CN103837871B - A kind of inverse beamforming method and system - Google Patents

Abstract

The present invention relates to a kind of inverse beamforming method, comprising: discrete Fourier transformation is done to received signal, obtain the frequency domain value of signal; Be that reception drags linear array to calculate cross-spectrum matrix according to frequency domain value; The main diagonal element of cross-spectrum matrix is revised; Toeplitz being done to the cross-spectrum matrix through revising average, being translated into the one dimension distributed data with space distribution: one dimension distributed data is compensated at the enterprising line phase of θ angle, then sues for peace, obtain the power spectrum in θ angle; In 0 ~ 179 °, spacescan is done to power spectrum, obtains the spatial spectrum P ˊ at current signal frequency f place _{dL_IBF}(f); Make P _{dL_IBF}=P _{dL_IBF}+ P ˊ _{dL_IBF}f (), f=f+1, as f≤f _maxtime, step before repetition, otherwise, end operation, thus complete broadband inverse beamforming.

Description

Reverse beam forming method and system

Technical Field

The invention relates to the field of sonar signal processing, in particular to a reverse beam forming method and system.

Background

With the continuous improvement of sonar technology, the energy of target radiation signals is lower and lower, and the signal-to-noise ratio (SNR) of signals received by each array element in the receiving towed array is lower and lower, so that the display effect of space spectrum data directly obtained by inverse beam forming (invert block area mimo, IBF) in an azimuth history chart is poor, and even the target azimuth cannot be well displayed.

Inverse beamforming requires the use of direction vector data and cross-spectral matrix data between array elements. Compared with other position elements, the main diagonal elements in the cross-spectrum matrix have large noise energy and basically the same signal energy, so the main and side lobe energy difference of the spatial spectrum obtained by beam forming under low signal-to-noise ratio is smaller, and the display effect of the spatial spectrum in the azimuth history chart is poor. Aiming at the defect of the existing inverse beamforming algorithm, in a new background equalization algorithm [ J ] acoustics report, 2000,25(1):5-9 "in reference 1" li-qi, pan-bao, Yi-li-digital sonar and in reference 2 "eweda-globalsibilizationftheastmaanfourthalgorithm [ J ]. trans-activity synthesis algorithm, 2012,60(3): 1473-1477", the technicians in the field adopt post-processing methods such as background equalization or gray level transformation on the spatial spectrum obtained by beamforming, so as to improve the display effect of the spatial spectrum, so that the workers can conveniently and intuitively obtain the target from the course azimuth map under low signal-to-noise ratio. The post-processing method has a good effect, but has the defects of complex algorithm, large computation amount and the like, and is not beneficial to realizing real-time processing.

Disclosure of Invention

The invention aims to overcome the defects of complex algorithm and large operation amount of the existing inverse beam forming method, thereby providing a simple and effective inverse beam forming method and system.

In order to achieve the above object, the present invention provides a reverse beam forming method comprising:

step 1), receiving signals by each array element in the towed linear array, performing discrete Fourier transform on the received signals to obtain frequency domain values of the signals, and obtaining a spatial spectrum P_{DL_IBF}Setting an initial value, and setting the frequency to f ═ f_min，f_minIs the lower frequency limit;

step 2), calculating a cross-spectrum matrix for the receiving array according to the frequency domain value of the signal received by each array element obtained in the step 1);

step 3), correcting main diagonal elements of the cross spectrum matrix obtained in the step 2);

step 4), conducting Toeplitz averaging on the modified cross-spectrum matrix of the main diagonal elements obtained in the step 3), and converting the Toeplitz averaging into one-dimensional distribution data with spatial distribution:

step 5), performing phase compensation on the one-dimensional distribution data obtained in the step 4) at the angle theta, and then summing to obtain a power spectrum at the angle theta; the theta angle is an included angle between a target radiation signal and a horizontal line where the array elements are located;

step 6), carrying out space scanning on the power spectrum obtained in the step 5) within 0-179 degrees to obtain a space spectrum P 'at the current signal frequency f'_{DL_IBF}(f) (ii) a Wherein the direction from the array element 1 to the array element N is 0 degree, and the anticlockwise rotation direction is 0-179 degree;

step 7), let P_{DL_IBF}＝P_{DL_IBF}+P'_{DL_IBF}(f) F is f +1, when f is less than or equal to f_maxIf so, repeating the step 2) to the step 6), otherwise, finishing the operation, thereby finishing the broadband inverse beam forming; wherein f is_maxIs the upper frequency limit.

In the above technical solution, in the step 1), the method further includes: after receiving the array element receiving signal in the dragging line array, FFT conversion is carried out on the broadband signal in the received signal, and therefore the operation of the narrow-band signal is obtained.

In the above technical solution, in the step 3), the modifying the main diagonal elements of the cross spectrum matrix includes setting the values of the main diagonal elements to zero.

In the above technical solution, in the step 3), the modifying the main diagonal element of the cross spectrum matrix includes setting a value of the main diagonal element as a constant having an absolute value smaller than 5.

The invention also provides a reverse beam forming system, which comprises a preprocessing module, a cross-spectrum matrix calculation module, a correction module, a Toeplitz average module, a power spectrum calculation module, a space spectrum calculation module and a circular execution module; wherein,

the preprocessing module is used for receiving the dragging linear arrayThe signal received by each array element is subjected to discrete Fourier transform to obtain the frequency domain value of the signal, and the frequency domain value is a space spectrum P_{DL_IBF}Setting an initial value, and setting the frequency to f ═ f_min，f_minIs the lower frequency limit;

the cross spectrum matrix calculation module calculates a cross spectrum matrix for the receiving array according to the frequency domain value of the signal received by each array element obtained by the preprocessing module;

the correction module corrects main diagonal elements of the cross spectrum matrix obtained by the cross spectrum matrix calculation module;

the Toeplitz averaging module performs Toeplitz averaging on the modified cross-spectrum matrix of the main diagonal elements obtained by the modification module, and converts the Toeplitz averaging into one-dimensional distribution data with spatial distribution:

the power spectrum calculation module performs phase compensation on the one-dimensional distribution data obtained by the Toeplitz averaging module at an angle theta, and then sums the one-dimensional distribution data to obtain a power spectrum at the angle theta; the theta angle is an included angle between a target radiation signal and a horizontal line where the array elements are located;

the space spectrum calculation module performs space scanning on the power spectrum obtained by the power spectrum calculation module within 0-179 degrees to obtain a space spectrum P 'at the current signal frequency f'_{DL_IBF}(f) (ii) a Wherein the direction from the array element 1 to the array element N is 0 degree, and the anticlockwise rotation direction is 0-179 degree;

the loop execution module order P_{DL_IBF}＝P_{DL_IBF}+P'_{DL_IBF}(f) F is f +1, when f is less than or equal to f_maxWhen the broadband inverse beam is formed, restarting the cross-spectrum matrix calculation module, otherwise, finishing the operation, thereby completing the broadband inverse beam forming; wherein f is_maxIs the upper frequency limit.

In the above technical solution, the apparatus further includes an FFT conversion module, where the FFT conversion module performs FFT conversion on a wideband signal in signals received by each array element in the receiving array to obtain a narrowband signal, and transmits the obtained narrowband signal to the preprocessing module.

In the above technical solution, the modifying module modifies the main diagonal elements of the cross-spectrum matrix, including setting the values of the main diagonal elements to zero.

In the above technical solution, the modifying module modifies the main diagonal element of the cross-spectrum matrix, including setting a value of the main diagonal element as a constant having an absolute value less than 5.

The invention has the advantages that:

1. the method of the invention can weaken the influence of noise contained in the main diagonal elements on the beam forming result, especially the condition that the noise energy on the main diagonal elements far exceeds the signal energy under the low signal-to-noise ratio.

2. After verification: compared with the existing IBF method, the space spectrum obtained by the inverse beam forming ((DIAGONALIBF: DLIBF) after the main diagonal elements are corrected by adopting the method of the invention has more stable side lobe and better display effect.

3. The method is simple and effective, and can meet the requirements of practical engineering application.

Drawings

Fig. 1 is a schematic diagram of a receiving towed array in accordance with the present invention;

FIG. 2 is a schematic diagram of the inverse beamforming method of the present invention;

FIG. 3(a) is P₂(iv) 0v, a spatial spectrogram snapshot obtained by the method of the invention and the prior art method under the condition of SNR-25 dB;

FIG. 3(b) is P₂0v, spatial spectral azimuth history obtained by prior art method under SNR-25 dB;

FIG. 3(c) is P₂The spatial spectrum azimuth history chart obtained by the method of the invention under the condition that SNR is-25 dB is 0 v;

FIG. 4(a) is P₂(iv) 0.5v, a spatial spectrogram snapshot obtained by the method of the invention and the method of the prior art under the condition that SNR is-23 dB;

FIG. 4(b) is P₂0.5v, spatial spectral azimuth history obtained by prior art method under SNR-23 dB;

FIG. 4(c) is P₂The spatial spectrum azimuth history chart obtained by the method of the invention under the condition that SNR is-23 dB is 0.5 v;

FIG. 5(a) is P₂1.0v, under the condition that SNR is-23 dB, obtaining a space spectrogram instantaneous graph by the method of the invention and the method of the prior art;

FIG. 5(b) is P₂1.0v, under the condition of SNR-23 dB, the space spectrum azimuth history chart obtained by the prior art method;

FIG. 5(c) is P₂1.0v, under the condition that SNR is-23 dB, the space spectrum azimuth history chart obtained by the method of the invention.

Detailed Description

The invention will now be further described with reference to the accompanying drawings.

Before describing the method of the present invention in detail, a description will be given of a receiving towed array to which the method of the present invention is applied. Fig. 1 is a schematic diagram of a receiving array, which is an equally spaced horizontal array with an array element number N, and a target radiates a signal from a direction θ and reaches the array element after being propagated through an underwater acoustic channel. Taking the receiving towed array as an example, the method of the present invention will be described in detail below.

Referring to fig. 2, the method of the present invention comprises the steps of:

step 1), receiving signals by each array element in the towed linear array, performing discrete Fourier transform on the received signals, setting initial values for a spatial spectrum,setting the frequency to f ═ f_min，f_minThe lower frequency limit.

The signals received by each array element can be described by equation (1):

x_i(t)＝P_i·s(t-Δt_i)+n_i(t)(1)

wherein: s (t) is the signal radiated by the target at time t, P_iFor the amplitude of the received signal of the ith array element, i is the ith array element, Δ t_iD ═ f ═ dcos θ c is the delay difference between the ith and 1 st array elements_cC is the array spacing, f_cIs the center frequency of the received signal, c is the average speed of sound, n_iAnd (t) is the background noise received by the ith array element at the time t.

Obtaining the frequency domain value X of the signal by discrete Fourier transform of the received signal_i(f) The frequency range of the signal is f_min≤f≤f_maxIn this step, the frequency of the signal is f_minWherein: f. of_minIs the lower limit of frequency, f_maxIs the upper frequency limit. Simultaneous output of spatial spectrum P_{DL_IBF}＝0。

In this embodiment, the signal received by each array element is a narrowband signal. In other embodiments, the signal received by each array element may also be a wideband signal, and in step 1), after the array element receives the signal, the FFT of the wideband signal is further included, so as to obtain a narrowband signal.

And step 2), calculating a cross-spectrum matrix for the receiving towed linear array.

The cross spectrum matrix among the array elements to be calculated is represented by R, and the expression is as follows:

where denotes the conjugate transpose.

And 3) correcting main diagonal elements of the cross-spectrum matrix obtained in the step 2).

In this embodiment, the correction includes zeroing the value of the main diagonal element, i.e.:

R(i,i)＝0(3)

wherein: and i is less than or equal to N, and N is the number of array elements.

In other embodiments, modifying the primary diagonal elements of the cross-spectrum matrix further comprises setting the values of the primary diagonal elements to a constant having an absolute value less than 5.

Step 4), conducting Toeplitz averaging on the modified cross-spectrum matrix R of the main diagonal elements obtained in the step 3), and converting the Toeplitz averaging into one-dimensional distribution data with spatial distribution:

$D\left(n\right)=\frac{1}{N-\left|n\right|}\underset{m=0}{\overset{N-1\left|n\right|}{\mathrm{\Σ}}}R(m,m+n)---\left(4\right)$

wherein: n is not less than N-1, m is not less than 0 and not more than N-1-N.

And 5) carrying out phase compensation on the one-dimensional distribution data obtained in the step 4) on the angle theta, and then summing to obtain a power spectrum on the angle theta.

The associated calculation is as follows:

$\mathrm{FIM}(f,\mathrm{\θ})=\underset{n=1-N}{\overset{N-1}{\mathrm{\Σ}}}D\left(n\right)e^{-\mathrm{jndk}\cos \mathrm{\θ}}---\left(5\right)$

and the angle theta is an included angle between the target radiation signal and a horizontal line where the array elements are located.

Step 6), spatially scanning the power spectrum obtained in the step 5) at 0-179 degrees (the direction from array element 1 to array element N is 0 degrees, and the counterclockwise rotation direction is 0-179 degrees), and obtaining a spatial spectrum P 'at the frequency f'_{DL_IBF}(f)。

Step 7), let P_{DL_IBF}＝P_{DL_IBF}+P'_{DL_IBF}(f) F is f +1, when f is less than or equal to f_maxAnd then, repeating the step 2) to the step 6) to finish the broadband inverse beam forming.

The above is a description of the basic steps of the method of the present invention, which are further described below.

In step 3), the main diagonal elements of the cross-spectrum matrix between the array elements are corrected. The necessity of such a correction is set forth below.

As is known, each array element in the receiver array receives a signal, and the received signal is represented as follows:

$X\left(t\right)=\left[\begin{array}{c}{x}_1\left(t\right)\\ x_2\left(t\right)\\ .\\ .\\ .\\ x_N\left(t\right)\end{array}\right]=\left[\begin{array}{c}{s}_1\left(t\right)+n_1\left(t\right)\\ s_2\left(t\right)+n_2\left(t\right)\\ .\\ .\\ .\\ s_N\left(t\right)+n_N\left(t\right)\end{array}\right]---\left(6\right)$

wherein: x (t) is the data stream, and N is the number of array elements.

The cross-spectrum matrix R can be derived in the time domain as follows:

$R=X\left(t\right)X^H\left(t\right)=\left[\begin{array}{c}{s}_1\left(t\right)+n_1\left(t\right)\\ s_2\left(t\right)+n_2\left(t\right)\\ .\\ .\\ .\\ s_N\left(t\right)+n_N\left(t\right)\end{array}\right]\·{[s}_1\left(t\right)+n_1\left(t\right),s_2\left(t\right)+n_2\left(t\right),...s_N\left(t\right)+n_N\left(t\right)]$

according to the formula (7), the total energy Ps of all signals, the total energy Pn of all noises, the energy Psn obtained by correlating the signals and the noises, the total energy Pow, the signal energy DL _ Ps contained in the main diagonal elements and the noise energy DL _ Pn in the cross spectrum matrix R are as follows:

$\left\{\begin{array}{c}\mathrm{Ps}=\underset{j=1}{\overset{N}{\mathrm{\Σ}}}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{s}_j\left(t\right)s_i^{*}\left(t\right)\\ \mathrm{Pn}=\underset{j=1}{\overset{N}{\mathrm{\Σ}}}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{n}_j\left(t\right)n_i^{*}\left(t\right)\\ \mathrm{Psn}=\underset{j=1}{\overset{N}{\mathrm{\Σ}}}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{s}_j\left(t\right)n_i^{*}\left(t\right)\\ \mathrm{Pow}=\mathrm{Ps}+\mathrm{Pn}+\mathrm{Psn}\\ \mathrm{DL}\_\mathrm{Ps}=\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{s}_i\left(t\right)s_i^{*}\left(t\right)\\ \mathrm{DL}\_\mathrm{Pn}=\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{n}_i\left(t\right)n_i^{*}\left(t\right)\end{array}\right.---\left(8\right)$

as can be seen from equation (8), when the background noise is white gaussian noise, the correlation between the background noise signals of different array elements is small, and the correlation between the signals of different array elements is large, that is: the background noise energy contained in each element on the main diagonal of the cross-spectrum matrix R is far larger than that contained in other elements, and the signal energy contained in each element on the main diagonal is similar to that contained in other elements. This conclusion can be expressed as follows:

$\left\{\begin{array}{c}{n}_j\left(t\right)n_t^{*}\left(t\right)<<n_i\left(t\right)n_i^{*}\left(t\right)\\ s_j\left(t\right)s_i^{*}\left(t\right)\&ap;s_i\left(t\right)s_i^{*}\left(t\right)\end{array},(j\&NotEqual;i,1\&le;i,j\&le;N)---\left(9\right)\right.$

at low signal-to-noise ratios, it can further be known that:

$s_i\left(t\right)s_i^{*}\left(t\right)<<n_i\left(t\right)n_t^{*}\left(t\right),(1\&le;i\&le;N)---\left(10\right)$

at low signal-to-noise ratios, it is possible to obtain from equations (8) - (11):

$\left\{\begin{array}{c}\mathrm{DL}\_\mathrm{Ps}\&ap;\frac{\mathrm{Ps}}{N}<<\mathrm{Ps}\\ \mathrm{DL}\_\mathrm{Ps}<<\mathrm{Pn}\\ \mathrm{DL}\_\mathrm{Pn}\&ap;\mathrm{Pn}\end{array}\right.---\left(12\right)$

it can be seen from this that: under the condition of low signal-to-noise ratio, the noise energy contained in the main diagonal elements in the cross-spectrum matrix R is larger and far exceeds the signal energy. Based on this conclusion, the main diagonal elements of the cross-spectrum matrix are modified in step 3) so as to reduce the influence of noise contained in the main diagonal elements on the inverse beamforming.

In order to further verify the above conclusion, the proportion of noise and signal energy contained in the main diagonal elements in the cross-spectrum matrix R in the total energy can be calculated.

And when the specific gravity of noise and signal energy contained in main diagonal elements in the cross-spectrum matrix R in the total energy is calculated, MATLAB numerical simulation can be adopted.

In a simulation process, the frequency is f₁Taking a sinusoidal signal of 500Hz as a target radiation signal, wherein the azimuth angle of the target radiation signal relative to the horizontal linear array is theta₁60 °; the phase difference between other array elements relative to the reference array element is reversely deduced according to the acoustic path difference between the target and other array elements and the target and reference array elements and is added to the received signals of each array element. Further, the background noise is made to be isotropic white Gaussian noise, and the frequency band is set to be 100 to 800 Hz. The number of horizontal linear array elements N is 16, 32, 64, the array spacing isWherein c is 1500m/s as effective sound velocity, and sampling rate is f_s5kHz, and 2000 quick beat. The data in table 1 below are the results obtained from 200 independent statistics according to equation (8).

TABLE 1 weight ratio of DL _ Ps and DL _ Pn in the Total energy of the Cross-spectral matrix R

When the signal and noise, and the noise and noise correlation are small, it can be seen from the data in table 1 that: when the SNR of the low signal-to-noise ratio is less than-10 dB, DL _ Pn is more than or equal to 10DL _ Ps; when N is 16 and SNR is less than-16 dB, $\frac{\mathrm{DL}\_\mathrm{Ps}}{\mathrm{Pow}}<1.26\%;$ when N is 32 and SNR is less than-16 dB, $\frac{\mathrm{DL}-\mathrm{Pn}}{\mathrm{Pow}}>33\%,$ $\frac{\mathrm{DL}\_\mathrm{Ps}}{\mathrm{Pow}}<0.83\%;$ when N is 64 and SNR is less than-16 dB, $\frac{\mathrm{DL}\_\mathrm{Pn}}{\mathrm{Pow}}>20\%,$ $\frac{\mathrm{DL}-\mathrm{Ps}}{\mathrm{Pow}}<0.5\%.$

so at N16 and SNR < -16dB,therefore, when the number of array elements is small and the signal-to-noise ratio is low, the noise energy contained in the main diagonal elements of the cross spectrum matrix accounts for about 50% of the total energy, so that the display effect of the space spectrum obtained by beam forming under the low signal-to-noise ratio can be improved by correcting the main diagonal elements of the cross spectrum matrix.

The method (DL _ IBF for short) of the present invention has significant advantages over prior art methods (IBF for short). The effect of the method of the present invention and the related method of the prior art are compared with each other by way of example.

In a comparative example, a frequency f is used₁500Hz, amplitude P₁Taking a sine signal of 1v as a target 1 radiation signal, wherein the azimuth angle of the target 1 radiation signal relative to the horizontal line array is a radiation angle theta₁＝60°；Frequency f₂500Hz, amplitude P₂Taking sinusoidal signals of 0v, 0.5v and 1v as target 2 radiation signals, wherein the azimuth angle of the target 2 radiation signals relative to the horizontal line array is a radiation angle theta₂70 °; the phase difference between other array elements relative to the reference array element is reversely deduced according to the acoustic path difference between the target and other array elements and between the target and the reference array element and is added to the signals received by each array element; further, the background noise is made to be isotropic white Gaussian noise, and the frequency band is set to be 100 to 800 Hz. The number of elements N of the horizontal linear array is equal to 32, and the array interval isWherein c is 150m/s as effective sound velocity, and sampling rate is f_s5kHz, and 2000 quick beat.

And (4) obtaining simulation graphs under different signal-to-noise ratios.

FIG. 3(a) is P₂Obtaining a space spectrogram instantaneous graph by two methods under the condition that SNR is-25 dB;

FIG. 3(b) is P₂A spatial spectrum azimuth history map obtained by IBF under the condition that SNR is-25 dB is 0 v;

FIG. 3(c) is P₂0v, under the condition that SNR is-25 dB, obtaining a space spectrum azimuth history chart by DL _ IBF;

from FIG. 3(a) it can be derived: under the condition of a single target, the energy difference between the target position of the space spectrum obtained by the inverse beam forming algorithm corrected by the main diagonal elements and other positions is much larger than the energy difference between the target position of the space spectrum obtained without any processing and other positions; comparing FIG. 3(b) with FIG. 3(c) at the same time yields: the effect of displaying the target azimuth of the azimuth history map shown in fig. 3(c) is better than that of displaying the target azimuth of the azimuth history map shown in fig. 3 (b).

FIG. 4(a) is P₂Obtaining a space spectrogram instantaneous graph by two methods under the condition that SNR is-23 dB;

FIG. 4(b) is P₂0.5v, SNR-23 dB, from IBF, obtaining a spatial spectrum azimuth history chart;

FIG. 4(c) is P₂A spatial spectrum azimuth history map obtained by DL _ IBF under the condition that SNR is-23 dB is 0.5 v;

also from fig. 4(a) can be derived: under the same multi-target condition, the energy difference between the strong target position of the space spectrum obtained by the inverse beam forming algorithm corrected by the main diagonal elements and other positions is much larger than the energy difference between the strong target position of the space spectrum obtained without any processing and other positions; however, since the weak target has relatively small energy, the detection of the weak target cannot be well achieved by the two methods shown in fig. 4(a), which is a problem to be discussed further; comparing fig. 4(b) with fig. 4(c) can result in: the azimuth history map target azimuth display effect shown in fig. 4(c) is better than the azimuth history map target azimuth display effect shown in fig. 4(b), and the azimuth of a weak target can be displayed better in fig. 4(c) than in fig. 4 (b).

FIG. 5(a) is P₂1.0v, under the condition that SNR is-23 dB, obtaining a space spectrogram instantaneous graph by two methods;

FIG. 5(b) is P₂1.0v, under the condition that SNR is-23 dB, obtaining a space spectrum azimuth history chart by IBF;

FIG. 5(c) is P₂1.0v, and under the condition that SNR is-23 dB, obtaining a space spectrum azimuth history chart by DL _ IBF.

Also from fig. 5(a) can be derived: the energy difference between the strong target position of the space spectrum obtained by the main diagonal element-corrected inverse beam forming algorithm and other positions is much larger than the energy difference between the strong target position of the space spectrum obtained without any processing and other positions; comparing fig. 5(b) with fig. 5(c) yields: the effect of displaying the target azimuth of the azimuth history map shown in fig. 5(c) is better than that of displaying the target azimuth of the azimuth history map shown in fig. 5 (b).

The simulation results according to fig. 3 to 5 can therefore yield: under the condition of low signal to noise ratio, the energy difference between the strong target position of the space spectrum obtained by the inverse beam forming algorithm corrected by the main diagonal elements and other positions is much larger than the energy difference between the strong target position of the space spectrum obtained without any processing and other positions; the target azimuth display effect of the azimuth history diagram of the space spectrum obtained by the inverse beam forming algorithm corrected by the main diagonal elements is better than that of the target azimuth display effect of the azimuth history diagram of the space spectrum obtained without any processing. In addition to the above method, the present invention also provides an inverse beamforming system, which in one embodiment includes a preprocessing module, a cross-spectrum matrix calculation module, a modification module, a Toeplitz averaging module, a power spectrum calculation module, a spatial spectrum calculation module, and a loop execution module; wherein,

the preprocessing module performs discrete Fourier transform on signals received by each array element in the receiving towed array to obtain frequency domain values of the signals, and the frequency domain values are spatial spectrums P_{DL_IBF}Setting an initial value, and setting the frequency to f ═ f_min，f_minIs the lower frequency limit;

the cross spectrum matrix calculation module calculates a cross spectrum matrix for the receiving array according to the frequency domain value of the signal received by each array element obtained by the preprocessing module;

the correction module corrects main diagonal elements of the cross spectrum matrix obtained by the cross spectrum matrix calculation module; in this embodiment, the modifying module modifies the main diagonal elements of the cross-spectrum matrix including setting the values of the main diagonal elements to zero. In other embodiments, the modifying module may modify the main diagonal elements of the cross-spectrum matrix by setting the values of the main diagonal elements to be constants with absolute values less than 5.

The Toeplitz averaging module performs Toeplitz averaging on the modified cross-spectrum matrix of the main diagonal elements obtained by the modification module, and converts the Toeplitz averaging into one-dimensional distribution data with spatial distribution:

the power spectrum calculation module performs phase compensation on the one-dimensional distribution data obtained by the Toeplitz averaging module at an angle theta, and then sums the one-dimensional distribution data to obtain a power spectrum at the angle theta; the theta angle is an included angle between a target radiation signal and a horizontal line where the array elements are located;

the space spectrum calculation module performs space scanning on the power spectrum obtained by the power spectrum calculation module within 0-179 degrees to obtain a space spectrum P 'at the current signal frequency f'_{DL_IBF}(f) (ii) a Wherein the direction from the array element 1 to the array element N is 0 degree, and the anticlockwise rotation direction is 0-179 degree;

the loop execution module order P_{DL_IBF}＝P_{DL_IBF}+P'_{DL_IBF}(f) F is f +1, when f is less than or equal to f_maxWhen the broadband inverse beam is formed, restarting the cross-spectrum matrix calculation module, otherwise, finishing the operation, thereby completing the broadband inverse beam forming; wherein f is_maxIs the upper frequency limit.

In another embodiment, the system of the present invention further includes an FFT conversion module, where the FFT conversion module performs FFT conversion on a wideband signal in signals received by each array element in the receive array to obtain a narrowband signal, and transmits the obtained narrowband signal to the preprocessing module.

Finally, it should be noted that the above embodiments are only used for illustrating the technical solutions of the present invention and are not limited. Although the present invention has been described in detail with reference to the embodiments, it will be understood by those skilled in the art that various changes may be made and equivalents may be substituted without departing from the spirit and scope of the invention as defined in the appended claims.

Claims (8)

1. A method of inverse beamforming, comprising:

step 1), receiving signals by each array element in the towed linear array, performing discrete Fourier transform on the received signals to obtain frequency domain values of the signals, and obtaining a spatial spectrum P_{DL_IBF}Setting an initial value, and setting the frequency to f ═ f_min，f_minIs the lower frequency limit;

step 2), calculating a cross-spectrum matrix for the receiving array according to the frequency domain value of the signal received by each array element obtained in the step 1);

step 3), correcting main diagonal elements of the cross spectrum matrix obtained in the step 2);

step 4), conducting Toeplitz averaging on the modified cross-spectrum matrix of the main diagonal elements obtained in the step 3), and converting the Toeplitz averaging into one-dimensional distribution data with spatial distribution:

step 5), performing phase compensation on the one-dimensional distribution data obtained in the step 4) at the angle theta, and then summing to obtain a power spectrum at the angle theta; the theta angle is an included angle between a target radiation signal and a horizontal line where the array elements are located;

step 6), carrying out space scanning on the power spectrum obtained in the step 5) within 0-179 degrees to obtain a space spectrum P 'at the frequency f'_{DL_IBF}(f) (ii) a Wherein the direction from the array element 1 to the array element N is 0 degree, and the anticlockwise rotation direction is 0-179 degree;

step 7), let P_{DL_IBF}＝P_{DL_IBF}+P'_{DL_IBF}(f) F is f +1, when f is less than or equal to f_maxIf so, repeating the step 2) to the step 6), otherwise, finishing the operation, thereby finishing the broadband inverse beam forming; wherein f is_maxIs the upper frequency limit.

2. The inverse beamforming method according to claim 1, further comprising, in step 1): after receiving the array element receiving signal in the dragging line array, FFT conversion is carried out on the broadband signal in the received signal, and therefore the operation of the narrow-band signal is obtained.

3. Method for inverse beamforming according to claim 1 or 2, wherein in step 3) the modification of the main diagonal elements of the cross-spectral matrix comprises zeroing the values of the main diagonal elements.

4. Method for inverse beamforming according to claim 1 or 2, wherein in step 3) the modification of the main diagonal elements of the cross-spectral matrix comprises setting the values of the main diagonal elements to a constant with an absolute value smaller than 5.

5. The inverse beam forming system is characterized by comprising a preprocessing module, a cross-spectrum matrix calculating module, a correcting module, a Toeplitz averaging module, a power spectrum calculating module, a space spectrum calculating module and a cyclic executing module; wherein,

the preprocessing module performs discrete Fourier transform on signals received by each array element in the receiving towed array to obtain frequency domain values of the signals, and the frequency domain values are spatial spectrums P_{DL_IBF}Setting an initial value, and setting the frequency to f ═ f_min，f_minIs the lower frequency limit;

the cross spectrum matrix calculation module calculates a cross spectrum matrix for the receiving array according to the frequency domain value of the signal received by each array element obtained by the preprocessing module;

the correction module corrects main diagonal elements of the cross spectrum matrix obtained by the cross spectrum matrix calculation module;

the Toeplitz averaging module performs Toeplitz averaging on the modified cross-spectrum matrix of the main diagonal elements obtained by the modification module, and converts the Toeplitz averaging into one-dimensional distribution data with spatial distribution:

the power spectrum calculation module performs phase compensation on the one-dimensional distribution data obtained by the Toeplitz averaging module at an angle theta, and then sums the one-dimensional distribution data to obtain a power spectrum at the angle theta; the theta angle is an included angle between a target radiation signal and a horizontal line where the array elements are located;

the space spectrum calculation module performs space scanning on the power spectrum obtained by the power spectrum calculation module within 0-179 degrees to obtain a space spectrum P 'at the frequency f'_{DL_IBF}(f) (ii) a Wherein the direction from the array element 1 to the array element N is 0 degree, and the anticlockwise rotation direction is 0-179 degree;

the loop execution module order P_{DL_IBF}＝P_{DL_IBF}+P'_{DL_IBF}(f) F is f +1, when f is less than or equal to f_maxWhen the broadband inverse beam is formed, restarting the cross-spectrum matrix calculation module, otherwise, finishing the operation, thereby completing the broadband inverse beam forming; wherein f is_maxIs the upper frequency limit.

6. The inverse beamforming system of claim 5, further comprising an FFT transformation module, where the FFT transformation module performs FFT transformation on a wideband signal in signals received by each array element in the receive array to obtain a narrowband signal, and transmits the obtained narrowband signal to the preprocessing module.

7. Inverse beamforming system according to claim 5 or 6, wherein the modification module modifies the main diagonal elements of the cross-spectrum matrix comprising zeroing the values of the main diagonal elements.

8. Inverse beamforming system according to claim 5 or 6, wherein the modification module modifies the main diagonal elements of the cross-spectrum matrix comprising setting the values of the main diagonal elements to a constant with an absolute value smaller than 5.