Joint detection method based on suppression jamming and multi-dimensional difference characteristics of target

Abstract

The present application relates to a kind of joint detection method based on suppressive jamming and target multidimensional difference characteristic, for passive sonar target, the method constructs frequency domain focusing matrix to frequency domain base array data is weighted, obtain the matrix of directional power spectral density;Based on the minimum variance criterion of guidance, construct adaptive beam forming optimal weight coefficient, obtain the beam domain frequency domain data of adaptive weighting;All beams are traversed, extract multidimensional difference characteristic, and obtain joint detection result based on multidimensional difference characteristic.The present application improves the anti-interference processing gain of weak target under suppressive jamming, effectively reduces the blind area range of suppressive jamming, greatly improves the detection efficiency of passive sonar target under suppressive underwater acoustic countermeasure jamming environment;The method is verified by simulation and sea trial data, and the weak target detection ability of the present application under suppressive jamming is obviously superior to traditional energy detection method, and effectively reduces the detection blind area range under suppressive jamming.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of sonar signal processing, and particularly relates to a joint detection method based on suppression jamming and multi-dimensional difference characteristics of targets, which is suitable for weak target detection under water acoustic counter-jamming. BACKGROUND

[0002] Suppression water acoustic jamming mainly suppresses and reduces the detection performance of sonar and torpedoes on targets, and is mainly equipped with water acoustic jamming devices, etc. According to the working frequency band, there are high-frequency acoustic jamming devices and low-frequency acoustic jamming devices. By emitting strong power noise into the water, the normal work of the opponent's sonar and torpedo acoustic homing is suppressed, aiming to weaken or destroy the acoustic detection performance, and improve the survival ability of the own ship platform.

[0003] At the same time, the target detection technology under water acoustic counter-jamming is a processing method for suppressing / eliminating the influence caused by water acoustic counter-jamming. The current development of water acoustic counter-jamming technology has made the existing sonar detection performance decrease sharply. Under the condition of strong suppression jamming radiation noise energy, not only the background noise will be increased, but also a large fan-shaped detection blind area will be formed, which will greatly reduce the sonar action distance and effective detection range, and reduce the parameter estimation accuracy of target azimuth, distance, etc.

[0004] In the prior art, the passive target detection method is generally based on energy detection, which is easily affected by strong suppression jamming. Although the existing high-resolution adaptive detection method can suppress the sidelobe leakage problem of strong jamming, it still has problems such as strong main lobe energy and wide beam, and the detection range is still limited. SUMMARY

[0005] The present application solves the problems in the prior art and provides an optimized joint detection method based on suppression jamming and multi-dimensional difference characteristics of targets. Starting from the characteristics of suppression jammer equipment, based on the characteristics that the frequency band range of suppression jamming is limited, the spectrum is uniform white and cannot cover low-frequency information, etc., the joint detection method is proposed by jointly using the difference characteristics such as high-low frequency energy proportion characteristics, narrowband line spectrum characteristics and spatial energy distribution.

[0006] The technical scheme adopted by the present application is a joint detection method based on suppression jamming and multi-dimensional difference characteristics of targets. For passive sonar targets, the method constructs a frequency domain focusing matrix to perform weighted processing on frequency domain array data to obtain a steering power spectral density matrix; an adaptive beam forming optimal weight coefficient is constructed based on a steering minimum variance criterion to obtain adaptive weighted beam domain frequency domain data; all beams are traversed, multi-dimensional difference characteristics are extracted, and a joint detection result is obtained based on the multi-dimensional difference characteristics.

[0007] Preferably, the construction of the frequency domain focusing matrix comprises the following steps:

[0008] Step 1.1: calculating array steering vector a n (f k ,θ);

[0009] Step 1.2: compensating phase of each channel frequency domain array data using array steering vector, obtaining compensated array output matrix Y(f k ,θ)。

[0010] Preferably, in the step 1.1, wherein a n (f k ,θ) is a direction vector of the nth array element at processing frequency f k , scan direction θ, n = 1, 2, …, N, N is the total number of array elements of the linear array, d is the array element spacing, c is the sound speed, f k ∈ [fL, fH], fL and fH are the upper and lower limits of the processing frequency band, and j is an imaginary number;

[0011] In the step 1.2, Y(f k ,θ) = T Η (f k ,θ)X(f k ),

[0012] wherein, is a diagonal matrix composed of array flow vector corresponding to f k frequency point and θ direction, X(f k ) is frequency domain array data, (·) Η is the conjugate transpose of a complex matrix.

[0013] Preferably, the steering power spectral density matrix is wherein R(f k ) is a covariance matrix of f k frequency point, M is the number of frequency domain array data snapshots, m is a positive integer from 1 to M, and K is the total number of processing frequency points.

[0014] Preferably, the optimal weight coefficient of adaptive beamforming is constructed based on the steering minimum variance criterion wherein, is the inverse of the steering power spectral density matrix, and I is a unit matrix.

[0015] Preferably, the beam output is Y(f k ,θ) = w Η (θ)X(f k ).

[0016] Preferably, the multi-dimensional difference feature includes a high-low frequency energy difference feature, a normalized narrowband line spectrum feature vector and a spatial energy distribution feature, and the multi-dimensional feature joint detection result is a product of the multi-dimensional difference features. F (θ)=α eLH (θ)•α Fline (θ)•α Beam (θ)。

[0017] Preferably, based on the beam-domain frequency-domain signal Y(f k ,θ), all beams are traversed to extract the high-low frequency energy difference feature, α eLH (θ)=E L (θ) / E H (θ), wherein the low frequency energy high frequency energy f o is a demarcation line for dividing the low frequency and the high frequency.

[0018] Preferably, based on the beam-domain frequency-domain signal Y(f k ,θ), all beams are traversed to extract the narrowband line spectrum, to obtain the number of narrowband line spectrums of each beam, and the normalized narrowband line spectrum feature vector α Fline (θ)=c1+(N Fline (θ)-min(N Fline ))·d1 / (max(N Fline )-min(N Fline )) is obtained by using the efficacy coefficient method, wherein c1=25 and d1=75 are the efficacy coefficient method parameters, N Fline (θ) is the narrowband line spectrum number vector, and max(·) and min(·) are maximum and minimum functions respectively.

[0019] Preferably, based on the beam-domain frequency-domain signal Y(f k ,θ), all beams are traversed to obtain the spatial energy distribution feature

[0020] The application relates to an optimized joint detection method based on suppression jamming and target multi-dimensional difference features, for a passive sonar target, the method constructs a frequency-domain focusing matrix to perform weighted processing on frequency-domain array data, to obtain a directional power spectrum density matrix; constructs an adaptive beam forming optimal weight coefficient based on a directional minimum variance criterion, to obtain adaptive weighted beam-domain frequency-domain data; traverses all beams to extract multi-dimensional difference features, and obtains a joint detection result based on the multi-dimensional difference features.

[0021] Compared with the traditional energy detection method, the present invention improves the anti-interference processing gain of weak targets under suppression interference, effectively reduces the blind area of ​​suppression interference, and greatly improves the detection efficiency of passive sonar targets in the suppression underwater acoustic counter-interference environment; this method is verified by simulation and sea trial test data. The method is significantly better than the traditional energy detection method in weak target detection capability under suppression interference, and effectively reduces the detection blind area under suppression interference. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Attachment Figure 1 is a flow chart of the method of the present invention;

[0023] Figure 2 To simulate the multi-dimensional feature joint detection results of a 64-element, 1.5m-spaced uniform linear array sonar under suppression interference;

[0024] Figure 3 The data processing results of the sea trial for passive target detection under suppression jamming are shown. DETAILED DESCRIPTION

[0025] The present invention is further described in detail below with reference to the embodiments, but the protection scope of the present invention is not limited thereto.

[0026] The present invention relates to a joint detection method based on suppressive interference and multidimensional difference characteristics of a target. For a passive sonar target, the method constructs a frequency domain focusing matrix to perform weighted processing on frequency domain array data to obtain a guided power spectrum density matrix; constructs optimal weight coefficients for adaptive beamforming based on a guided minimum variance criterion to obtain adaptively weighted beam domain frequency domain data; traverses all beams to extract multidimensional difference characteristics, and obtains a joint detection result based on the multidimensional difference characteristics.

[0027] (1) Calculate the array steering vector: a n (f k ,θ) is the processing frequency f of the nth array element k , the direction vector at the scanning azimuth θ, n=1,2,…,N, N is the total number of array elements, d is the array element spacing, c is the sound speed, f k ∈[fL,fH], fL, fH are the upper and lower limits of the processing frequency band, j is an imaginary number, -j2πf k ndcosθ / c is the phase information of the nth array element.

[0028] (2) Use the array steering vector to perform phase compensation on the frequency domain array data of each channel to obtain the compensated array output matrix Y(f k ,θ)=T Η (f k ,θ)X(f k ),

[0029] where (·) Η is the conjugate transpose of a complex matrix, X(f k ) is the frequency domain array data, T(f k , θ) is a diagonal matrix composed of array flow vector corresponding to f k frequency point and θ direction, called focusing matrix, defined as:

[0030]

[0031] (3) Focus on the frequency domain array data using the focusing matrix to get the steering power spectral density matrix,

[0032] where R(f k ) is the covariance matrix of f k frequency point, where M is the frequency domain data snapshot number, m is a positive integer from 1 to M, and K is the total frequency point number.

[0033] (4) Calculate the optimal weight vector and beam output of adaptive waveform recovery

[0034] Optimal weight vector

[0035] Beam output Y(f k , θ) is Y(f k , θ) = w Η (θ)X(f k ),

[0036] where, is the inverse of the steering power spectral density matrix, and I is the unit matrix.

[0037] (5) Based on the beam domain frequency domain signal Y(f k , θ), traverse all beams, extract the high and low frequency energy difference feature, α eLH (θ) = E L (θ) / E H (θ),

[0038] where, low frequency energy high frequency energy f o is the low and high frequency demarcation line.

[0039] (6) Based on the beam domain frequency domain signal Y(f k , θ), traverse all beams, extract narrowband line spectrum, get the number of narrowband line spectrum of each beam, use the efficacy coefficient method to get the normalized narrowband line spectrum feature vector, α Fline (θ) = c1+ (N Fline (θ) - min(NFline ))·d1 / (max(N Fline )-min(N Fline )),

[0040] where c1=25, d1=75 are the efficacy coefficient method parameters, N Fline is the narrowband line spectrum number vector, max(·) and min(·) are the maximum and minimum functions respectively.

[0041] (7) Based on the beam domain frequency domain signal Y(f k , θ), all beams are traversed to obtain the spatial energy distribution feature,

[0042] (8) The high and low frequency energy difference feature, the narrowband line spectrum feature and the spatial energy distribution feature are jointly utilized to obtain a multi-dimensional feature joint detection result, B F (θ) = α eLH (θ)·α Fline (θ)·α Beam (θ).

[0043] The embodiments of the application are as follows Figure 2 , 3 .

[0044] As Figure 2 , the simulation 64-element, 1.5m interval uniform linear array sonar in the multi-dimensional feature joint detection result under the suppression type jamming; the processing frequency is 20-500Hz, the equal angle scanning, the simulation target direction is near 70°, the input signal to noise ratio is -5dB, the simulation suppression type jamming direction is near 80°, and the input signal to noise ratio is 10dB;

[0045] Wherein, (a) is a spatial energy distribution feature, a weak target is suppressed in the spatial energy by a strong jamming at 70° direction;

[0046] (b) is a high and low frequency energy proportion feature, the feature vector forms a clear null at the strong jamming direction;

[0047] (c) is a narrowband line spectrum feature;

[0048] (d) is a multi-dimensional feature joint detection spatial spectrum result;

[0049] (e) is a traditional energy detection direction history, wherein the weak target track is suppressed by the sidelobe of the strong jamming and cannot be detected;

[0050] (f) is a multi-dimensional feature joint detection direction history, it can be seen from (d) that the strong jamming is suppressed obviously, the weak target detection output is greatly enhanced, the signal to jamming ratio changes from the input -15dB to 2dB, and the jamming suppression ratio can be improved by 17dB.

[0051] As Figure 3 shown, the processing results of passive target detection sea trial data under the suppression jamming, the sonar array is a 128-element half-wavelength linear array, and the 0-180° cosine scanning is used.

[0052] Among them, (a) is the traditional regular energy detection result;

[0053] (b) is the traditional adaptive energy detection result, the suppression jamming is located at 60° azimuth, and the weak target is located at 80° azimuth. It can be found that the weak target track is completely covered in the traditional energy detection result, especially the sidelobe influence of the regular energy detection of the strong jamming is more serious;

[0054] (c) is the processing result of the multi-dimensional feature joint detection method. It can be found that the suppression strong jamming is obviously suppressed, the blind area influence range is greatly reduced, and the weak target track is clear and distinguishable;

[0055] (d) is the spectral feature of the suppression jamming and the weak target. It can be found that the energy of the suppression jamming is concentrated in the medium and high frequencies, and the low frequency cannot be covered, and there is no obvious line spectrum information;

[0056] (e) and (f) are the spatial spectrum snapshot results at different times. By comparing the output energy of the suppression jamming and the output energy of the weak target, it can be found that the interference suppression ratio of the joint detection method based on multi-dimensional features reaches about 20 dB, and the weak target detection ability is better than that of the traditional energy detection method.

Claims

1. A joint detection method based on suppressive interference and multi-dimensional target difference characteristics, characterized by: For passive sonar targets, the method constructs a frequency domain focusing matrix to perform weighted processing on frequency domain array data to obtain a guided power spectrum density matrix; constructs an adaptive beamforming optimal weight coefficient based on a guided minimum variance criterion to obtain adaptively weighted beam domain frequency domain data; traverses all beams to extract multidimensional difference features, the multidimensional difference features including high- and low-frequency energy difference features, standardized narrowband line spectrum feature vectors and spatial energy distribution features; traverses all beams based on beam domain frequency domain signals to extract high- and low-frequency energy difference features; traverses all beams based on beam domain frequency domain signals to extract narrowband line spectra to obtain the number of narrowband line spectra of each beam; utilizes an efficiency coefficient method to obtain standardized narrowband line spectrum feature vectors; traverses all beams based on beam domain frequency domain signals to obtain spatial energy distribution features; A joint detection result is obtained based on the multi-dimensional difference features, and the multi-dimensional feature joint detection result is the product of the multi-dimensional difference features.

2. The joint detection method based on suppressive interference and target multi-dimensional difference characteristics according to claim 1, characterized in that: Constructing the frequency domain focusing matrix includes the following steps: Step 1.1: Calculate the array steering vector a n (f k , θ); Step 1.2: Use the array steering vector to perform phase compensation on the frequency domain array data of each channel to obtain the compensated array output matrix Y(f k , θ).

3. The joint detection method based on suppressive interference and target multi-dimensional difference characteristics according to claim 2, characterized in that: In step 1.1, ,in, For the The array element processes the frequency , Scan direction The direction vector at , N is the total number of linear array elements, is the array element spacing, is the speed of sound, , To process the upper and lower limits of the frequency band, is an imaginary number; In step 1.2, , in, , corresponding to Frequency, The diagonal array of popular vectors corresponding to the direction, is the frequency domain array data, is the conjugate transpose of the complex matrix.

4. The joint detection method based on suppressive interference and target multi-dimensional difference characteristics according to claim 3, characterized in that: The guided power spectral density matrix is ,in, for The covariance matrix of the frequency points, , is the number of snapshots of frequency domain array data, m is a positive integer from 1 to M, To process the total number of frequency points.

5. The joint detection method based on suppressive interference and target multi-dimensional difference characteristics according to claim 4, characterized in that: Constructing optimal weight coefficients for adaptive beamforming based on the steered minimum variance criterion ,in, is the inverse of the oriented power spectral density matrix, For the unit array.

6. The joint detection method based on suppressive interference and target multi-dimensional difference characteristics according to claim 5, characterized in that: The beam output is .

7. The joint detection method based on suppressive interference and target multi-dimensional difference characteristics according to claim 1, characterized in that: 。 8. The joint detection method based on suppressive interference and target multi-dimensional difference characteristics according to claim 7, characterized in that: Based on beam domain frequency domain signal , traverse all beams and extract the high and low frequency energy difference characteristics, , where low-frequency energy , high-frequency energy , It is the dividing line between low frequency and high frequency.

9. The joint detection method based on suppressive interference and target multi-dimensional difference characteristics according to claim 7, characterized in that: Based on beam domain frequency domain signal , traverse all beams, extract narrowband line spectra, obtain the number of narrowband line spectra of each beam, and use the power coefficient method to obtain the standardized narrowband line spectrum feature vector ,in 、 is the parameter of the power coefficient method, is the number vector of narrowband line spectra, and are the maximum and minimum functions respectively.

10. The joint detection method based on suppressive interference and target multi-dimensional difference characteristics according to claim 7, characterized in that: Based on beam domain frequency domain signal , traverse all beams and obtain spatial energy distribution characteristics .

Images