduced into null broadening for self-adaptive judgment, the problem of mismatching of the null depth and the interference coming direction in special motion scenes such as a large off-axis attitude is solved, meanwhile, the anti-interference limit performance in a conventional operation scene is not affected, and the self-adaptive capacity of the anti-interference navigation terminal is effectively improved.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of satellite navigation, and particularly relates to a space-frequency anti-interference method based on adaptive null broadening. BACKGROUND

[0002] In order to solve the problem of satellite navigation anti-suppression interference, the space-frequency anti-interference method based on an array antenna is generally used in navigation receivers at home and abroad to suppress multi-direction and multi-style suppression interference.

[0003] However, in a large off-axis or special motion scenario, the direction of interference relative to the receiver can change rapidly, so that the interference direction of the weight value training is mismatched to a certain extent with the interference direction in actual application, and the interference is easy to move out of the null direction, resulting in failure of the traditional adaptive interference suppression algorithm.

[0004] In view of the rapid change of the interference direction in the special motion scenario, the null broadening algorithm has good effect on solving this problem by expanding the interference null. SUMMARY

[0005] Therefore, the application aims to provide a space-frequency anti-interference method based on adaptive null broadening, which can solve the problem of mismatch between null depth and interference direction in a large off-axis attitude or other special motion scenarios, without affecting the anti-interference limit performance in the conventional operation scenario, and effectively improve the adaptive ability of the anti-interference navigation terminal.

[0006] In a first aspect, the embodiments of the application provide a space-frequency anti-interference method based on adaptive null broadening, which comprises the following steps:

[0007] 1. A space-frequency anti-interference method based on adaptive null broadening, and the specific steps are as follows:

[0008] S1. Obtain a digital intermediate frequency signal, and convert the time domain signal into a frequency domain signal;

[0009] S2. Calculate the covariance matrix of each channel frequency domain data;

[0010] S3. Judge the special motion scenario of the large off-axis attitude, and reconstruct the interference plus noise covariance matrix;

[0011] S4. Calculate the sub-band weight coefficient according to the covariance matrix, and realize matrix inversion through the Gauss-Jordan elimination method;

[0012] S5. Perform space domain filtering on each narrow sub-band according to the weight value and convert the time domain data.

[0013] Compared with the prior art, the application has the beneficial effects that:

[0014] (1) The application adopts Doppler change rate to identify whether the current scene is a special motion scene such as a large off-axis posture, solves the problem of zero-depth mismatch and interference direction mismatch in the special motion scene such as a large off-axis posture, does not affect the anti-interference limit performance in the conventional running scene, and effectively improves the adaptive ability of the anti-interference navigation terminal.

[0015] (2) It is difficult to directly perform high-dimensional matrix inversion in engineering, the application introduces Gauss-Jordan elimination method to realize high-dimensional matrix inversion, has good numerical stability, low algorithm complexity, and completes lightweight implementation in the inversion process. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the application or the prior art, the drawings needed to be used in the embodiments or the prior art description will be briefly introduced as follows. Obviously, for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.

[0017] Figure 1 is a work flow diagram of the application; DETAILED DESCRIPTION

[0018] The technical solutions in the embodiments of the application will be described clearly and completely in combination with the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, not all the embodiments. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor are within the protection scope of the application.

[0019] The application proposes a space-frequency anti-interference method based on adaptive null broadening, aiming to solve the problem of zero-depth mismatch and interference direction mismatch in the special motion scene such as a large off-axis posture, without affecting the anti-interference limit performance in the conventional running scene, and effectively improving the adaptive ability of the anti-interference navigation terminal.

[0020] As shown in Figure 1 , the application proposes a space-frequency anti-interference method based on adaptive null broadening, including the following steps:

[0021] S1: Collecting digital intermediate frequency data, converting time domain signals into frequency domain signals:

[0022] (1) Digital intermediate frequency signal: obtaining M-way antenna array digital intermediate frequency signals obtained through an array antenna, down-conversion and analog-to-digital conversion, and denoted as , wherein M represents the number of antenna array elements, m is the serial number of the antenna array element, ; This represents the intermediate frequency signal of the m-th element;

[0023] (2) Perform a data delay sampling operation to convert the time-domain signal into a frequency-domain signal through windowing / DFT operation;

[0024] Delay the digital intermediate frequency signal for each sub-channel separately. After a sampling time, the delayed signal can be obtained. , , N is the sampling length; This represents the time-domain signal of the m-th antenna element after an n-sampling delay. It is converted to the frequency domain through windowing and Discrete Fourier Transform (DFT), with the following conversion formula:

[0025] ;

[0026] Where N is the sampling length, n is the sampling sequence number, M is the number of antenna elements, and m is the sequence number of the antenna elements. Let f be a window function of length N, where k is the data index, i.e., let f k Let k be the frequency domain sub-narrowband. For frequency domain sub-narrowband signals, representing the first... After DFT, the individual element digital intermediate frequency data corresponds to the frequency domain sub-narrowband f. k The frequency domain value; Let j be the time-domain signal of the m-th antenna element after n sampling delays, where j is the imaginary part.

[0027] S2: Calculate the covariance matrix of the frequency domain data for each channel. ;

[0028] For each frequency domain narrow subband f k Calculate the covariance matrix of the frequency domain data for each channel under normal operating conditions. And using L frequency domain data block samples, the covariance matrix is ​​estimated. The theoretical calculation formula is:

[0029] ;

[0030] in, Represents frequency domain data, , It is a frequency domain sub-narrowband signal. Let L be the covariance matrix, and L be the number of samples in the frequency domain data block.

[0031] S3: Determine special motion scenarios with large off-axis attitudes and reconstruct the interference plus noise covariance matrix;

[0032] S3.1: Use Doppler rate of change to determine whether the current scene is a special motion scene such as a large off-axis posture:

[0033] The Doppler rate of change Dr is used to identify whether the current scene is a special motion scene such as a large off-axis posture. When Dr > β (β is a set threshold that needs to be set according to engineering experience parameters), the process proceeds to step S3.2; otherwise, it proceeds directly to step S4. This step is the core step of this patent. By using the Doppler rate of change, it determines whether the current scene needs null widening, thereby avoiding the impact of null widening on the anti-interference limit performance under normal operating scenarios.

[0034] S3.2: Reconstruct the interference plus noise covariance matrix using the spatial spectrum estimate of the interference region;

[0035] The interference plus noise covariance (INC) matrix is ​​reconstructed using the spatial spectrum estimate of the interference region. The matrix is ​​reconstructed to achieve the zero-trap broadening function. The reconstructed matrix expression is as follows:

[0036] ;

[0037] in Represents the reconstructed matrix. This represents the Capon power spectrum. This represents the angular range containing the desired signal angle of arrival. express The complement of the set, i.e., the angular interval that does not contain the desired signal. The angle of the guide vector, and They represent angles respectively. The guiding vector and its conjugate transpose.

[0038] This step estimates the number of interfering signals by comparing the ratios of adjacent eigenvalues ​​of the sampled covariance matrix. Then, based on the number of interfering signals and the null broadening requirements, the spatial spectrum estimate for the region near the direction of the interfering signal is reset. Finally, the INC matrix is ​​reconstructed using the reset spatial spectrum estimate, and the adaptive array weight vector is solved. This reconstruction method avoids both estimating the direction of the desired GNSS signal and the impact of noise power estimation errors on the INC matrix reconstruction accuracy, thus solving the problem of desired GNSS signals contaminating the array's received data.

[0039] S4: Calculate the sub-band weight coefficients based on the covariance matrix, and invert the matrix using the Gauss-Jordan elimination method;

[0040] S4.1: Calculate the anti-interference weights for each sub-band;

[0041] Among them, the sub-band weight coefficients are calculated based on the covariance matrix. This is the anti-interference weight, calculated as follows:

[0042] ;

[0043] in, This represents the subband weighting coefficients of the narrow subband in the frequency domain. For the covariance matrix or using the reconstructed matrix In substitution, s is the constraint vector. This represents the conjugate transpose of the constraint vector.

[0044] S4.2: Matrix inversion using the Gauss-Jordan elimination method;

[0045] To address the computational complexity issue of multidimensional matrix inversion in this step, the Gauss-Jordan elimination method is employed to achieve matrix inversion. This method features good numerical stability and low algorithm complexity. Specifically, assume the target matrix is... Square Array First use identity matrix of order Constructing augmented matrices :

[0046] ;

[0047] Then, the matrix in the above equation is transformed using elementary row operations. Transformed into:

[0048] ;

[0049] In the formula, To find the augmented matrix with the inverse completed, That is, the target matrix The inverse matrix.

[0050] for 1st order target matrix Construct an augmented matrix, and start from the first row of the matrix. The main steps to solve this problem are as follows:

[0051] (1) For Augmented matrix Divide all elements in the row by the pivot, making the pivot equal to 1;

[0052] (2) The first Multiply by each row The column elements of the column containing the pivot (excluding the pivot itself) are subtracted from the corresponding row elements, and so on, up to the nth column. All other columns in the column containing the pivot element are set to 0;

[0053] (3) Repeat the first two steps until... ;

[0054] Based on the main computational steps of the Gauss-Jordan elimination method, it can be seen that the Gauss-Jordan inversion process only includes three basic mathematical operations: addition / subtraction, multiplication, and division. It does not require matrix decomposition and matrix multiplication, and the algorithm structure is simple and suitable for pipelined implementation.

[0055] S5: Based on the anti-interference weights, perform spatial filtering on each narrow sub-band and convert it into time domain data;

[0056] Based on anti-interference weights Spatial filtering is performed on each narrow subband, and the calculation formula is as follows: ,

[0057] in This represents all frequency domain data after interference suppression.

[0058] Then the frequency domain data after spatial filtering is processed. The Inverse Discrete Fourier Transform (IDFT) is used to transform frequency domain data into the time domain, and then an inverse window is applied to the transformed time domain data.

[0059] ;

[0060] in This represents the time-domain data after interference suppression. This represents the k-th frequency domain data after anti-interference processing.

[0061] Finally, weighting is performed to increase the number of effective bits in the signal, thus obtaining the final anti-interference data.

[0062] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0063] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention.

[0064] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention without departing from the principles and spirit of the present invention.

Claims

1. A space-frequency anti-interference method based on adaptive null broadening, characterized in that, The specific steps are as follows: S1. Obtain the digital intermediate frequency signal and convert the time-domain signal into a frequency-domain signal; S2. Calculate the covariance matrix of the frequency domain data for each channel; S3. Determine the special motion scenarios with large off-axis attitudes and reconstruct the interference plus noise covariance matrix; S4. Calculate the sub-band weight coefficients based on the covariance matrix, and invert the matrix using the Gauss-Jordan elimination method. S5. Perform spatial filtering on each narrow subband according to the weights and convert it into time domain data.

2. The space-frequency anti-interference method based on adaptive null broadening according to claim 1, characterized in that, In step S1, windowing and discrete Fourier transform convert the time-domain signal into a frequency-domain signal for processing. The formula is as follows: ； Where N is the sampling length, n is the sampling sequence number, M is the number of antenna elements, and m is the sequence number of the antenna elements. Let f be a window function of length N, where k is the data index, i.e., let f k Let k be the frequency domain sub-narrowband. For frequency domain sub-narrowband signals, representing the first... After DFT, the individual element digital intermediate frequency data corresponds to the frequency domain sub-narrowband f. k The frequency domain value; Let j be the time-domain signal of the m-th antenna element after n sampling delays, where j is the imaginary part.

3. The space-frequency anti-interference method based on adaptive null broadening according to claim 1, characterized in that, In S2, for each frequency domain sub-narrowband f k Calculate the covariance matrix of the frequency domain data for each channel under normal operating conditions. And using L frequency domain data blocks as samples, the covariance matrix is ​​estimated. The theoretical calculation formula is: ； in, Represents frequency domain data, It is a frequency domain sub-narrowband signal. Let L be the covariance matrix, and L be the number of samples in the frequency domain data block.

4. The space-frequency anti-interference method based on adaptive null broadening according to claim 1, characterized in that, The specific steps in S3 are as follows: S3.1: Use the Doppler rate of change to determine whether the current scene is a special motion scene such as a large off-axis posture; specifically, when the Doppler rate of change Dr > the set threshold β, proceed to step S3.2; otherwise, proceed directly to S4; S3.2: Reconstruct the interference plus noise covariance matrix using the spatial spectrum estimate of the interference region; (1) The number of interference signals is estimated by the ratio of adjacent eigenvalues ​​of the sampled covariance matrix; (2) Based on the number of interference signals and the requirements for null broadening, the spatial spectrum estimate of the area near the direction of the interference signal is reset; (3) Reconstruct the INC matrix using the reset spatial spectrum estimate and solve for the adaptive array weight vector.

5. The space-frequency anti-interference method based on adaptive null broadening according to claim 4, characterized in that, In S3.2, the covariance matrix is ​​reconstructed at this point. Using reconstruction matrix The substitution calculation is performed, and the expression is as follows: ； in, Represents the reconstructed matrix. Represents the Capon power spectrum. This represents the angular range containing the desired signal angle of arrival. express The complement of the set, i.e., the angular interval that does not contain the desired signal. The angle of the guide vector, and They represent angles respectively. The guiding vector and its conjugate transpose.

6. The space-frequency anti-interference method based on adaptive null broadening according to claim 1, characterized in that, In S4, the sub-band weight coefficients are calculated based on the covariance matrix. The sub-band weighting coefficient is the anti-interference weight, and the calculation method is as follows: ； in, This represents the subband weighting coefficients of the narrow subband in the frequency domain. Covariance matrix or reconstructed matrix In substitution, s is the constraint vector. This represents the conjugate transpose of the constraint vector.

7. The space-frequency anti-interference method based on adaptive null broadening according to claim 1, characterized in that, In S4, matrix inversion is achieved using the Gauss-Jordan elimination method. Specifically, assume the target matrix is Square Array First use identity matrix of order Constructing augmented matrices : ； Then, the matrix in the above equation is transformed using elementary row operations. Transformed into: ； In the formula, To find the augmented matrix with the inverse completed, That is, the target matrix The inverse matrix; for 1st order target matrix Construct an augmented matrix, and start from the first row of the matrix. The main steps to begin solving for its inverse matrix are as follows: (1) For Augmented matrix Divide all elements in the row by the pivot, making the pivot equal to 1; (2) The first Multiply by each row The column elements of the column containing the pivot (excluding the pivot itself) are subtracted from the corresponding row elements, and so on, up to the nth column. All other columns in the column containing the pivot element are set to 0; (3) Repeat the first two steps until... ; Ultimately, the matrix In This is the obtained inverse matrix.

8. The space-frequency anti-interference method based on adaptive null broadening according to claim 1, characterized in that, The specific steps of S5 are as follows: (1) Based on the sub-band weight coefficients Perform spatial filtering on each narrow subband; (2) Perform the frequency domain data after spatial filtering. The point-based discrete Fourier inverse transform converts frequency domain data to the time domain, and then applies an inverse window to the transformed time-domain data; the formula is: ； in, This represents the time-domain data after interference suppression. This represents the k-th frequency domain data after anti-interference processing. The window function represents the length N; k represents the data index, i.e., the kth data point, and j is the imaginary part; (3) Finally, weighting is performed to increase the number of effective bits of the signal, and the final anti-interference data can be obtained.

Images