CN111414580B - Reverberation suppression method under low signal-to-mixing ratio condition - Google Patents
Reverberation suppression method under low signal-to-mixing ratio condition Download PDFInfo
- Publication number
- CN111414580B CN111414580B CN202010219935.4A CN202010219935A CN111414580B CN 111414580 B CN111414580 B CN 111414580B CN 202010219935 A CN202010219935 A CN 202010219935A CN 111414580 B CN111414580 B CN 111414580B
- Authority
- CN
- China
- Prior art keywords
- reverberation
- matrix
- iteration
- result
- mixing ratio
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 230000001629 suppression Effects 0.000 title claims abstract description 49
- 238000000034 method Methods 0.000 title claims abstract description 46
- 238000001514 detection method Methods 0.000 claims abstract description 20
- 230000006641 stabilisation Effects 0.000 claims abstract description 12
- 238000011105 stabilization Methods 0.000 claims abstract description 12
- 239000011159 matrix material Substances 0.000 claims description 79
- 238000001228 spectrum Methods 0.000 claims description 29
- 238000000354 decomposition reaction Methods 0.000 claims description 20
- 238000004364 calculation method Methods 0.000 claims description 14
- 230000009977 dual effect Effects 0.000 claims description 9
- 239000013598 vector Substances 0.000 claims description 9
- 238000010276 construction Methods 0.000 claims description 2
- 230000003595 spectral effect Effects 0.000 claims description 2
- 230000000087 stabilizing effect Effects 0.000 abstract description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- OAICVXFJPJFONN-UHFFFAOYSA-N Phosphorus Chemical compound [P] OAICVXFJPJFONN-UHFFFAOYSA-N 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000003313 weakening effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Computational Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computing Systems (AREA)
- Algebra (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Measurement Of Velocity Or Position Using Acoustic Or Ultrasonic Waves (AREA)
Abstract
The invention relates to a reverberation suppression method under the condition of low signal-to-mixing ratio, which decomposes reverberation into a reverberation stable part and a reverberation dynamic part. The reverberation stabilization part is mainly a stabilization part of the sea bottom reverberation, the sea surface reverberation and the volume reverberation. The reverberation dynamics part is mainly the dynamics part of the sea reverberation and the volume reverberation (including the reverberation caused by the moving object). The reverberation stabilizing part is a main component of reverberation, so that the signal-to-mixing ratio of the obtained reverberation dynamic part is remarkably improved, and the reverberation suppression under the condition of low signal-to-mixing ratio is realized. The method has better universality because the method has no obvious requirements on the form of an active detection transmitting signal, the motion speed of a target and the like. The method has good reverberation suppression effect and wide application range, and is greatly helpful for improving the detection performance of a sonar system.
Description
Technical Field
The invention belongs to the fields of sonar technology, radar technology, image processing, signal processing and the like, and relates to a reverberation suppression method under the condition of low signal-to-mixing ratio, which is suitable for reverberation suppression and interference suppression in active detection.
Background
In active detection of ocean, the boundary of ocean and the object in water can generate scattering or reflection action on the actively emitted sound wave, resulting in the existence of active detection reverberation. In a low signal-to-noise ratio environment, the characteristics of the scattered or reflected signals and the target echo are not obviously different in a time domain and a frequency domain. On the other hand, since the reverberation intensity is often higher than the target echo signal intensity, reverberation suppression cannot be performed by simply setting a threshold.
For static targets, the main method of reverberation suppression is to use short pulses or to use pulse compression to reduce the surface area of the scatterers at any one time, thereby weakening the intensity of the reverberation. When the LFM signal is used as an active detection signal, based on the difference between the target reflection capability and the scattering capability of marine scatterers (including organisms on the sea surface, the sea bottom, and in a water body), the fractional fourier transform is used to perform good focusing characteristics on the LFM signal, and Sparse processing is used in the fractional fourier domain to perform reverberation suppression and target azimuth estimation, which is disclosed in Applied acoustics. Experimental data analysis results show that in a shallow sea strong reverberation environment, the method has a good reverberation suppression effect on both static and dynamic targets, and can achieve a reverberation suppression function, but for weak target signals, the reverberation suppression method is prone to failure. For dynamic targets, reverberation is suppressed in The Doppler domain based on The Doppler characteristics of The target signal, see "gap reduction using Doppler sensor in a polar environment", published in The Journal of The acoustic Society of America. Experiments show that the main factors influencing the Doppler characteristic of the target comprise: 1) Using a form of signal such as a CW signal, an LFM signal, or the like; 2) The frequency of the active probing signal and the speed of motion of the target. When the parameters of the detection signal are fixed, the moving speed of the target directly affects the strength of the doppler characteristic of the received signal, and when the target speed is low, the doppler characteristic caused by sea surface fluctuation often overwhelms the doppler characteristic of the echo of the target, resulting in a lower signal-to-noise ratio, and thus reverberation suppression cannot be performed.
The reverberation suppression method based on the fractional order fourier domain has a requirement on the strength of a target signal, and the reverberation suppression method based on the doppler domain also has a requirement on the speed of a target, the form of an active detection signal and the like. Therefore, under the environment with low signal-to-mixing ratio, the current method is difficult to obtain better reverberation suppression effect.
Disclosure of Invention
Technical problem to be solved
In order to avoid the defects of the prior art, the invention provides a reverberation suppression method under the condition of low signal-to-reverberation ratio, which aims at the conditions that the current reverberation suppression method has high requirements on target signal strength, active detection signal form and target speed and has limited reverberation suppression capability. In order to improve the strong reverberation suppression capability under the condition of low signal-to-reverberation ratio, the invention provides a reverberation suppression method under the condition of low signal-to-reverberation ratio, which is suitable for various active detection signal forms, has universality on the motion speed of a target, and can remarkably suppress reverberation.
Technical scheme
A reverberation suppression method under the condition of low signal-to-mixing ratio is characterized by comprising the following steps:
step 1: for actively detecting received data D j Performing space spectrum estimation to obtain a space spectrum matrix Z of the j frame for actively detecting different distances and different directions j With dimension B 0 ×R 0 ，R 0 Is the number of distance grids, B 0 The number of azimuth grids; j is the number of detection frames, j =1,2 \ 8230n;
step 2: spatial spectral matrix Z obtained from N-frame active sounding j To select N 0 Frame construction dimension of N 0 ×B 0 R 0 Reverberation matrix Q of (a):
wherein:
is a reverberation submatrix of the reverberation matrix Q, and N 0 ≤N；
And step 3: decomposing the reverberation matrix Q into a reverberation stabilization part matrix R s And a reverberation dynamic part matrix R d The specific decomposition process is as follows:
1) Let M be the reverberation matrix variable and matrix Y be the dual matrix of M, whose dimensions are all N 0 ×B 0 R 0 (ii) a Let M, R d And Y is initialized to M 0 ＝Q、
And &>
Selecting an iteration termination constant delta, wherein the value of the iteration termination constant delta meets delta e (10) -5 ,10 -1 ) (ii) a Let k be the number of iterations, and take k =1;
2) Evaluating the k-th iteration result L of the intermediate variable L k . Iterative computation of the result M with the k-1 th iteration of M k-1 、R d Is calculated as a result of the (k-1) th iteration of (R) d,k-1 And the k-1 th iterative computation result Y of Y k-1 Obtaining the k-th iteration result L of the intermediate variable L k The calculation expression is:
L k ＝M k-1 -R d,k-1 -Y k-1
3) Solving the reverberation stability part matrix R s Is calculated as a result R of the kth iteration of (1) s,k . To L k Performing singular value decomposition, and calculating an expression:
U∑V H ＝svd(L k )
then based on the decomposition result of the singular value, obtaining R according to calculation s,k ：
Wherein rho is a step factor, and the initial value of the rho satisfies rho 0 E (0.1, 5); svd (-) represents singular value decomposition of matrix, U and V are respectively left singular value vector and right singular value vector of singular value decomposition, and dimensionality is N 0 ×N 0 And B 0 R 0 ×B 0 R 0 (ii) a Sigma means N 0 ×B 0 R 0 A singular value matrix of dimensions; diag (·) represents the operation of taking the diagonal elements of the matrix; (. Cndot.) H Representing a transpose; h ξ [·]For a threshold operator, let x and ξ be the to-be-processed quantity and the corresponding threshold value of the operator, respectively, then the operator can be expressed as:
4) Solving the reverberation dynamic part matrix R d Of the kth iteration result R d,k ：
Wherein
a 1 Is a constant, the value of which satisfies a 1 ∈(0.01,5)；
5) Calculating the kth iteration result of the dual matrix Y:
Y k ＝Y k-1 +ρ k-1 [M k-1 -R s,k -R d,k ]
6) Calculate the kth iteration residual E of M M,k And dual residual E d,k ：
E M,k ＝M k-1 -(R s,k +R d,k )
E d,k ＝R d,k -R d,k-1
7) Updating the k-th iteration calculation values ρ of the parameters ρ, α and M k 、α k And M k ：
ρ k ＝τρ k-1
γ k ＝||M k-1 || max /a 2
M k ＝M k-1 +γ k E
Wherein tau is a constant, and the value of tau belongs to (0.1, 5); i | · | purple wind max Representing operations taking the maximum value of the matrix elements, a 2 Is constant, the value satisfies a 2 ∈(1,200)；γ 1 Is constant, and the value satisfies gamma 1 E is (0.0001, 0.1), E is an identity matrix with dimension N 0 ×B 0 R 0 ；
8) Judging whether the decomposition calculation of the reverberation matrix Q is finished: judge E | | M,k || 2 If delta is not more than or equal to delta, entering the step 4 if delta is not more than delta, otherwise increasing the iteration number k by 1, and continuing to circulate the operation from the step 2) to the step 8) until the condition of | | E is met M,k || 2 δ is less than or equal to, and the number of iterations at this time is marked as k 0 (ii) a Wherein | · | purple 2 Is 1 2 Performing norm operation; completing the decomposition calculation of the reverberation matrix Q and the reverberation dynamic part matrix
Reverberation stabilization part matrix
Step 4, obtaining a reverberation suppression result: to pair
(j 1 ＝1,2,…,N 0 ) Setting hard threshold Th respectively 0 ＝a 3 ，a 3 Is a constant whose value satisfies->
Will->
Each element of the matrix is ≥ er>
(n 1 ,n 2 ) And a threshold value Th 0 And judging that:
wherein n is 1 ＝1,2,…,B 0 ，n 2 ＝1,2,…,R 0 。
Is paired with>
And (5) performing reverberation suppression.
The spatial spectrum estimation in the step 1 adopts a beam forming method CBF to carry out spatial spectrum azimuth estimation to obtain a spatial spectrum matrix Z of the j frame active detection j 。
In the step 2, N is selected from the N frame spatial spectrum matrixes 0 When frames, adjacent framesThe following relationships are required:
wherein v is 0 Is the target movement speed, T 0 For the time interval of two adjacent frames, e 0 Is the minimum distinguishability of the difference of the target positions between two adjacent frames.
The spatial spectrum matrix is used in the step 2
When forming the reverberation matrix Q, Z j1 One column vector forming Q with the orientation as a basic unit, i.e.
Wherein m is 1 ＝1,2,…,B 0 ；m 2 ＝1,2,…,R 0 ；
Denotes the j (th) 1 Frame space spectrum matrix->
M th 1 An azimuth R 0 An orientation matrix composed of spatial spectra of the distances; />
Represents->
M in 2 An azimuth spectrum of individual distances.
Advantageous effects
According to the method for suppressing the reverberation under the condition of the low signal-to-mixing ratio, the reverberation of the marine environment has stability, and the reverberation caused by the moving target has dynamics, so that the reverberation is decomposed into a reverberation stable part and a reverberation dynamic part by using the method for suppressing the reverberation under the condition of the low signal-to-mixing ratio. The reverberation stabilization part is mainly a stabilization part of the seabed reverberation, the sea surface reverberation and the volume reverberation. The reverberation dynamics part is mainly the dynamics part of the sea reverberation and the volume reverberation (including the reverberation caused by the moving object). The reverberation stabilizing part is a main component of reverberation, so that the signal-to-mixing ratio of the obtained reverberation dynamic part is remarkably improved, and the reverberation suppression under the condition of low signal-to-mixing ratio is realized.
FIG. 2 shows the effect of the reverberation suppression method under the condition of low signal-to-mixing ratio according to the present invention, wherein FIG. 2 (a) shows the reverberation space spectrum of the 5 th frame, i.e. Z 5 The result of (1); FIG. 2 (b) shows the reverberant dynamics part of frame 5, i.e., Z 5,(d,19) The result of (2); FIG. 2 (c) shows the reverberation stabilization part of frame 5, i.e. Z 5,(s,19) The result of (1); FIG. 2 (d) shows the result of suppressing the spatial spectrum reverberation of the reverberation space of frame 5, i.e. psi 5,d The result of (1). After reverberation suppression under the low signal-to-reverberation ratio condition in fig. 2 (a), there can be the result of fig. 2 (d), which includes a clear target echo signal. Therefore, the reverberation suppression method under the condition of low signal-to-mixing ratio achieves the suppression effect on reverberation under the environment with low signal-to-mixing ratio.
The reverberation suppression method under the condition of low signal-to-mixing ratio performs reverberation suppression by taking the general characteristics of the response of the ocean and the moving target to the active detection signal as a starting point, so that the reverberation suppression method has a good reverberation suppression effect. Meanwhile, the method has no obvious requirements on the form of the active detection transmitting signal, the motion speed of the target and the like, so the method has better universality. The method has good reverberation suppression effect and wide application range, and is greatly helpful for improving the detection performance of a sonar system.
Drawings
Fig. 1 is a flowchart of a reverberation suppression method under a low signal-to-mixing ratio condition according to the present invention; the matrix of the reverberation dynamic part is obtained through iteration, a proper threshold value is set according to the strength of the reverberation background of the part, and the method related by the invention can realize the suppression of the reverberation.
Fig. 2 (a) is a spatial spectrum result of reverberation of the 5 th frame in the reverberation suppression method under the condition of low signal-to-mixing ratio according to the present invention;
fig. 2 (b) is a 5 th frame reverberation dynamic part obtained by the reverberation suppression method under the condition of low signal-to-mixing ratio according to the present invention;
fig. 2 (c) is a 5 th frame reverberation stabilization part obtained by the reverberation suppression method under the condition of low signal-to-mixing ratio according to the present invention;
fig. 2 (d) is a reverberation suppression result of the 5 th frame obtained by the reverberation suppression method under the condition of low signal-to-mixing ratio according to the present invention;
1 is a target echo signal; 2 is a reverberant dynamic background
Detailed Description
The invention will now be further described with reference to the following examples, and the accompanying drawings:
the reverberation suppression method under the condition of low signal-to-mixing ratio comprises the following steps:
the method comprises the following steps: for data D received by each active detection j (j is the number of detection frames, j =1,2 \ 823030; 30); performing space spectrum orientation estimation by using a traditional beam forming method (CBF) to obtain a space spectrum matrix Z of the j frame active detection j With dimension B 0 ×R 0 Wherein R is 0 ＝40，B 0 ＝361。
Step two: the time interval of actively detecting two adjacent frames is T 0 =2s, underwater target movement velocity v 0 =1m/s, from Z according to the requirements of the following formula j Selecting 20 frames of spatial spectrum matrix:
and Z is j Constructing a reverberation matrix Q according to the following two formulas:
the dimensions of Q are 20 x 14440,
Q＝[Z 1 ,Z 2 ,…Z j1 ,…,Z 20 ],j 1 ＝1,2,…,20 ， (1)
then
Is the reverb submatrix of the reverb matrix Q.
Wherein: v. of 0 Is the target movement speed, T 0 For the time interval of two adjacent frames, e 0 The minimum distinguishability of the target position difference between two adjacent frames is obtained; m is 1 ＝1,2,…,B 0 ；m 2 ＝1,2,…,R 0 ；
Represents the j1 th frame spatial spectrum matrix->
M th 1 An azimuth R 0 An orientation matrix composed of spatial spectra of the distances; />
Represents->
M in m 2 An orientation spectrum of the individual distances;
step three: decomposing the reverberation matrix Q into a reverberation stabilization part matrix R s And a reverberation dynamic part matrix R d The specific decomposition process is as follows:
1) Let M be the reverberation matrix variable and matrix Y be the dual matrix of M, all of whose dimensions are 20 × 14440; let M, R d And Y is initialized to M 0 ＝Q，R d,0 ＝0 20×14440 And Y 0 ＝0 20×14440 (ii) a Take iteration termination constant δ =10 -3 (ii) a Let k be the number of iterations and take k =1.
2) Evaluating the k-th iteration result L of the intermediate variable L k . Iterative computation of the result M with the k-1 th iteration of M k-1 、R d Is calculated as a result of the (k-1) th iteration of (R) d,k-1 And the k-1 th iterative computation result Y of Y k-1 Obtaining the k-th iteration result L of the intermediate variable L k The concrete calculation is shown as the formula (2)
L k ＝M k-1 -R d,k-1 -Y k-1 ， (2)
3) Solving the reverberation stability part matrix R s Is calculated as a result R of the kth iteration of s,k . To L k Performing singular value decomposition, calculating expression as shown in (3), and obtaining R according to calculation expression (4) based on the decomposition result of the singular value s,k ，
U∑V H ＝svd(L k )， (3)
Wherein rho is a step factor, and the value of the initial value of rho satisfies rho 0 =0.5; svd (-) represents singular value decomposition of matrix, U and V are respectively left singular value vector and right singular value vector of singular value decomposition, and dimensionality is N 0 ×N 0 And B 0 R 0 ×B 0 R 0 (ii) a Sigma denotes N 0 ×B 0 R 0 A singular value matrix of dimensions; diag (·) represents the operation of taking the diagonal elements of the matrix; (.) H Representing a transpose; h ξ [·]Is a threshold operator, and the specific expression is shown as formula (5) in the claims:
4) Solving the reverberation dynamic part matrix R d The result of the k-th iteration of (1)R d,k
Where λ =0.05.
5) Calculating the kth iteration result of the dual matrix Y:
Y k ＝Y k-1 +ρ k-1 [M k-1 -R s,k -R d,k ]， (7)
6) Calculate the kth iteration residual E of M M,k And dual residual E d,k ：
E M,k ＝M k-1 -(R s,k +R d,k )， (8)
E d,k ＝R d,k -R d,k-1 ， (9)
7) Updating the k-th iteration calculation values ρ of the parameters ρ, α and M k 、α k And M k ：
ρ k ＝τρ k-1 ， (10)
γ k ＝||M k-1 || max /a 2 (11)
M k ＝M k-1 +γ k E， (12)
Wherein τ =3, a 2 ＝100、γ 1 ＝0.01；||·|| max Representing the maximum operation of the matrix elements, E is the identity matrix with dimensions 20 x 14440.
8) Judging whether the decomposition calculation of the reverberation matrix Q is finished. Judge | | | E M,k || 2 If the delta is less than or equal to the preset value, entering the step four if the delta is less than or equal to the preset value, otherwise, increasing the iteration number k by 1, and continuing to circulate the operation from the process 2) to the process 8) until the < I > E is met M,k || 2 Less than or equal to delta, wherein | · |. Non-woven phosphor 2 Is 1 2 And (5) performing norm operation. Obtaining the final iteration number k 0 =19; there may be reverberation dynamic partial matrix
Reverberation stabilization part matrix
Step four: a reverberation suppression result is obtained. To pair
(j 1 =1,2, \ 8230;, 20) hard threshold value, respectively
Is paired and/or matched>
Each element of the matrix is ≥ er>
(n 1 ,n 2 ) Respectively using threshold values Th 0 Make a judgment of whether there is
Wherein n is 1 ＝1,2,…,361，n 2 ＝1,2,…,40。
I.e. is paired>
And (5) performing reverberation suppression. />
Claims (4)
1. A reverberation suppression method under the condition of low signal-to-mixing ratio is characterized by comprising the following steps:
step 1: for actively detecting received data D j Performing space spectrum estimation to obtain a space spectrum matrix Z of the j frame for actively detecting different distances and different directions j With dimension B 0 ×R 0 ，R 0 Is a distance gridNumber, B 0 The number of azimuth grids; j is the number of detection frames, j =1,2 \ 8230n;
step 2: spatial spectral matrix Z obtained from N-frame active sounding j To select N 0 Frame construction dimension of N 0 ×B 0 R 0 Reverberation matrix Q of (a):
wherein:
is a reverberation submatrix of the reverberation matrix Q, and N 0 ≤N；
And step 3: decomposing the reverberation matrix Q into a reverberation stabilization part matrix R s And a reverberation dynamic part matrix R d The specific decomposition process is as follows:
1) Let M be the reverberation matrix variable and matrix Y be the dual matrix of M, whose dimensions are all N 0 ×B 0 R 0 (ii) a Let M, R d And Y is initialized to M 0 ＝Q、
And
selecting an iteration termination constant delta, wherein the value of the iteration termination constant delta meets delta e (10) -5 ,10 -1 ) (ii) a Let k be the number of iterations, and take k =1;
2) Evaluating the k-th iteration result L of the intermediate variable L k : iterative computation of the result M with the k-1 th iteration of M k-1 、R d Is calculated as a result of the (k-1) th iteration of (R) d,k-1 And the k-1 th iterative computation result Y of Y k-1 Obtaining the k-th iteration result L of the intermediate variable L k The calculation expression is:
L k ＝M k-1 -R d,k-1 -Y k-1
3) Solving the reverberation stability part matrix R s Is calculated as a result R of the kth iteration of s,k : to L k Performing singular value decomposition, and calculating an expression:
U∑V H ＝svd(L k )
then based on the decomposition result of the singular value, obtaining R according to calculation s,k ：
Wherein rho is a step factor, and the initial value of the rho satisfies rho 0 E (0.1, 5); svd (-) represents singular value decomposition of the matrix, U and V are respectively a left singular value vector and a right singular value vector of the singular value decomposition, and the dimensionalities of the vectors are respectively N 0 ×N 0 And B 0 R 0 ×B 0 R 0 (ii) a Sigma means N 0 ×B 0 R 0 A singular value matrix of dimensions; diag (·) represents the operation of taking the diagonal elements of the matrix; (.) H Representing a transpose; h ξ [·]For a threshold operator, let x and ξ be the to-be-processed quantity of the operator and the corresponding threshold, respectively, then the operator can be expressed as:
4) Solving the reverberation dynamic part matrix R d Of the kth iteration result R d,k ：
Wherein
a 1 Is a constant, the value of which satisfies a 1 ∈(0.01,5)；
5) Calculating the kth iteration result of the dual matrix Y:
Y k ＝Y k-1 +ρ k-1 [M k-1 -R s,k -R d,k ]
6) Calculate the kth iteration residual E of M M,k And dual residual E d,k ：
E M,k ＝M k-1 -(R s,k +R d,k )
E d,k ＝R d,k -R d,k-1
7) Updating the k-th iteration calculation values ρ of the parameters ρ, α and M k 、α k And M k ：
ρ k ＝τρ k-1
γ k ＝||M k-1 || max /a 2
M k ＝M k-1 +γ k E
Wherein tau is a constant, and the value of tau belongs to (0.1, 5); i | · | live through max Representing maximum operation of matrix elements, a 2 Is constant, the value satisfies a 2 ∈(1,200)；γ 1 Is constant, and the value satisfies gamma 1 E is (0.0001, 0.1), E is an identity matrix with dimension N 0 ×B 0 R 0 ；
8) Judging whether the decomposition calculation of the reverberation matrix Q is finished: judge E | | M,k || 2 If delta is not more than or equal to delta, entering the step 4 if delta is not more than delta, otherwise increasing the iteration number k by 1, and continuing to circulate the operation from the step 2) to the step 8) until the condition of | | E is met M,k || 2 δ is less than or equal to, and the number of iterations at this time is marked as k 0 (ii) a Wherein | · | purple 2 Is 1 2 Performing norm operation; completing the decomposition calculation of the reverberation matrix Q and the reverberation dynamic part matrix
Reverberation stabilization part matrix
Step 4, obtaining a reverberation suppression result: to pair
Setting hard threshold Th respectively 0 ＝a 3 ，a 3 Is a constant having a value satisfying
Will be provided with
Each element of the matrix
And a threshold value Th 0 And judging that:
wherein n is 1 ＝1,2,…,B 0 ，n 2 ＝1,2,…,R 0 ，
Is a pair of
And (5) performing reverberation suppression.
2. The method for suppressing reverberation under low signal-to-mixing ratio as claimed in claim 1, wherein: the spatial spectrum estimation of the step 1 adopts a beam forming method CBF to carry out spatial spectrum azimuth estimation to obtain a spatial spectrum matrix Z of the j frame active detection j 。
3. The method for suppressing reverberation under low signal-to-mixing ratio as claimed in claim 1, wherein: in the step 2, N is selected from the N frame spatial spectrum matrixes 0 When frames are processed, the relationship between adjacent frames needs to be satisfied:
wherein v is 0 Is the target movement speed, T 0 Is the time interval of two adjacent frames, e 0 Is the minimum distinguishable degree of the target position difference between two adjacent frames.
4. The method for suppressing reverberation under low signal-to-mixing ratio as claimed in claim 1, wherein: the spatial spectrum matrix Z is used in the step 2 j1 When the reverberation matrix Q is constructed,
one column vector forming Q with the orientation as a basic unit, i.e.
Wherein m is 1 ＝1,2,…,B 0 ；m 2 ＝1,2,…,R 0 ；
Denotes the j (th) 1 Frame space spectrum matrix
M th 1 In a direction of R 0 An orientation matrix composed of spatial spectra of the distances;
to represent
M in 2 An azimuth spectrum of individual distances.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010219935.4A CN111414580B (en) | 2020-03-25 | 2020-03-25 | Reverberation suppression method under low signal-to-mixing ratio condition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010219935.4A CN111414580B (en) | 2020-03-25 | 2020-03-25 | Reverberation suppression method under low signal-to-mixing ratio condition |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111414580A CN111414580A (en) | 2020-07-14 |
| CN111414580B true CN111414580B (en) | 2023-03-24 |
Family
ID=71494707
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010219935.4A Active CN111414580B (en) | 2020-03-25 | 2020-03-25 | Reverberation suppression method under low signal-to-mixing ratio condition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111414580B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114818773B (en) * | 2022-03-12 | 2024-04-16 | 西北工业大学 | Low-rank matrix sparsity compensation method for improving reverberation suppression robustness |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101387701A (en) * | 2008-10-24 | 2009-03-18 | 西北工业大学 | Passive time reversal reverberation suppression method based on forward prediction |
| JP2013113866A (en) * | 2011-11-25 | 2013-06-10 | Nippon Telegr & Teleph Corp <Ntt> | Reverberation removal method, reverberation removal device and program |
| CN103487794A (en) * | 2012-06-13 | 2014-01-01 | 中国科学院声学研究所 | Underwater reverberation suppression method based on wavelet packet transforming |
| CN110189761A (en) * | 2019-05-21 | 2019-08-30 | 哈尔滨工程大学 | A kind of single channel speech dereverberation method based on greedy depth dictionary learning |
-
2020
- 2020-03-25 CN CN202010219935.4A patent/CN111414580B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101387701A (en) * | 2008-10-24 | 2009-03-18 | 西北工业大学 | Passive time reversal reverberation suppression method based on forward prediction |
| JP2013113866A (en) * | 2011-11-25 | 2013-06-10 | Nippon Telegr & Teleph Corp <Ntt> | Reverberation removal method, reverberation removal device and program |
| CN103487794A (en) * | 2012-06-13 | 2014-01-01 | 中国科学院声学研究所 | Underwater reverberation suppression method based on wavelet packet transforming |
| CN110189761A (en) * | 2019-05-21 | 2019-08-30 | 哈尔滨工程大学 | A kind of single channel speech dereverberation method based on greedy depth dictionary learning |
Non-Patent Citations (2)
| Title |
|---|
| 基于预白化和子空间方法的主动声呐混响抑制;洪小洋等;《网络新媒体技术》;20190515(第03期);全文 * |
| 被动时反混响抑制方法研究;郭国强等;《自然科学进展》;20090115(第01期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111414580A (en) | 2020-07-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109283536B (en) | Multi-beam sounding sonar water body imaging beam forming method | |
| CN101387701B (en) | Passive time reversal reverberation suppression method based on forward prediction | |
| CN109521430B (en) | Distance extension target detection method for inhibiting narrow-band interference | |
| CN111025273B (en) | Distortion drag array line spectrum feature enhancement method and system | |
| CN112612006B (en) | Deep learning-based non-uniform clutter suppression method for airborne radar | |
| CN105974400A (en) | Robust space-time detection method based on symmetric spectral characteristic | |
| CN111414580B (en) | Reverberation suppression method under low signal-to-mixing ratio condition | |
| Blomberg et al. | Adaptive sonar imaging using aperture coherence | |
| CN105676211B (en) | A kind of more bright spot target space-time detection methods based on contraction technology | |
| CN115993591A (en) | Self-adaptive interference suppression method based on characteristic subspace judgment | |
| Carbone et al. | A novel normalization algorithm based on the three-dimensional minimum variance spectral estimator | |
| Saucan et al. | Enhanced sonar bathymetry tracking in multi-path environment | |
| Das et al. | Analysis of effective signal design for active sensing of undersea objects/bottoms in tropical shallow waters | |
| CN113050098A (en) | Anti-frogman sonar reverberation suppression method based on block sparse robust principal component analysis | |
| Wang et al. | Coarseness in OTHR image and its application for diagonal loading factor determination | |
| Fang et al. | Experimental Demonstration of Depth-dependent Acoustic Detection Performance by Underwater Gliders | |
| Shi et al. | An improved generalized inverse beamforming-noise source localization method using acoustic vector sensor arrays | |
| Sun et al. | Robust adaptive beamforming method for active sonar in single snapshot | |
| CN114578333B (en) | Active sonar target dynamic and static identification method | |
| Guo et al. | Review of research on sonar imaging technology in China | |
| Bird et al. | Bathymetric sidescan sonar bottom estimation accuracy: Tilt angles and waveforms | |
| CN117907998B (en) | Shallow sea broadband sound source ranging method, medium and system | |
| Shang et al. | Research on Feature Extraction and Classification Recognition of Underwater Small Target Active Sonar | |
| CN117250604B (en) | Separation method of target reflection signal and shallow sea reverberation | |
| Lorenson et al. | 3D-Sonar image formation and shape recognition techniques |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |