Space-time-frequency domain zeroing anti-interference method and device for emergency satellite communication

Abstract

The invention discloses a space-time-frequency domain zeroing anti-interference method and device for emergency satellite communication, and the method comprises the following steps: S1, receiving a signal, and sequentially carrying out frequency domain filtering, low-noise amplifier and analog-to-digital conversion to obtain a sampling sequence; s2, carrying out amplitude and phase calibration on the sampling sequence; s3, estimating an interference direction by adopting an ESPRIT algorithm to obtain K signal sequences; s4, screening the K signals one by one by adopting an LCMV interference suppression algorithm to obtain a signal S (n) after interference suppression; s5, performing baseband signal processing on the S (n), wherein the baseband signal processing comprises frequency domain cross-correlation; s6, determining an azimuth angle of the sampling signal based on a cross-correlation result, and obtaining an optimal zeroing weight matrix W; and S7, completing space-time anti-interference suppression processing of the received signal according to the optimal zero setting weight matrix W. ESPRIT and LCMV are combined, and self-adaptive interference suppression of beam direction estimation and satellite communication signal receiving is achieved more accurately.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of emergency satellite communication terminals, and particularly relates to a space-time-frequency domain nulling anti-interference method and device for an emergency satellite communication terminal. BACKGROUND

[0002] Satellite communication terminal nulling anti-interference technology is a key means to ensure communication stability and security, especially in complex electromagnetic environments and multi-interference source scenarios. Current technology development shows a multi-dimensional fusion trend, covering space domain, time domain, frequency domain and AI empowerment and other directions.

[0003] Space domain nulling technology focuses on two core links of direction of arrival estimation and weight optimization, and forms a radiation null in the interference direction through an array antenna. For example, CN114422013A (adaptive anti-interference satellite communication method) proposes a cascade architecture of two-dimensional MUSIC algorithm and LCMV criterion: first, a low-precision fast spectral peak search is performed using the MUSIC algorithm, then P candidate signals are filtered one by one through the LCMV algorithm, and the real satellite signal is identified in combination with the frequency domain correlation of the baseband spread spectrum signal. This scheme improves the processing speed by 60% by reducing the search precision of the S frequency band (5° step), but is limited to static environment, and in dynamic scenarios, the null depth decreases due to direction misalignment.

[0004] CN118300635A (nulling antenna interference suppression device) uses an attitude angle prediction and active null polling mechanism: based on ephemeris prediction, the angles of visible stars are predicted and the active null of the real satellite direction is generated, the related peaks in the residual signal are detected to identify deception interference, and then a null polling is formed in the suspicious space domain. This scheme solves the problem of deception interference suppression in the full power interval, but frequent weight updating increases the processing delay, and actual measurement shows that the interference suppression stability is insufficient in dynamic environment. Further research shows that the weight update rate of a seven-element uniform circular array needs to be increased to more than 100Hz to meet the needs of high dynamic environment.

[0005] Space-time processing (STAP) technology combines time domain tap extension and space domain filtering, significantly improving the degree of freedom and interference suppression dimension. Patent application No. 20171004561.3 discloses a space-time-frequency architecture satellite navigation anti-interference method, which proposes a cascade processing framework: first, narrowband interference is suppressed by frequency domain FFT notch, and then wideband interference is suppressed by space-time adaptive processing (STAP). The innovation lies in the segmented average power threshold method: the 20.46MHz bandwidth is divided into N segments and the M frame power median is calculated as a dynamic threshold, effectively solving the multi-intensity narrowband interference detection problem, and the actual measurement interference suppression ratio is improved by 12dB. However, this scheme needs to cascade multiple processing modules, resulting in a doubling of hardware resource consumption, and the logic resource occupancy rate is as high as 78% when implemented on an FPGA.

[0006] To solve the problem of computational complexity bottleneck, the patent application with the application number 202111592685.X introduces a beam synthesis-blocking joint processing structure: the baseband signal is split into beam synthesis signal and blocking signal, the interference components are reconstructed by adaptive weighting and cancellation is performed. This structure avoids the inverse of the covariance matrix, and the operation amount is reduced by 40%, but the signal leakage problem causes the expected signal loss of about 2dB.

[0007] The prior art generally has the defects of weak dynamic adaptability and large hardware overhead, and the weight update algorithm in the space-time processing (STAP) technology is complex, and frequent update brings high delay of signal processing. Therefore, a technical solution is needed to balance the computational complexity and hardware overhead, while having a good interference suppression ratio. SUMMARY

[0008] The purpose of the present application is to overcome the shortcomings of the prior art, and to provide an ESPRIT+LCMV combined space-time-frequency domain nulling anti-jamming method that more accurately realizes beam direction estimation and adaptive interference suppression of received satellite communication signals, and a device for implementing the space-time-frequency domain nulling anti-jamming method.

[0009] The purpose of the present application is achieved by the following technical solution: a space-time-frequency domain nulling anti-jamming method for emergency satellite communication, comprising the following steps:

[0010] S1, receiving signals through an antenna array, and sequentially passing the signals received by each antenna through a frequency domain filter, a low noise amplifier and an analog-to-digital converter to obtain a sampling sequence x(n) of each antenna received signal;

[0011] S2, amplitude and phase calibration is performed on the sampling sequence x(n) to obtain a calibrated data sequence;

[0012] S3, the calibrated data sequence is used to estimate the interference direction by using the ESPRIT algorithm to obtain K signal sequences x K (n);

[0013] S4, the K signal sequences x K (n) are screened one by one by using the LCMV interference suppression algorithm to obtain interference suppressed signals S(n);

[0014] S5, baseband signal processing is performed on the interference suppressed signals S(n), which includes frequency domain cross-correlation;

[0015] S6, based on the cross-correlation result, the azimuth of the received signal is determined, and the optimal nulling weight matrix W is obtained;

[0016] S7, complete the space-time anti-jamming suppression processing of the received signal according to the optimal zero-adjusting weight matrix W.

[0017] The S3 comprises the following steps:

[0018] S31, data preprocessing: calculating the covariance matrix of the signal sequence x(n) X is the discrete Fourier transform of x(n), and H is the conjugate transpose;

[0019] S32, performing eigenvalue decomposition on , and retaining the eigenvectors U corresponding to the first K large eigenvalues s ;

[0020] S33, dividing U s into U s1 and U s2 two vectors;

[0021] S34, obtaining Ψ by using total least squares TLS-ESPRIT:

[0022] S35, performing eigenvalue decomposition on Ψ, arranging the eigenvalues in descending order, taking the first K signals to form a signal sequence x K (n), and estimating the beam incident angle θ k corresponding to each signal, k=1,…,K.

[0023] The S4 comprises:

[0024] S41, inverting the covariance matrix of the sampling sequence x(n);

[0025] S42, calculating the optimal weight vector by using the Lagrange multiplier method:

[0026]

[0027] Wherein, W LCMV is the optimal weight vector of the LCMV beamformer, a(θ) is the steering vector, is the inverse matrix of the covariance matrix; wherein, θ={θ0, θ1,…, θ k ,…,θ K}.

[0028] S43, multiplying the conjugate transpose of W LCMV with the K signals to obtain the signal S(n) after interference suppression:

[0029] S(n)=W LCMV H *x K (n).

[0030] In the S41, the inverse of the covariance matrix of the sampling sequence x(n) is solved by using a recursive least square (RLS) or a conjugate gradient method for acceleration.

[0031] The S7 comprises: delaying the sampling sequence after analog-to-digital conversion respectively to obtain signals x(1,n), x(1,n-τ)…x(1,n-(N-1)*τ), multiplying the delayed signals by weight coefficients w(1,1), w(1,2)…w(1,N) respectively, and summing to obtain an output signal Y(n); which is expressed as:

[0032]

[0033] Wherein, M is the number of elements of the antenna array; N is the number of time delay units, that is, the order of time domain filtering; the time interval of each time delay unit is τ; τ≤1 / B, wherein B is the receiver processing bandwidth; Y(n) is the fusion output signal of the antenna array data x(n) after space-time-frequency domain interference suppression, {w(i,j)} is the element of the i-th row and the j-th column of the optimal zeroing weight matrix W, i=1,2…M, j=1,2,…,N.

[0034] Another object of the present application is to provide a space-time-frequency domain zeroing anti-interference device for emergency satellite communication, which is used to realize the space-time-frequency domain zeroing anti-interference method, comprising:

[0035] A data acquisition module is configured to acquire a sampling sequence x(n) of an antenna receiving signal.

[0036] An amplitude and phase calibration module is configured to perform amplitude and phase calibration on the sampling sequence x(n).

[0037] An estimation module is configured to estimate interference directions based on the sampling sequence by using an ESPRIT algorithm to obtain K signal sequences.

[0038] A screening and suppression module is configured to screen the K signal sequences by using an LCMV interference suppression algorithm to obtain interference-suppressed signals S(n), n=1,2,…,K.

[0039] A baseband signal processing module is configured to perform baseband signal processing on the obtained signals S(n), and the baseband signal processing comprises frequency domain cross-correlation.

[0040] A weight determination module is configured to determine an azimuth angle of the sampling signal based on the frequency domain cross-correlation result and obtain an optimal zeroing weight matrix W.

[0041] A power inversion module is configured to implement space-time-frequency domain interference suppression of the receiving signal based on the optimal zeroing weight W.

[0042] The present application has the following beneficial effects:

[0043] 1. The ESPRIT algorithm is used to estimate the direction of interference. Compared with the MUSIC algorithm, it has lower complexity and lower signal processing delay and hardware overhead.

[0044] 2. The ESPRIT algorithm is integrated with a neural network model to improve algorithm accuracy and make the direction estimation of interference beams more accurate;

[0045] 3. By combining ESPRIT and LCMV, beam direction estimation and adaptive interference suppression of received satellite communication signals are achieved with greater accuracy. Attached Figure Description

[0046] Figure 1 This is a flowchart of a spatiotemporal frequency domain zeroing anti-interference method for emergency satellite communication according to the present invention. Detailed Implementation

[0047] The technical solution of the present invention will be further described below with reference to the accompanying drawings.

[0048] like Figure 1 As shown, this invention provides a space-time-frequency domain nulling anti-interference method for emergency satellite communication, applied in emergency satellite communication terminals to achieve interference suppression in three-dimensional space, time, and frequency domains. Specifically, it includes the following steps:

[0049] S1. Receive signals through the antenna array, and sequentially process the signals received by each antenna through frequency domain filtering, low-noise amplifier and analog-to-digital conversion to obtain the sampling sequence x(n) of the signals received by each antenna;

[0050] S2. Perform amplitude and phase calibration on the sampled sequence x(n) to obtain the calibrated data sequence;

[0051] S3. For the calibrated data sequence, the ESPRIT algorithm is used to estimate the interference direction, resulting in K signal sequences x. K (n); includes the following steps:

[0052] S31. Data preprocessing: Calculate the covariance matrix of the signal sequence x(n). X is the discrete Fourier transform of x(n), and H is the conjugate transpose;

[0053] S32, to Perform eigenvalue decomposition and retain the eigenvectors U corresponding to the top K largest eigenvalues. s ;

[0054] S33. Will U s Divided into U s1 and U s2 Two vectors can be divided according to rows or columns, or they can be divided randomly;

[0055] S34, solve for by total least squares TLS-ESPRIT: Rotational Invariance Subspace Algorithm (Rotational Invariance Subspace Algorithm) is a kind of parameter estimation method based on signal subspace structure, its core idea is to use the rotational invariance of signal subspace to estimate signal parameters (such as direction of arrival, frequency, etc.). ESPRIT (Estimation of Signal Parameters via Rotational Invariance Techniques) is a typical representative of this kind of algorithm.

[0056] S35, eigenvalue decomposition is carried out on Ψ, the eigenvalues are arranged in descending order, the first K signals are taken to form a signal sequence x K (n), and the beam incident angle θ k of each signal is estimated.

[0057] The method for estimating the beam incident angle is:

[0058] 1) extract the phase angle of each eigenvalue λ K in x k (n): Wherein, ∠ represents taking the phase of complex number (radian system);

[0059] 2) according to the phase shift relationship of subarray (element spacing d and wavelength λ):

[0060]

[0061] Get:

[0062]

[0063] S4, the LCMV interference suppression algorithm is used to screen the K signal sequences x K (n) one by one, to obtain the signal S(n) after interference suppression, n=1, 2, …, K; LCMV interference suppression algorithm (linear constraint minimum variance) algorithm is a classical adaptive beam forming technology, which is widely used in radar, communication, acoustics and other fields, for suppressing interference and noise while ensuring target signal distortion free. The present application introduces the algorithm and makes certain improvements, including:

[0064] S41, the covariance matrix of the sampling sequence x(n) is inverted;

[0065] Preferably, before inverting the covariance matrix, the covariance matrix can be corrected by vector first, and the specific correction method can be: Wherein, Γ is greater than 0, I is a unit matrix; the modified covariance matrix can improve the stability of the numerical value and suppress noise.

[0066] Preferably, the operation of the above matrix inversion can be accelerated by using recursive least squares (RLS) or conjugate gradient method.

[0067] S42, the optimal weight vector is calculated by Lagrange multiplier method:

[0068]

[0069] Wherein, W LCMV is the optimal weight vector of the LCMV beamformer, which determines the weighting coefficient of each channel of the array and is used to synthesize the output signal; a (θ) is a steering vector, which describes the response mode of the array to the signal with the direction of arrival θ; is the inverse matrix of the covariance matrix; wherein, θ={θ0, θ1,…, θ k ,…,θ K};

[0070] S43, the conjugate transpose of W LCMV is multiplied by K signals to obtain the interference suppressed signal S(n):

[0071] S(n)=W LCMV H *x K (n)。

[0072] S5, baseband signal processing is performed on the interference suppressed signal S(n), and the baseband signal processing includes frequency domain cross correlation.

[0073] S6, the azimuth angle of the received signal is determined based on the cross correlation result, and the optimal zeroing weight matrix W is obtained.

[0074] S7, the space-time anti-interference suppression processing of the received signal is completed according to the optimal zeroing weight matrix W; including: respectively performing time delay on the analog-to-digital converted sampling sequence to obtain signals x(1,n), x(1,n-τ)…x(1,n-(N-1)*τ), then respectively multiplying the time delayed signals by weight coefficients w(1,1), w(1,2)…w(1,N), and then summing to obtain an output signal Y(n); represented as:

[0075]

[0076] Wherein, M is the number of antenna array elements; N is the number of time delay units, that is, the time domain filter order; the time interval of each time delay unit is τ, τ≤1 / B, wherein B is the receiver processing bandwidth. Y(n) is the fusion output signal of the antenna array data x(n) after space-time-frequency domain interference suppression, {w(i,j)} is the element of the i-th row and the j-th column of the optimal zeroing weight matrix W, i=1,2…M, j=1,2,…,N; w(i,j) is a space-time two-dimensional weighting coefficient, that is, a tap coefficient in a multi-stage time domain filter.

[0077] Through the technical scheme provided by the application, at least 90% of the interference of the Ku-band satellite terminal is suppressed; at the same time, compared with the MUSIC-LCMV, the search time is reduced by at least 30%, the processing delay of anti-interference suppression is reduced, and the algorithm complexity and hardware overhead are low.

[0078] The application also provides a space-time-frequency domain zeroing anti-interference device for emergency satellite communication, which is used to realize the space-time-frequency domain zeroing anti-interference method, and comprises:

[0079] A data acquisition module is configured to acquire a sampling sequence x(n) of an antenna receiving signal.

[0080] An amplitude and phase calibration module is configured to perform amplitude and phase calibration on the sampling sequence x(n).

[0081] An estimation module is configured to estimate an interference direction based on the sampling sequence by using an ESPRIT algorithm to obtain K signal sequences.

[0082] A screening and suppression module is configured to screen the K signal sequences by using an LCMV interference suppression algorithm to obtain a signal S(n) after interference suppression, n=1,2,…,K.

[0083] A baseband signal processing module is configured to perform baseband signal processing on the obtained signal S(n), and the baseband signal processing comprises frequency domain cross-correlation.

[0084] A weight determination module is configured to determine an azimuth angle of the sampling signal based on the frequency domain cross-correlation result and obtain an optimal zeroing weight matrix W.

[0085] A power inversion module is configured to implement space-time-frequency domain interference suppression of the receiving signal based on the optimal zeroing weight W.

[0086] Through signal testing of actual products, the application realizes beam direction estimation and adaptive interference suppression of receiving satellite communication signals more accurately by using the method combining ESPRIT and LCMV. Meanwhile, the robustness and real-time performance can be improved through machine learning and hardware optimization.

[0087] Those skilled in the art will appreciate that the embodiments described herein are presented for purposes of illustration and that the inventive principles are not limited to these particular embodiments. Other variations and modifications can be made to the embodiments without departing from the spirit and scope of the inventive principles.

Claims

1. A spatiotemporal frequency domain nulling anti-interference method for emergency satellite communication, characterized in that, Includes the following steps: S1. Receive signals through the antenna array, and sequentially process the signals received by each antenna through frequency domain filtering, low-noise amplifier and analog-to-digital conversion to obtain the sampling sequence x(n) of the signals received by each antenna; S2. Perform amplitude and phase calibration on the sampled sequence x(n) to obtain the calibrated data sequence; S3. For the calibrated data sequence, the ESPRIT algorithm is used to estimate the interference direction, resulting in K signal sequences x. K (n); S4. The LCMV interference suppression algorithm is used to process the K signal sequences x. K (n) are filtered one by one to obtain the signal S(n) after interference suppression; S5. Perform baseband signal processing on the interference suppression signal S(n). Baseband signal processing includes frequency domain cross-correlation. S6. Determine the azimuth angle of the received signal based on the cross-correlation results, and obtain the optimal zero-adjustment weight matrix W; S7. Perform the space-time anti-interference suppression processing of the received signal according to the optimal zero-adjustment weight matrix W.

2. The spatiotemporal frequency domain nulling anti-interference method for emergency satellite communication according to claim 1, characterized in that, S3 includes the following steps: S31. Data preprocessing: Calculate the covariance matrix of the signal sequence x(n). X is the discrete Fourier transform of x(n), and H is the conjugate transpose; S32, to Perform eigenvalue decomposition and retain the eigenvectors U corresponding to the top K largest eigenvalues. s ; S33. Will U s Divided into U s1 and U s2 Two vectors; S34. Obtain Ψ by using total least squares TLS-ESPRIT: S35. Perform eigenvalue decomposition on Ψ, arrange the eigenvalues ​​in descending order, and take the first K signals to form a signal sequence x. K (n), and estimate the beam incidence angle θ corresponding to each signal. k , k=1,…,K.

3. The spatiotemporal frequency domain nulling anti-interference method for emergency satellite communication according to claim 1, characterized in that, S4 includes: S41. Invert the covariance matrix of the sampled sequence x(n); S42. The optimal weight vector is obtained by calculating the Lagrange multiplier method: Among them, W LCMV Let a(θ) be the optimal weighting vector for the LCMV beamformer, and a(θ) be the steering vector. Let θ be the inverse of the covariance matrix; where θ = {θ0, θ1, ..., θ...} k ,…,θ K }; S43, W LCMV Multiplying the conjugate transpose of the signal by the K signals yields the interference-suppressed signal S(n): S(n)=W LCMV H *x K (n)。 4. The spatiotemporal frequency domain nulling anti-interference method for emergency satellite communication according to claim 3, characterized in that, In step S41, the inversion of the covariance matrix of the sampling sequence x(n) is accelerated by recursive least squares or conjugate gradient method.

5. The spatiotemporal frequency domain nulling anti-interference method for emergency satellite communication according to claim 1, characterized in that, S7 includes: delaying the sampled sequences after analog-to-digital conversion to obtain signals x(1,n), x(1,n-τ)…x(1,n-(N-1)*τ), multiplying the delayed signals by weight coefficients w(1,1), w(1,2)…w(1,N) respectively, and then summing them to obtain the output signal Y(n); expressed as: Where M is the number of elements in the antenna array; N is the number of time delay units, i.e., the time domain filter order; the time interval of each time delay unit is τ; Y(n) is the fused output signal of the antenna array data x(n) after spatiotemporal interference suppression; and {w(i,j)} is the element in the i-th row and j-th column of the optimal zeroing weight matrix W, i = 1, 2, ..., M, j = 1, 2, ..., N.

6. The spatiotemporal frequency domain nulling anti-interference method for emergency satellite communication according to claim 5, characterized in that, τ≤1 / B, where B is the receiver processing bandwidth.

7. A space-time-frequency domain zeroing anti-interference device for emergency satellite communication, used to implement the space-time-frequency domain zeroing anti-interference method according to any one of claims 1 to 6, characterized in that, include: The data acquisition module is used to acquire the sampling sequence x(n) of the signal received by the antenna; The amplitude and phase calibration module is used to calibrate the amplitude and phase of the sampled sequence x(n); The estimation module is used to estimate the direction of interference based on the sampled sequence using the ESPRIT algorithm, and obtain K signal sequences; The filtering and suppression module is used to filter K signal sequences using the LCMV interference suppression algorithm to obtain the interference-suppressed signal S(n), n=1,2,…,K; A baseband signal processing module is used to perform baseband signal processing on the obtained signal S(n), wherein the baseband signal processing includes frequency domain cross-correlation; The weight determination module is used to determine the azimuth angle of the sampled signal based on the frequency domain cross-correlation results, and to obtain the optimal zero-adjustment weight matrix W; The power inversion module is used to suppress spatial-temporal-frequency interference of the received signal based on the optimal zeroing weight W.

Images