CN1900738A - High resolution detection depth side scan sonar signal processing method - Google Patents

Abstract

The invention relates to a method for processing high-resolution sounding side-scan sonar signals using direction-of-arrival estimation technology to obtain seabed micro-topography, which is called a multi-subarray subspace fitting method. The invention shows that when the number of sonar array elements is fixed, there is a sub-array pair formation mode, and the standard deviation of phase estimation is the smallest when there is mutual coupling or no mutual coupling between the array elements. The invention overcomes the phase error of the sonar array caused by the influence of mutual coupling between the line array elements, and obtains the optimal combination of the number of sub-arrays and the number of line array elements contained in each sub-array under the condition of a given number of array elements, and improves the performance of the sonar. Sounding accuracy.

Description

A kind of high resolution detection depth side scan sonar signal processing method

Technical field

The present invention relates to a kind of high-resolution sonar signal processing method, particularly relate to a kind of direction of arrival that utilizes and estimate (Directions of Arrival, be called for short DOA) technology, the high resolution detection depth side scan sonar signal processing method of acquisition seabed mima type microrelief landforms.

Background technology

In later stage nineteen fifties, people begin to develop side-scan sonar.Its field angle of sonar battle array in vertical plane is very wide, general widely reaches 100 °, even wideer.Very narrow in surface level, be generally about 1 °.Sound pulse is penetrated in the sonar paroxysm, and the seabed produces the backscattering sound wave, is received by the sonar battle array according to the priority of time.The sonar battle array is in the process of advancing, and constantly emission constantly receives, and has just obtained the sound spectrogram in seabed, the landforms state in its reflection seabed.Obtained after side-scan sonar is succeeded in developing to use widely, produced remarkable economical and social benefit.In the real work, not only need the landforms in seabed, also need submarine topography simultaneously, often side-scan sonar is used with the multibeam echosounding sonar.For simplified apparatus, increase the benefit, people remember naturally makes side-scan sonar when obtaining sea-bed topography, also can obtain submarine topography, this detection depth side scan sonar that just has been born.Later stage nineteen nineties, detection depth side scan sonar adopts many parallel linear arrays, measure the phase differential of sound echo between them, once emission can obtain hundreds of to thousands of depth measurement points, it is better than multibeam sounding system on resolution, but it has two important defectives, its development of The Long-term Effect.At first, near under the sonar battle array, its total accuracy of sounding is very poor; Secondly, can not distinguish the echo that different directions arrives simultaneously, therefore, at the waters detection depth side scan sonar cisco unity malfunction that has many ways of channel and landform more complicated.

On March 29th, 2005 licensed among people's such as Zhu Weiqing, Liu Xiaodong U.S. Pat 6873570 B2 " Highresolution bathymetric sonar system and measuring method for measuring thephysiognomy of the seabed ", disclose two kinds of technology, solved two major defects of above-mentioned detection depth side scan sonar.First, the sonar battle array is made up of many equidistant linear arrays that are parallel to each other, adjacent linear array spacing is between a λ and λ/2, and λ is the underwater acoustic wave wavelength of sonar centre frequency correspondence, and near total accuracy of sounding has reached the precision of high-resolution digital sounder under the sonar; The second, the seabed is detected automatically---and the multiple submatrixes Wave arrival direction estimating method, its adopts direction of arrival to estimate (DOA) technology, extracts the information of sound wave by the time-space correlation function matrix of sonar battle array, comprises sound wave incident angle and amplitude etc.

Except echo signal, also has noise signal in the signal that the sonar battle array receives.Therefore sonar battle array time-space correlation function matrix can resolve into signal subspace and noise subspace in functional space, and two sub spaces are vertical mutually.The general two big classes of dividing of signal processing method to sonar battle array time-space correlation function matrix: a class is the spectrum based method, and it comprises the noise subspace method, claims Zero Space Method and Its again, and at small sample, when low signal-to-noise ratio and high signal coherency, the performance of these class methods descends; Another kind of is parametric method, and it comprises the signal subspace method.The performance of parametric method obviously is better than composing based method.Among U.S. Pat 6873570 B2, disclosing the seabed detects automatically---the multiple submatrixes Wave arrival direction estimating method, and it belongs to signal subspace method, function admirable, overcome the multi-path signals interference that underwater acoustic channel and complicated submarine topography cause, extracted the through echoed signal in seabed.But it does not have the open mutual coupling (Mutul Coupling) that how to solve between parallel linear array to influence problem.

Licensed to people's such as P.Kraeutner middle kernel (the being noise subspace) method that adopts in direction of arrival estimation (DOA) technology of United States Patent (USP) 613041 " Imagingmethods and apparatus using model-based array signal processing " on October 10th, 2000, sonar battle array time-space correlation function matrix is handled, obtained the resolution higher than conventional beam-forming technology.But near total accuracy of sounding under this patent does not provide is not studied the mutual coupling influence between parallel linear array yet.

Document 1:W Xu, the article of Stewart W K. " Coherent source direction estimation forthree-row bathymetric sidescan sonars ", OCEANS ' 99, MTS/IEEE, Seattle, Washington, estimate (Coherent Source Direction Estimator 299-304. proposed the coherent source direction in the literary composition, be called for short CSDE) method, for the detection depth side scan sonar of three winding displacement battle arrays and the situation of 2 signal sources, carried out analog computation, and compared with ESPRIT (Estimating Signal Parameters via Rotational InvarianceTechniques is called for short ESPRIT) method.Analog computation shows, for the signal source of the height correlation of 10 ° of angular spacings, more than the signal to noise ratio (S/N ratio) 10db, CSDE work is good.ESPRIT has good robustness for the noncoherent signal source simultaneously.Document suggestion replaces existing difference phase scheme to two kinds of method use in conjunction.

In sum, the deficiency that exists in the existing technology mainly contains:

1. high resolution detection depth side scan sonar has strict demand to the phase propetry of sonar battle array, and it is the how high important parameter of decision resolution.Mutual coupling influence between many parallel linear battle arrays, the especially influence of the mutual coupling between adjacent lines array element are important phase error sources.How to reduce this class error, in the prior art, comprise above-mentioned two patents of this patent and one piece of document, all do not mention by method for processing signals.

2. high resolution detection depth is surveyed the reception sonar battle array sweep sonar and is made up of many equidistant parallel lines array elements, receives gust to be divided into several submatrixs, and each submatrix comprises several linear array units, and a good combination is wherein arranged, its DOA estimation error minimum.In the existing technology, comprise above-mentioned two patents of this patent and one piece of document, all do not mention the method that obtains best combination.

Summary of the invention

The objective of the invention is to overcome the above-mentioned deficiency of prior art, proposed the signal processing method of multiple submatrixes subspace fitting.

In order to achieve the above object, the present invention takes following technical scheme:

A kind of high resolution detection depth side scan sonar signal processing method comprises the steps:

A) determine that each submatrix chooses the submatrix in the mode;

B) calculate each submatrix and choose the incident angle of the submatrix of mode for target;

C) calculate the variance that each submatrix is chosen the incident angle of mode;

D) submatrix of choosing incident angle standard deviation minimum is chosen the working method of mode as the sonar battle array.

In technique scheme, further, step a) specifically comprises following substep:

1) for mutual coupling matrix degree of freedom be the sonar battle array of h, head and the tail array element is removed in h 〉=2, gets sonar battle array output signal X, is expressed from the next:

X＝AGS+N (3)

Wherein

X＝[x _h?x _h+1?…?x _L-(h-1)z] ^T (4a)

S＝[s ₁?s ₂?…?s _M] ^T (4b)

$G=\mathrm{diag}\left\{\begin{array}{ccc}\underset{q=-(h-1)}{\overset{h-1}{\mathrm{\Σ}}}{b}_{\left|q\right|}{e}^{j\left(q{\mathrm{\φ}}_1\right)}g\left({\mathrm{\θ}}_1\right)& ...& \underset{q=-(h-1)}{\overset{h-1}{\mathrm{\Σ}}}{b}_{\left|q\right|}{e}^{j\left(q{\mathrm{\φ}}_M\right)}g\left(\mathrm{\θM}\right)\end{array}\right\},b_0=1---\left(5\right)$

Wherein, M is a number of targets, i.e. signal source number, and L is an array number, and N is a zero-mean space white noise, and X and N are that L-2 (h-1) dimensional vector A is the flow pattern matrix of space array; Signal S is made of M independent gaussian signal source;

2) ask the estimation of sonar battle array related function

Promptly

$\hat{R}=\frac{1}{N}\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{X}_n{X_n}^H---\left(6\right)$

3) related function of sonar battle array output signal is estimated

Carry out characteristic value decomposition:

$\hat{R}=\mathrm{AG}{\hat{R}}_S{G}^H{A}^H+{\mathrm{\σ}}^{2}I={\hat{U}}_S{\widehat{\mathrm{\Σ}}}_S{{\hat{U}}_S}^H+{\hat{U}}_N{\widehat{\mathrm{\Σ}}}_N{{\hat{U}}_N}^H---\left(7\right)$

Wherein

Be the estimated value of signal correction function, σ ²Be the estimated value of noise variance, With Be respectively the proper vector of signal and the estimated value of eigenwert,

With It is respectively the estimated value of characteristics of noise vector sum eigenwert; Subscript H represents the conjugate transpose computing; I is a unit matrix.

4) ask the submatrix computing:

J ₁＝[I _l-1?0] _(l-1)×l，J ₂＝[0?I _l-1] _(l-1)×l (8)

Wherein, M＜l≤L-2 (h-1);

In technique scheme, further, step b) specifically comprises following substep:

5) ask submatrix characteristic of correspondence vector estimated value;

${\hat{U}}_{S1}=J_1{\hat{U}}_S---\left(9\right)$

${\hat{U}}_{S2}=J_2{\hat{U}}_S$

6) by previous step rapid 5) can obtain estimated value

With

Obtain

Separating of multiple submatrixes subspace fitting algorithm when having mutual coupling:

$\widehat{\mathrm{\Ψ}}={\left({{\hat{U}}_{S1}}^H{\hat{U}}_{S1}\right)}^{-1}{{\hat{U}}_{S1}}^H{\hat{U}}_{S2}---\left(10\right)$

7) ask

Computing:

$\widehat{\mathrm{\Ψ}}=C^{-1}\widehat{\mathrm{\Φ}}C,---\left(11\right)$

Wherein $\widehat{\mathrm{\Φ}}=\mathrm{diag}\left\{\begin{array}{ccc}{e}^{i{\widehat{\mathrm{\φ}}}_1}& ...& e^{i{\widehat{\mathrm{\φ}}}_M}\end{array}\right\}---\left(12a\right)$

$C=G{\hat{R}}_S$ $G^H{A}^H{\hat{U}}_S{{\widehat{\mathrm{\Σ}}}_S}^{\′}---\left(12b\right)$

${{\widehat{\mathrm{\Σ}}}_S}^{\′}={\widehat{\mathrm{\Σ}}}_S-{\mathrm{\σ}}^{2}I---\left(12c\right)$

8) ask sound wave incident angle estimated value

Computing:

${\widehat{\mathrm{\θ}}}_i={\sin }^{-1}({\widehat{\mathrm{\φ}}}_i/\mathrm{kd}),i=\mathrm{1,2},...M---\left(13\right)$

In technique scheme, further, step c) specifically comprises following substep:

9) at fast umber of beats N greater than 100 o'clock, the evaluated error of the multiple submatrixes subspace fitting algorithm of even linear array Be the associating Gaussian distribution, its average and variance are respectively:

$E\{{\widehat{\mathrm{\φ}}}_i-{\mathrm{\φ}}_i\}=0---\left(14a\right)$

$E\left({({\widehat{\mathrm{\φ}}}_i-{\mathrm{\φ}}_i)}^2\right)=\frac{{\mathrm{\σ}}^2}{2N}\mathrm{Re}\left\{\left({{\mathrm{\ρ}}_i}^H{\mathrm{\ρ}}_i\right)({\left[P^{-1}\right]}_{i,i}+{\mathrm{\σ}}^2{\left[P^{-1}{\left(A^{H}A\right)}^{-1}{P}^{-1}\right]}_{i,i})\right\}---\left(14b\right)$

I=1 wherein ..., M;

$P=G{\hat{R}}_s$ $G^H---\left(15a\right)$

${{\mathrm{\ρ}}_i}^H={{\left[{\left(A_1^H{A}_1\right)}^{-1}{A}_1^H{F}_i\right]}_i}^{\left(\mathrm{\γ}\right)}---\left(15b\right)$

A in the formula ₁=[I _l0] A, F _iBe (l-1) * l matrix, [X] _i ^(γ)The i of representing matrix X is capable, Expression signal correction Function Estimation value;

Square root according to variance is a standard deviation, calculates the right incident angle estimated standard deviation of all submatrixs, and the incident angle estimated standard deviation right to all submatrixs carries out arithmetic mean, and this mean value is as final estimated standard deviation;

10) by step 8) and step 9) calculating sound incident angle estimated value

Final estimated standard deviation;

11) all possible submatrix is chosen mode, repeating step 4)-step 10), obtain the final estimated standard deviation of incident angle that all possible submatrix is chosen mode.

In technique scheme, further, in the step d): all possible submatrix that draws comparison previous step rapid 11) is chosen the final estimated standard deviation of incident angle of mode, determine that submatrix of the final estimated standard deviation minimum of incident angle chooses mode, choose this submatrix and choose the working method of mode as sonar.

Compared with prior art, the invention has the advantages that:

(1) with general only from sensor array, or only unilaterally suppress the mutual coupling existing between elements difference from the DOA signal Processing, the present invention suppresses mutual coupling between array element jointly by two aspects, just can develop well behaved sonar on this basis.The present invention is for the directive significance of sonar design: one side reduces the mutual coupling matrix degree of freedom h of sonar battle array as far as possible, needs the array element of deletion few like this, and cost is little; Multiple submatrixes subspace fitting technology still has superperformance when there is mutual coupling in array element on the other hand.

(2) the present invention shows, under the certain situation of sonar array number, when mutual coupling being arranged between array element or not having mutual coupling, all there is the right constituted mode of a kind of submatrix, its phase estimation standard deviation minimum adopts the right constituted mode of this submatrix can be so that sonar estimating target orientation more accurately.

Description of drawings

There is not mutual coupling in concerning of Fig. 1 incident angle estimated standard deviation (degree) and submatrix number between even linear array array element, signal to noise ratio (S/N ratio) 20dB, and 8 of array numbers, the incident angle that fast umber of beats 100, signal number 1, curve are corresponding from top to bottom is 80 °, 60 °, 40 °, 20 ° and 0 °;

There is mutual coupling in concerning between even linear array array element of Fig. 2 incident angle estimated standard deviation (degree) and submatrix number, degree of freedom 3, coupling coefficient b ₀=1, b ₁=0.2exp (j π/6), b ₂=0.05exp (j π/5), signal to noise ratio (S/N ratio) 20dB, 8 of array numbers, the incident angle that fast umber of beats 100, signal number 1, curve are corresponding from top to bottom is 80 °, 60 °, 40 °, 20 ° and 0 °;

Fig. 3 is the process flow diagram of the test procedure of application one embodiment of the invention.

Embodiment

Below in conjunction with the drawings and specific embodiments the present invention is described in further detail:

At first, the present invention provides a kind of DOA signal processing technology, overcomes the sonar battle array phase error that the influence of line mutual coupling existing between elements causes.

If arrowband plane wave s (t) incides on the even reception linear array of being made up of L array element, suppose only to consider the interaction between adjacent array element, then receive battle array input s (t) and output x _i, i=1 ... L, relation be expressed from the next

$\left[\begin{array}{c}{x}_1\\ x_2\\ x_2\\ .\\ .\\ .\\ .\\ .\\ .\\ x_{L-2}\\ x_{L-1}\\ x_L\end{array}\right]=\left[\begin{array}{cccccccccc}1& b_1& 0& 0& 0& 0& ...& ...& ...& 0\\ b_1& 1& b_1& 0& 0& 0& ...& ...& ...& 0\\ 0& b_1& 1& b_1& 0& 0& ...& ...& ...& 0\\ ...& ...& ...& ...& ...& ...& ...& ...& ...& ...\\ ...& ...& ...& ...& ...& ...& ...& ...& ...& ...\\ 0& ...& 0& 0& 0& 0& b_1& 1& b_1& 0\\ 0& ...& 0& 0& 0& 0& 0& b_1& 1& b_1\\ 0& ...& 0& 0& 0& 0& 0& 0& b_1& 1\end{array}\right]\left[\begin{array}{c}1\\ e^{\mathrm{i\φ}}\\ e^{i2\mathrm{\φ}}\\ .\\ .\\ .\\ .\\ .\\ .\\ e^{i(L-3)\mathrm{\φ}}\\ e^{i(L-2)\mathrm{\φ}}\\ e^{i(L-1)\mathrm{\φ}}\end{array}\right]g\left(\mathrm{\θ}\right)s\left(t\right)---\left(1\right)$

Wherein, g (θ) is the directive property of linear array unit, and they are identical, b ₁Be coupling factor, the mutual coupling influence between expression adjacent lines array element.φ=kd sin θ, k is a wave number, and d is the spacing of adjacent lines array element, and θ is the sound wave incident angle.Following formula is the effect of discussing between the adjacent lines array element, so do not consider The noise.First matrix of formula (1) the right is mutual coupling square B, and wherein the number of the first row nonzero element is called the degree of freedom h of mutual coupling matrix, h=2 in the formula (1).In the matrix B first row and last column are removed, get matrix B ₁, Dui Ying linear array unit is output as x with it _i, i=2 ... L-1.At i=2 ... appoint among the L-1 and get two submatrixs, the line array number that each submatrix comprises is identical, for example gets i=2 ... L-2, and i=3 ... L-1.Then

$\left[\begin{array}{c}{x}_3\\ x_4\\ .\\ .\\ .\\ .\\ .\\ .\\ x_{L-2}\\ x_{L-1}\end{array}\right]=\left[\begin{array}{ccccccccc}{b}_1& 1& b_1& 0& ...& ...& ...& ...& ...\\ 0& b_1& 1& b_1& ...& ...& ...& ...& ...\\ ...& ...& ...& ...& ...& ...& ...& ...& ...\\ ...& ...& ...& ...& ...& ...& ...& ...& ...\\ ...& ...& ...& ...& ...& b_1& 1& b_1& 0\\ 0& 0& 0& 0& ...& ...& b_1& 1& b_1\end{array}\right]\left[\begin{array}{c}1\\ e^{\mathrm{i\φ}}\\ e^{i2\mathrm{\φ}}\\ .\\ .\\ .\\ .\\ .\\ .\\ e^{i(L-3)\mathrm{\φ}}\\ e^{i(L-2)\mathrm{\φ}}\end{array}\right]g\left(\mathrm{\θ}\right)s\left(t\right)e^{\mathrm{i\φ}}=\left[\begin{array}{c}{x}_2\\ x_3\\ .\\ .\\ .\\ .\\ .\\ .\\ x_{L-1}\\ x_{L-2}\end{array}\right]e^{\mathrm{i\φ}}---\left(2\right)$

Formula (2) shows, is not counted in head and the tail array element, is divided into two adjacent submatrixs receiving battle array, and each submatrix comprises identical array number, and then the ratio of the output signal of submatrix is e ^{I φ}, the result when not having mutual coupling with the sonar battle array is consistent.That is to say, by method of the present invention e when having mutual coupling and not having mutual coupling ^{I φ}Average be identical, it is not have partially to estimate.The sound wave incident angle θ that obtains of φ thus.Sound wave is to the glancing angle α=θ+θ in seabed _m, θ _mBe sonar battle array established angle, it equals sonar battle array plane and the interplanar angle of spatial vertical.

And then, the invention provides a kind of multiple submatrixes DOA signal processing technology, under the given situation of array number, obtain the optimal combination of the line array number that submatrix number and each submatrix comprise, improve the total accuracy of sounding of sonar.For example be similar to (2) formula, get i=2 ... L-3, or 3 ... L-2 or 4 ... L-1 etc.

A kind of high resolution detection depth side scan sonar signal processing method comprises the steps:

1) for mutual coupling matrix degree of freedom be the sonar battle array of h, head and the tail array element is removed in h 〉=2, gets sonar battle array output signal X, is expressed from the next:

X＝AGS+N (3)

Wherein

X＝[x _h?x _h+1…x _L-(h-1)] ^T (4a)

S＝[s ₁?s ₂…s _M] ^T (4b)

$G=\mathrm{diag}\left\{\begin{array}{ccc}\underset{q=-(h-1)}{\overset{h-1}{\mathrm{\Σ}}}{b}_{\left|q\right|}{e}^{j\left(q{\mathrm{\φ}}_1\right)}g\left({\mathrm{\θ}}_1\right)& ...& \underset{q=-(h-1)}{\overset{h-1}{\mathrm{\Σ}}}{b}_{\left|q\right|}{e}^{j\left(q{\mathrm{\φ}}_M\right)}g\left({\mathrm{\θ}}_M\right)\end{array}\right\},b_0=1---\left(5\right)$

Wherein, M is a number of targets, i.e. signal source number, and L is an array number, and N is a zero-mean space white noise, and X and N are L-2 (h-1) dimensional vector; A is the flow pattern matrix of space array; Signal S is made of M independent gaussian signal source;

2) ask the estimation of sonar battle array related function

Promptly

$\hat{R}=\frac{1}{N}\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{X}_n{X_n}^H---\left(6\right)$

3) related function of sonar battle array output signal is estimated

Carry out characteristic value decomposition:

$\hat{R}=\mathrm{AG}{\hat{R}}_S{G}^H{A}^H+{\mathrm{\σ}}^{2}I={\hat{U}}_S{\widehat{\mathrm{\Σ}}}_S{{\hat{U}}_S}^H+{\hat{U}}_N{\widehat{\mathrm{\Σ}}}_N{{\hat{U}}_N}^H---\left(7\right)$

Wherein

Be the estimated value of signal correction function, σ ²Be the estimated value of noise variance, With

Be respectively the proper vector of signal and the estimated value of eigenwert,

With It is respectively the estimated value of characteristics of noise vector sum eigenwert; Subscript H represents the conjugate transpose computing; I is a unit matrix.

4) ask the submatrix computing:

J ₁＝[I _l-1?0] _(l-1)×l，J ₂＝[0?I _l-1]( _l-1)×l (8)

Wherein, M＜l≤L-2 (h-1);

Disclosed technology in the document 1 is adopted in the submatrix computing of asking in this step;

5) ask submatrix characteristic of correspondence vector estimated value;

${\hat{U}}_{S1}=J_1{\hat{U}}_S---\left(9\right)$

${\hat{U}}_{S2}=J_2{\hat{U}}_S$

6) by previous step rapid 5) obtain estimated value With

Obtain

Separating of multiple submatrixes subspace fitting algorithm when having mutual coupling:

$\widehat{\mathrm{\Ψ}}={\left({{\hat{U}}_{S1}}^H{\hat{U}}_{S1}\right)}^{-1}{{\hat{U}}_{S1}}^H{\hat{U}}_{S2}---\left(10\right)$

7) ask Computing:

$\widehat{\mathrm{\Ψ}}=C^{-1}\widehat{\mathrm{\Φ}}C,---\left(11\right)$

Wherein $\widehat{\mathrm{\Φ}}=\mathrm{diag}\left\{\begin{array}{ccc}{e}^{i{\widehat{\mathrm{\φ}}}_1}& ...& e^{i{\widehat{\mathrm{\φ}}}_M}\end{array}\right\}---\left(12a\right)$

$C=G{\hat{R}}_S$ $G^H{A}^H{\hat{U}}_S{{\widehat{\mathrm{\Σ}}}_S}^{\′}---\left(12b\right)$

${{\widehat{\mathrm{\Σ}}}_S}^{\′}={\widehat{\mathrm{\Σ}}}_S-{\mathrm{\σ}}^{2}I---\left(12c\right)$

8) ask sound wave incident angle estimated value

Computing:

${\widehat{\mathrm{\θ}}}_i={\sin }^{-1}\left({\widehat{\mathrm{\φ}}}_i\text{/kd}\right),i=\mathrm{1,2},...M---\left(13\right)$

9) at fast umber of beats N greater than 100 o'clock, the evaluated error of the multiple submatrixes subspace fitting algorithm of even linear array Be the associating Gaussian distribution of a zero-mean, the evaluated error of the multiple submatrixes subspace fitting algorithm of even linear array Be the associating Gaussian distribution, its average and variance are respectively:

$E\{{\widehat{\mathrm{\φ}}}_1-{\mathrm{\φ}}_i\}=0---\left(14a\right)$

$E\left({({\widehat{\mathrm{\φ}}}_i-{\mathrm{\φ}}_i)}^2\right)=\frac{{\mathrm{\σ}}^2}{2N}\mathrm{Re}\left\{\left({{\mathrm{\ρ}}_i}^H{\mathrm{\ρ}}_i\right)({\left[P^{-1}\right]}_{i,i}+{\mathrm{\σ}}^2{\left[P^{-1}{\left(A^{H}A\right)}^{-1}{P}^{-1}\right]}_{i,i})\right\}---\left(14b\right)$

I=1 wherein ..., M;

$P=G{\hat{R}}_S{G}^H---\left(15a\right)$

${{\mathrm{\ρ}}_i}^H={{\left[{\left(A_1^H{A}_1\right)}^{-1}{A}_1^H{F}_i\right]}_i}^{\left(\mathrm{\γ}\right)}---\left(15b\right)$

A in the formula ₁=[I _l0] A, F _iBe (l-1) * l matrix, [X] _i ^(γ)The i of representing matrix X is capable,

Expression signal correction Function Estimation value;

For zero-mean be illustrated in consider have mutual coupling between array element after, the same when not having mutual coupling, multiple submatrixes subspace fitting algorithm is not have partially to estimate; The incident angle estimated standard deviation right to all submatrixs carries out arithmetic mean, and this average is as final estimated standard deviation;

10) by step 8) and step 9) calculating sound incident angle estimated value

Final estimated standard deviation;

11) all possible submatrix is chosen mode, repeating step 4)-step 10); Obtain all possible submatrix and choose the final estimated standard deviation of incident angle of mode;

12) all possible submatrix that draws comparison previous step rapid 11) is chosen the incident angle estimated standard deviation of mode, determines to exist that submatrix of incident angle estimated standard deviation minimum to choose mode, chooses this submatrix and chooses the working method of mode as sonar.

In above-mentioned steps 1) in, if target is irrelevant, mutual coupling matrix degree of freedom h=2 removes head and the tail array element, gets sonar battle array output X, is expressed from the next

X＝AGS+N

Wherein

X＝[x ₂?x ₃…x _L-1] ^T

S＝[s ₁?s ₂…s _M] ^T

$G=\mathrm{diag}\left\{\begin{array}{ccc}\underset{q=-1}{\overset{1}{\mathrm{\Σ}}}{b}_{\left|q\right|}{e}^{j\left(q{\mathrm{\φ}}_1\right)}g\left({\mathrm{\θ}}_1\right)& ...& \underset{q=-1}{\overset{1}{\mathrm{\Σ}}}{b}_{\left|q\right|}{e}^{j\left(q{\mathrm{\φ}}_M\right)}g\left({\mathrm{\θ}}_M\right)\end{array}\right\},b_0=1$

Wherein, L is an array number, and N is a zero-mean space white noise, is the L-2 n dimensional vector n; A is the flow pattern matrix of space array.

Technology of the present invention is used for test or sea trial on the lake, can takes following steps:

A) the emission battle array is launched ping in water;

B) receive the backscattering echoed signal that battle array receives water body and seabed;

C) echo that receives is carried out filtering and sampling, obtain sonar battle array output signal X by formula (3) expression;

D) step 4)-10 of repetition the foregoing description); Obtain possible submatrix choose the incident angle estimated standard deviation of mode;

E) all possible submatrix that draws the rapid D of comparison previous step) is chosen the incident angle estimated standard deviation of mode, chooses the sort of submatrix of incident angle estimated standard deviation minimum and chooses the working method of mode as sonar.

Sonar signal processing method of the present invention can be used as the special measurement program and is applied to sonar system, generally this special measurement program is loaded in the main control computer of sonar system.Program flow diagram as shown in Figure 3, herein based on the embodiment of the invention described above, the symbol among the figure as

Or the like meaning same with the above-mentioned embodiment, program process is as follows:

Step 401 is beginning steps, starts the program of computing machine, makes sonar in running order.

In step 402 and 403, the software and hardware of sonar system is carried out initialization.

In step 404, main control computer generates and transmits.

In step 405, launch sound pulse in fluid media (medium), for example in the seawater to the seabed.

In step 406, receive from the backscattered echoed signal of fluid media (medium).

In step 407, echoed signal is carried out demodulation filtering.

In step 408, echoed signal is become digital signal from analog signal conversion, and to each echoed signal execution in step 409～step 413 one by one.

In step 409, calculate sonar battle array related function and estimate

In step 410, carry out

The characteristic value decomposition computing.

In step 411, carry out the submatrix computing, calculate submatrix characteristic of correspondence vector estimated value.

In step 412, calculate

In step 413, calculate

In step 414, calculate Variance.

In step 415, calculate Standard deviation.

In step 416, calculate sound wave incident angle estimated value

In step 417, calculate the glancing angle estimated value of sound wave to the seabed

In step 418, storage

With

Standard deviation.Feed back to step 411 simultaneously, repeating step 411 is to the calculating of step 417, chooses until all possible submatrix mode to finish, and chooses

With The submatrix of standard deviation minimum choose the working method of mode as sonar.

Need to prove that above-mentioned steps 401-step 408 all adopts the technology that well known to a person skilled in the art, so it is not elaborated among the present invention.All the other step 409-steps 418 adopt the formula in the previous embodiment to calculate.

Some concrete data instances that obtain with the application present embodiment illustrate effect of the present invention below.

Carry out the step 1)-step 10) in the present embodiment, obtain the incident angle estimated standard deviation that all possible submatrix is chosen mode, typical consequence is illustrated in Fig. 1 and Fig. 2.What Fig. 1 and Fig. 2 showed is the relation of incident angle estimated standard deviation and submatrix number.Among Fig. 1, do not have coupling between even linear array array element, the degree of freedom h=1 of mutual coupling matrix, array number are 8, and signal number M is 1, g (θ)=1, and the information source incident angle is respectively 80 °, and 60 °, 40 °, 20 ° and 0 °.Fig. 1 shows that incident angle is bigger, and the incident angle estimated standard deviation is bigger; There is certain submatrix number, its incident angle estimated standard deviation minimum.What Fig. 2 represented is to have coupling, the degree of freedom h=3 of mutual coupling matrix, b between even linear array array element ₀=1, b ₁=0.2e ^{J π/6}, b ₂=0.05e ^{J π/5}, other parameter is identical with Fig. 1.G has been carried out normalization, and the maximal value of its mould is 1.Wherein 8 of array numbers are meant the array number that actual participation is calculated, and do not comprise four head and the tail array elements removing.As shown in Figure 2 owing to there is mutual coupling, some increase of incident angle estimated standard deviation, incident angle is bigger, and the incident angle estimated standard deviation is bigger; Existence submatrix number same as in figure 1, its incident angle estimated standard deviation minimum.

Situation for two targets is M=2, and is similar to the result of a target.

It should be noted last that above embodiment is only unrestricted in order to technical scheme of the present invention to be described.Although the present invention is had been described in detail with reference to embodiment, those of ordinary skill in the art is to be understood that, technical scheme of the present invention is made amendment or is equal to replacement, do not break away from the spirit and scope of technical solution of the present invention, it all should be encompassed in the middle of the claim scope of the present invention.

Claims (5)

1. A method for processing high-resolution sounding side-scan sonar signals, comprising the steps of: a) determining the sub-arrays in each sub-array selection mode; b) calculating the angle of incidence of the sub-array of each sub-array selection mode for the target; c) calculating the variance of the angle of incidence of each sub-array selection mode; d) Select the sub-array selection method with the smallest standard deviation of incident angle as the working method of the sonar array. 2. The high-resolution sounding side-scan sonar signal processing method according to claim 1, wherein step a) specifically comprises the following sub-steps: 1) For the sonar array with the mutual coupling matrix degree of freedom h, h≥2, the first and last array elements are removed to obtain the output signal X of the sonar array, which is expressed by the following formula: X=AGS+N in X＝[x _h x _h+1 ... x _L-(h-1) ] ^T S＝[s ₁ s ₂ …s _M ] ^T

$\mathrm{<span\; class="notranslate">\; G</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; diag</span>}\left\{\begin{array}{ccc}\underset{\mathrm{<span\; class="notranslate">\; q</span>}<span\; class="notranslate">\; =</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; h</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>}{\overset{\mathrm{<span\; class="notranslate">\; h</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; b</span>}}_{<span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; q</span>}<span\; class="notranslate">\; |</span>}{\mathrm{<span\; class="notranslate">\; e</span>}}^{\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; q\&phi;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span>}\mathrm{<span\; class="notranslate">\; g</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span>& <span\; class="notranslate">\; \&Center\; Dot;</span><span\; class="notranslate">\; \&Center\; Dot;</span><span\; class="notranslate">\; \&Center\; Dot;</span>& \underset{\mathrm{<span\; class="notranslate">\; q</span>}<span\; class="notranslate">\; =</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; h</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>}{\overset{\mathrm{<span\; class="notranslate">\; h</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; b</span>}}_{<span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; q</span>}<span\; class="notranslate">\; |</span>}{\mathrm{<span\; class="notranslate">\; e</span>}}^{\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; q</span>}{\mathrm{<span\; class="notranslate">\; \&phi;</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}<span\; class="notranslate">\; )</span>}\mathrm{<span\; class="notranslate">\; g</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}<span\; class="notranslate">\; )</span>\end{array}\right\}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; b</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}$ Wherein, M is the number of targets, that is, the number of signal sources, L is the number of array elements, N is zero-mean space white noise, X and N are L-2(h-1) dimensional vectors; A is the flow matrix of the space array; The signal S is composed of M independent Gaussian signal sources; 2) Find the estimation of the correlation function of the sonar array Right now $\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; R</span>}}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; N</span>}}\underset{\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; N</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}}{{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}}}^{\mathrm{<span\; class="notranslate">\; h</span>}}$ 3) Estimation of the correlation function of the output signal of the sonar array

Perform an eigenvalue decomposition: $\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; R</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; AG</span>}{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; R</span>}}}_{\mathrm{<span\; class="notranslate">\; S</span>}}{\mathrm{<span\; class="notranslate">\; G</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; A</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&sigma;</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; =</span>{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; u</span>}}}_{\mathrm{<span\; class="notranslate">\; S</span>}}{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}_{\mathrm{<span\; class="notranslate">\; S</span>}}{{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; u</span>}}}_{\mathrm{<span\; class="notranslate">\; S</span>}}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; +</span>{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; u</span>}}}_{\mathrm{<span\; class="notranslate">\; N</span>}}{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}_{\mathrm{<span\; class="notranslate">\; N</span>}}{{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; u</span>}}}_{\mathrm{<span\; class="notranslate">\; N</span>}}}^{\mathrm{<span\; class="notranslate">\; h</span>}}$ in

is the estimated value of the signal correlation function, ^σ2 is the estimated value of the noise variance,

and are the estimated eigenvectors and eigenvalues of the signal, respectively, and are the estimated values of the eigenvectors and eigenvalues of the noise, respectively; the superscript H represents the conjugate transpose operation; I is the identity matrix; 4) Subarray operation: J ₁ =[I _l-1 0] _(l-1)×l , J ₂ =[0 I _l-1 ] _(l-1)×l Wherein, M<l≤L-2(h-1). 3. The high-resolution sounding side-scan sonar signal processing method according to claim 2, wherein step b) specifically includes the following sub-steps: 5) Find the estimated value of the eigenvector corresponding to the sub-array; ${\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; u</span>}}}_{\mathrm{<span\; class="notranslate">\; S</span>}\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; J</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; u</span>}}}_{\mathrm{<span\; class="notranslate">\; S</span>}}$ ${\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; u</span>}}}_{\mathrm{<span\; class="notranslate">\; S</span>}\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; J</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; u</span>}}}_{\mathrm{<span\; class="notranslate">\; S</span>}}$ 6) The estimated value can be obtained from the previous step 5)

and get The solution of the multi-subarray subspace fitting algorithm in the presence of mutual coupling: $\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; \&Psi;</span>}}<span\; class="notranslate">\; =</span>{<span\; class="notranslate">\; (</span>{{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; u</span>}}}_{\mathrm{<span\; class="notranslate">\; S</span>}\mathrm{<span\; class="notranslate">\; 1</span>}}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; u</span>}}}_{\mathrm{<span\; class="notranslate">\; S</span>}\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; )</span>}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; u</span>}}}_{\mathrm{<span\; class="notranslate">\; S</span>}\mathrm{<span\; class="notranslate">\; 1</span>}}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; u</span>}}}_{\mathrm{<span\; class="notranslate">\; S</span>}\mathrm{<span\; class="notranslate">\; 2</span>}}$ 7) seek

Operation: $\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; \&Psi;</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; C</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; \&Phi;</span>}}\mathrm{<span\; class="notranslate">\; C</span>}<span\; class="notranslate">\; ,</span>$ in $\widehat{\mathrm{\&Phi;}}=\mathrm{diag}\{e^{i{\widehat{\mathrm{\&phi;}}}_1}\&Center\; Dot;\&CenterDot;\&CenterDot;e^{i{\widehat{\mathrm{\&phi;}}}_{l1}}\}$ $\mathrm{<span\; class="notranslate">\; C</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; G</span>}{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; R</span>}}}_{\mathrm{<span\; class="notranslate">\; S</span>}}{\mathrm{<span\; class="notranslate">\; G</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; A</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; u</span>}}}_{\mathrm{<span\; class="notranslate">\; S</span>}}{{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}_{\mathrm{<span\; class="notranslate">\; S</span>}}}^{<span\; class="notranslate">\; \&prime;</span>}$ ${{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}_{\mathrm{<span\; class="notranslate">\; S</span>}}}^{<span\; class="notranslate">\; \&prime;</span>}<span\; class="notranslate">\; =</span>{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}_{\mathrm{<span\; class="notranslate">\; S</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&sigma;</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}\mathrm{<span\; class="notranslate">\; I</span>}$ 8) Find the estimated value of the incident angle of the sound wave

Operation: ${\widehat{\mathrm{\&theta;}}}_i={\sin }^{-1}({\widehat{\mathrm{\&phi;}}}_i/\mathrm{kd})$ i=1, 2, . . . M. 4. The method for processing high-resolution sounding side-scan sonar signals according to claim 3, wherein said step c) specifically includes the following sub-steps: 9) When the number of snapshots N is greater than 100, the estimation error of the multi-subarray subspace fitting algorithm of the uniform linear array

is a joint Gaussian distribution with mean and variance: $\mathrm{<span\; class="notranslate">\; E.</span>}<span\; class="notranslate">\; \{</span>{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; \&phi;</span>}}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&phi;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; \}</span><span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}$ $\mathrm{<span\; class="notranslate">\; E.</span>}<span\; class="notranslate">\; (</span>{<span\; class="notranslate">\; (</span>{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; \&phi;</span>}}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&phi;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; \&sigma;</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}}{\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; N</span>}}\mathrm{<span\; class="notranslate">\; Re</span>}<span\; class="notranslate">\; \{</span><span\; class="notranslate">\; (</span>{{\mathrm{<span\; class="notranslate">\; \&rho;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; \&rho;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; (</span>{<span\; class="notranslate">\; [</span>{\mathrm{<span\; class="notranslate">\; P</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; ]</span>}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&sigma;</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; [</span>{\mathrm{<span\; class="notranslate">\; P</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; A</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}\mathrm{<span\; class="notranslate">\; A</span>}<span\; class="notranslate">\; )</span>}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; P</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{<span\; class="notranslate">\; ]</span>}_{\mathrm{<span\; class="notranslate">\; l</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \}</span>$ where i=1,...,M; $\mathrm{<span\; class="notranslate">\; P</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; G</span>}{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; R</span>}}}_{\mathrm{<span\; class="notranslate">\; the\; s</span>}}{\mathrm{<span\; class="notranslate">\; G</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}$ ${{\mathrm{<span\; class="notranslate">\; \&rho;</span>}}_{\mathrm{<span\; class="notranslate">\; l</span>}}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; =</span>{{<span\; class="notranslate">\; [</span>{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; A</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; A</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; )</span>}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; A</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; f</span>}}_{\mathrm{<span\; class="notranslate">\; l</span>}}<span\; class="notranslate">\; ]</span>}_{\mathrm{<span\; class="notranslate">\; i</span>}}}^{<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; \&gamma;</span>}<span\; class="notranslate">\; )</span>}$ In the formula, A ₁ =[I _l 0]A, F _l is (l-1)×l matrix, [X] _i ^(γ) represents the ith row of matrix X, Indicates the estimated value of the signal correlation function; Arithmetic average the estimated standard deviations of the incident angles of all subarray pairs, and this average is used as the final estimated standard deviation; 10) by step 8) and step 9) calculate the acoustic incidence angle estimated value The final estimated standard deviation of ; 11) Repeat step 4)-step 10) for all possible subarray selection methods to obtain the final estimated standard deviation of the incident angles for all possible subarray selection methods. 5. The method for processing high-resolution depth-scanning side-scan sonar signals according to claim 4, wherein in step d): comparing the incidence of all possible sub-array selection modes obtained in the previous step 11) The standard deviation of the final estimated standard deviation of the incident angle is determined, and the selection method of the sub-array with the smallest standard deviation of the final estimated standard deviation of the incident angle is determined, and this sub-array selection method is selected as the working mode of the sonar.