CN104007453A - Frequency domain and space domain joint anti-interference method aided by probability search - Google Patents

Abstract

The invention discloses a probability search assisted joint anti-jamming method in the frequency domain and air domain, which is implemented according to the following steps: step 1, signal preprocessing, and finally divide them into I-channel and Q-channel signals for output; step 2, frequency domain processing, The I-channel and Q-channel signals finally output in step 1 are divided into two branches for processing respectively, and then the delayed signal and the obtained signal are synthesized by an adder; step 3, spatial processing, and the final synthesized signal in step 2 The output signal is subjected to spatial power inversion adaptive filtering, and the broadband interference in the signal is suppressed to obtain the final output signal. The method of the invention solves the problem that the performance of the anti-jamming algorithm is reduced or fails when the narrowband interference is submerged in the wide-band interference when the frequency-domain and air-space joint anti-jamming algorithm is used, and the applicable range of the frequency-domain and air-space joint anti-jamming algorithm is expanded.

Description

Probabilistic search-assisted joint anti-jamming method in frequency domain and space domain

technical field

The invention belongs to the technical field of anti-jamming of global navigation satellite systems, and relates to a joint anti-jamming method of frequency domain and air domain aided by probability search.

Background technique

With the maturity of the Global Navigation Satellite System (GNSS) technology, GNSS has shown its incomparable importance in both the military and civilian fields, and the research, development and application of GNSS technology are increasingly Many people's attention. However, the signal power of navigation satellites is low, and they are easily affected by various intentional or unintentional interferences. Especially in wartime, they will be suppressed by strong interference and cannot work normally.

In the prior art, in a complex electromagnetic environment where narrowband interference and broadband interference coexist, algorithms such as spatial filtering, space-time joint (STAP), and frequency-domain concatenated airspace are usually used for anti-interference processing. Although the pure spatial filtering algorithm can filter out narrowband interference and broadband interference at the same time, narrowband interference will reduce the degree of freedom of the array antenna and reduce the number of anti-interference. The algorithm of space-time combination is complicated, the calculation amount is too large, and the hardware requirement is high. Frequency domain cascaded air domain first uses frequency domain filtering technology to filter out narrowband interference, and then filters out more interference through air domain filtering, but when narrowband and broadband interference coexist, and the narrowband interference power is lower than broadband interference, frequency domain filtering detection is performed If the location of the narrowband interference is not found, the common frequency domain cascaded air domain anti-jamming algorithm will fail. Therefore, the existing technologies have their own advantages and disadvantages, and it is necessary to develop new anti-jamming methods for navigation satellites.

Contents of the invention

The purpose of the present invention is to provide a probability search assisted joint anti-jamming method in the frequency domain and space domain, which solves the problem of coexistence of narrowband interference and broadband interference in the case of interference signals in the prior art, especially that the center frequency of narrowband interference falls within the bandwidth of broadband interference Medium, and the power is lower than the broadband interference, resulting in obvious system performance degradation or malfunction.

The technical solution adopted in the present invention is a frequency-domain-space-domain joint anti-jamming method assisted by probability search, which is implemented according to the following steps:

Step 1, signal preprocessing;

Step 2, frequency domain processing;

Step 3, airspace processing.

The beneficial effects of the present invention are: A/D acquisition, digital down-conversion, filtering, and interpolation are performed on the intermediate frequency signal obtained by the RF front-end processing of the interfered GNSS signal, and then the frequency domain filtering of overlapping and windowed FFT is used to suppress narrow-band interference, and finally The signal is then passed through the airspace adaptive filtering module for broadband interference filtering to complete the purpose of filtering out interference signals. When the above-mentioned frequency-domain suppression technology of overlapping and windowing FFT suppresses narrow-band interference, since the narrow-band interference may be submerged in the broadband interference for a period of time, the peak value of the narrow-band interference cannot be found after the FFT transform. It is easy to be detected and effectively filtered out, and a statistical algorithm with good real-time performance can also be used according to requirements. For the spatial filtering module, a fast adaptive filtering algorithm is used for power inversion filtering, which greatly improves the dynamic performance of the system. This algorithm improves the stability and real-time performance of the frequency-domain-space-domain joint anti-jamming algorithm, expands the applicable range of the frequency-space-space joint anti-jamming algorithm, and improves the system's anti-jamming ability.

Description of drawings

Fig. 1 is the system block diagram that the inventive method depends on;

Fig. 2 is the functional block diagram of the signal preprocessing part in the inventive method;

Fig. 3 is the functional block diagram of the frequency domain filtering part in the method of the present invention;

Fig. 4 is a functional block diagram of the spatial filtering part in the method of the present invention.

Detailed ways

The present invention will be described in detail below in conjunction with the accompanying drawings and specific embodiments.

Referring to Fig. 1, it is the frequency-domain-space-domain joint anti-jamming method assisted by the probability search of the present invention, which relies on three parts of the system, namely the signal preprocessing part, the frequency domain filtering part and the spatial domain filtering part, wherein,

The structure of the signal preprocessing part is, including multi-channel RF front-end, A/D data collector, each A/D data collector is connected with digital down-converter, low-pass filter and decimator in sequence;

The structure of the frequency domain filtering part is that it includes two branches, one of which is composed of a window adder, an FFT processing module, a threshold generation and spectral line processing module, an IFFT processing module, and a 50% delay processing module connected in sequence ; The other branch is composed of 50% delay processing module, window adder, FFT processing module, threshold generation and spectral line processing module, and IFFT processing module connected in sequence, and the I and Q two paths of the two branches are finally simultaneously It is connected with the adder, and the output signal corresponding to the addition is used for the next step of spatial filtering.

The windowing device (Blackman window) adopts the Blackman window function with better sidelobe suppression effect to significantly reduce the leakage of the signal spectrum; the probability statistics are compared with the set threshold according to the statistical value of the position of the maximum value in each frame, If it is greater than the threshold, it is single-frequency interference, and it will be processed;

The structure of the spatial filtering part is composed of a filter structure and an adaptive filtering algorithm. The adaptive filtering algorithm is the innovative gradient variable step size least mean square algorithm (GVSS-NLMS for short) of the present invention.

The joint anti-jamming method in the frequency domain and air domain assisted by the probability search of the present invention relies on the above-mentioned structure and principle, and is implemented according to the following steps:

Step 1. Signal preprocessing

Referring to Figure 2, A/D data acquisition is performed on the intermediate frequency signal obtained after the RF front-end processing, and digital down-conversion processing is performed on the collected data signal; Low-pass filtering, and then extraction processing at the same time, and finally divided into I-channel and Q-channel signals for output for subsequent processing;

Step 2, frequency domain processing

With reference to Fig. 3, the specific steps of frequency domain filter processing are: divide the I road and Q road signal of step 1 last output into two branches and process respectively,

One branch is: 2.11) The I-channel and Q-channel signals are windowed. The introduction of the window function makes the boundary of the truncated sequence obtained by the subsequent fast Fourier transform (FFT transform) smooth, so it can reduce the energy of the subsequent FFT Give way;

2.12) Carry out FFT transformation to the signal output in step 1, output signal spectrum;

2.13) Probabilistic statistical processing is performed on the signal spectrum data to detect the narrowband interference in the spectrum, compare the ratio of the maximum value of the narrowband interference to the wideband interference, and according to whether the position of the maximum value is the same, if the same, add 1 to the number of statistics n, and if it is different, count Number of times n minus 1;

Set an upper limit N _max and a lower limit N _min , if the number of statistics exceeds the upper limit or is lower than the lower limit, n remains unchanged; set two thresholds α and β, n is greater than α (β<α<N _max ) indicating that there is a For single-frequency interference, perform zero-setting processing; if n is less than β (β>N _min ), it is considered that there is no single-frequency interference, and multiple trials are required to obtain a suitable threshold;

2.14) Perform spectral line processing on the spectral data according to the generated threshold;

2.15) performing IFFT transformation (inverse FFT transformation) on the data signal after spectral line processing, and performing IFFT transformation on the signal after frequency domain processing into a time domain signal;

2.16) Perform 50% delay processing on the time domain signal;

Another branch is:

2.21) Delay the signal output by step 1 by 50%;

2.22) Windowing is performed on the I-channel and Q-channel signals. The introduction of the window function makes the boundary of the truncated sequence obtained by the fast Fourier transform (FFT transform) smooth, so the energy leakage of the subsequent FFT can be reduced.

2.23) Carry out FFT transformation, and output the signal spectrum after windowing;

2.24) Probabilistic statistical processing is performed on the signal spectrum data to detect the narrowband interference in the spectrum, compare the ratio of the maximum value of the narrowband interference to the wideband interference, and according to whether the position of the maximum value is the same, if the same, add 1 to the number of statistics n, and if it is different, count Number of times n minus 1;

Set an upper limit N _max and a lower limit N _min , if the number of statistics exceeds the upper limit or is lower than the lower limit, n remains unchanged; set two thresholds α and β, n is greater than α (β<α<N _max ) indicating that there is a For single-frequency interference, perform zero-setting processing; if n is less than β (β>N _min ), it is considered that there is no single-frequency interference, and multiple trials are required to obtain a suitable threshold;

2.25) Perform spectral line processing on the spectral data according to the generated threshold;

2.26) Perform IFFT transformation (inverse FFT transformation) on the data after spectral line processing, and perform IFFT transformation on the signal after frequency domain processing into time domain signal;

2.27) Synthesizing the delayed signal in step 2.16) with the signal obtained in step 2.26) through an adder;

Step 3. Spatial filtering processing

Referring to Figure 4, the spatial power inversion adaptive filtering is performed on the output signal synthesized at the end of step 2, and the broadband interference in the signal is suppressed to obtain the final output signal,

The update formula of the weight of the power inversion adaptive filtering algorithm based on the GVSS-LMS algorithm is:

$\left\{\begin{array}{c}\mathrm{<span\; class="notranslate">\; W</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; W</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; \&mu;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; S</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}{{\mathrm{<span\; class="notranslate">\; S</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}}^{\mathrm{<span\; class="notranslate">\; T</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; /</span><span\; class="notranslate">\; (</span>{{\mathrm{<span\; class="notranslate">\; S</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}}^{\mathrm{<span\; class="notranslate">\; T</span>}}{\mathrm{<span\; class="notranslate">\; S</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ]</span>\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; T</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; W</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; /</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; \&gamma;</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; T</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; )</span>\\ \mathrm{<span\; class="notranslate">\; \&mu;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; \&mu;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; \&mu;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; g</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; g</span>}{<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; T</span>}}\\ \mathrm{<span\; class="notranslate">\; g</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; \&beta;g</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; \&beta;</span>}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; T</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; W</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; /</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; \&gamma;</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; T</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; )</span>\\ \mathrm{<span\; class="notranslate">\; W</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 0</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; 1,0</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; L</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; 0</span>}<span\; class="notranslate">\; ]</span>}^{\mathrm{<span\; class="notranslate">\; T</span>}}\\ {\mathrm{<span\; class="notranslate">\; S</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; =</span>{<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; 1,0</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; L</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; 0</span>}<span\; class="notranslate">\; ]</span>}^{\mathrm{<span\; class="notranslate">\; T</span>}}\end{array}\right.$

Where X(n) is the input signal vector, X ^T (n) is the transpose of the input signal vector, W(n) is the weight vector, μ(n) is the variable step size, s ₀ is the constraint vector, W(0 )s ₀ =1 is to ensure that the first output of spatial filtering is not zero, β is a smooth parameter constant (close to 1), γ is a constant greater than zero, g(n) is a smooth gradient vector, and I is an identity matrix.

The joint anti-jamming method in frequency domain and air domain assisted by probability search of the present invention, its working principle is:

The processing method of mapping the time-domain signal to the frequency domain for interference suppression will make the frequency characteristics of the signal more prominent, thereby obtaining better processing effects. Since narrowband interference and broadband interference coexist and the power of narrowband interference is lower than that of broadband interference, the frequency point of narrowband interference cannot be detected or detected inaccurately after frequency domain filtering FFT. Use the method assisted by probability and statistics to detect narrowband interference. Comparing the energy mean value with the maximum value of narrowband interference can effectively detect the location of narrowband interference and filter it out. Finally, the processed signal is transformed into the time domain by IFFT for further analysis. Spatial filtering. Spatial filtering uses a power inversion algorithm based on linearly constrained least mean square (LCMV) for algorithm iteration, aiming at the minimum power output. The GVSS-NLMS algorithm used in power inversion uses a first-order filter to smooth the gradient vector to reduce the impact of noise, and uses the smoothed gradient vector to update the step size, which can effectively estimate the mean square norm of the coefficient error in real time The minimized optimal step size improves the convergence speed and achieves rapid convergence without reducing the steady-state error. Compared with the existing variable step size LMS algorithm, it has a faster convergence speed and greatly enhances real-time performance.

Claims (4)

1. A joint anti-jamming method in the frequency domain and air domain aided by a probability search is characterized in that it is implemented according to the following steps: Step 1, signal preprocessing; Step 2, frequency domain processing; Step 3, airspace processing. 2. the joint anti-jamming algorithm based on frequency domain and space domain based on probability search according to claim 1, is characterized in that: In the described step 1, carry out A/D data acquisition to the intermediate frequency signal obtained after the radio frequency front-end processing, carry out the processing of digital down-conversion to the data signal collected; Then the I road and the Q road obtained after the digital down-conversion processing The signals are low-pass filtered at the same time, and then extracted at the same time, and finally divided into I-channel and Q-channel signals for output. 3. the joint anti-jamming algorithm based on frequency domain and space domain based on probability search according to claim 1, is characterized in that, In the described step 2, the specific steps of the frequency domain filtering process are: The I-channel and Q-channel signals finally output in step 1 are divided into two branches for processing respectively. One of the branches is: 2.11) Windowing processing is performed on the I-channel and Q-channel signals; 2.12) Carry out FFT transformation to the signal output in step 1, output signal spectrum; 2.13) Probabilistic statistical processing is performed on the signal spectrum data to detect the narrowband interference in the spectrum, compare the ratio of the maximum value of the narrowband interference to the wideband interference, and according to whether the position of the maximum value is the same, if the same, add 1 to the number of statistics n, and if it is different, count The number of times n minus 1; set an upper limit N _max and a lower limit N _min , if the statistical number exceeds the upper limit or is lower than the lower limit, n remains unchanged; set two thresholds α and β, n is greater than α, β<α<N _max , indicating that there is a single-frequency interference, and perform zero-setting processing; if n is less than β, β>N _min , it is considered that there is no single-frequency interference, and a suitable threshold is generated; 2.14) Perform spectral line processing on the spectral data according to the generated threshold; 2.15) performing IFFT transformation on the data signal after spectral line processing, and performing IFFT transformation on the signal after frequency domain processing into a time domain signal; 2.16) Perform 50% delay processing on the time domain signal; Another branch is: 2.21) Delay the signal output by step 1 by 50%; 2.22) Windowing is performed on the I-way and Q-way signals, 2.23) Carry out FFT transformation, and output the signal spectrum after windowing; 2.24) Probabilistic statistical processing is performed on the signal spectrum data to detect the narrowband interference in the spectrum, compare the ratio of the maximum value of the narrowband interference to the wideband interference, and according to whether the position of the maximum value is the same, if the same, add 1 to the number of statistics n, and if it is different, count The number of times n minus 1; set an upper limit N _max and a lower limit N _min , if the statistical number exceeds the upper limit or is lower than the lower limit, n remains unchanged; set two thresholds α and β, n is greater than α, β<α<N _max , indicating that there is a single-frequency interference, and perform zero-setting processing; if n is less than β, β>N _min , it is considered that there is no single-frequency interference, and an appropriate threshold is generated; 2.25) Perform spectral line processing on the spectral data according to the generated threshold; 2.26) performing IFFT transformation on the data after spectral line processing, and performing IFFT transformation on the signal after frequency domain processing into a time domain signal; 2.27) Combining the delayed signal in step 2.16) and the signal obtained in step 2.26) through an adder. 4. the joint anti-jamming algorithm based on frequency domain and space domain based on probability search according to claim 1 is characterized in that: in described step 3, carry out spatial power inversion adaptive filtering to the output signal that step 2 synthesizes at last, to The broadband interference in the signal is suppressed to obtain the final output signal, The update formula of the weight of the power inversion adaptive filtering algorithm based on the GVSS-LMS algorithm is: $\left\{\begin{array}{c}W(\mathrm{no}+1)=W\left(\mathrm{no}\right)-2\mathrm{\&mu;}\left(\mathrm{no}\right)[I-\left(S_0{S_0}^T\right)/\left({S_0}^T{S}_0\right)]x\left(\mathrm{no}\right)x^T\left(\mathrm{no}\right)W\left(\mathrm{no}\right)/(\mathrm{\&gamma;}+x^T\left(\mathrm{no}\right)x\left(\mathrm{no}\right))\\ \mathrm{\&mu;}\left(\mathrm{no}\right)=1-\mathrm{\&mu;}(\mathrm{no}-1)+\mathrm{\&mu;}(\mathrm{no}-1)g\left(\mathrm{no}\right)g{(\mathrm{no}-1)}^T\\ g\left(\mathrm{no}\right)=\mathrm{\&beta;g}(\mathrm{no}-1)+(1-\mathrm{\&beta;})x^T\left(\mathrm{no}\right)W\left(\mathrm{no}\right)/(\mathrm{\&gamma;}+x^T\left(\mathrm{no}\right)x\left(\mathrm{no}\right))\\ W\left(0\right)={[\mathrm{1,0},L,0]}^T\\ S_0={[\mathrm{1,0},L,0]}^T\end{array}\right.$ Where X(n) is the input signal vector, X ^T (n) is the transpose of the input signal vector, W(n) is the weight vector, μ(n) is the variable step size, s ₀ is the constraint vector, W(0 )s ₀ =1 is to ensure that the first output of spatial filtering is not zero, β is a smoothing parameter constant, γ is a constant greater than zero, g(n) is a smooth gradient vector, and I is an identity matrix.