A Sound Source Localization Method in Shallow Sea Environment

Abstract

The present invention relates to a method for sound source localization in a shallow sea environment, comprising: acquiring the sound pressure of a target signal; acquiring shallow sea environment parameters and calculating an initial sound speed according to the shallow sea environment parameters; constructing a four-layer horizontally stratified sound speed linear model according to the shallow sea environment parameters and the initial sound speed; constructing a first sound wave equation with sound speed variation according to the four-layer horizontally stratified sound speed linear model; performing Hankel transform on the first sound wave equation to obtain a second sound wave equation related only to depth, and then obtaining a numerical solution of the Green's function; substituting the numerical solution of the Green's function into an iterative Bayesian focusing algorithm for calculation to obtain the Bayesian posterior mean, i.e., the estimated position of the target signal. Compared with the prior art, the present invention can effectively improve the efficiency and accuracy of sound source localization in a shallow sea environment.

Description

Technical Field

[0001] The present invention relates to the field of shallow - sea positioning, and particularly to a method for sound - source positioning in a shallow - sea environment. Background Art

[0002] Due to the complexity of the ocean environment, such as complex sound - speed profiles and irregular bottom reflections, sound - source positioning in shallow seas is a challenging task. Due to the boundary and medium undulation effects in the shallow - sea environment, multi - path time - delay spread and Doppler frequency - shift spread are serious, and it shows fast time - varying characteristics. On the other hand, most traditional sound - source positioning algorithms are based on the subspace theory framework, and there are many deficiencies in classical subspace - type algorithms, and some of these deficiencies cannot be compensated for by algorithm improvement. For example, they require a large number of independent and identically - distributed sampling data, the estimation accuracy is limited by the "Rayleigh limit", they are sensitive to noise, have high signal - to - noise ratio requirements, etc., resulting in a significant decline in estimation performance. Summary of the Invention

[0003] The purpose of the present invention is to overcome the above - mentioned defects existing in the prior art and provide a method for sound - source positioning in a shallow - sea environment.

[0004] The purpose of the present invention can be achieved through the following technical solutions:

[0005] A method for sound - source positioning in a shallow - sea environment includes the following steps:

[0006] S1. Obtain the sound pressure of the target signal;

[0007] S2. Obtain shallow - sea environment parameters and calculate the initial sound speed according to the shallow - sea environment parameters;

[0008] S3. Construct a four - layer horizontally stratified sound - speed linear model according to the shallow - sea environment parameters and the initial sound speed;

[0009] S4. Construct a first sound - wave equation with sound - speed variation according to the four - layer horizontally stratified sound - speed linear model;

[0010] S5. Perform Hankel transform on the first sound - wave equation to obtain a second sound - wave equation related only to depth, and then obtain the numerical solution of the Green's function;

[0011] S6. Substitute the numerical solution of the Green's function into the iterative Bayesian focusing algorithm for calculation to obtain the Bayesian posterior mean, that is, the estimated position of the target signal.

[0012] Further, the shallow - sea environment parameters include seawater salinity, seawater temperature, seawater density, and seawater depth.

[0013] Further, the calculation expression of the initial sound speed is:

[0014] c(z, T, S) = 1449.2 + 4.6T - 0.055T 2 + 0.00029T 3

[0015] +(1.34 - 0.01T)(S - 35) + 0.016z

[0016] In the formula, S represents the seawater salinity, T represents the seawater temperature, and z represents the seawater depth.

[0017] Furthermore, the expression of the four - layer horizontal stratified sound speed linear model is:

[0018]

[0019] In the formula, c 1 represents the sound speed of the first horizontal layer, c 2 represents the sound speed of the third horizontal layer, c 3 represents the sound speed of the fourth horizontal layer, and z represents the depth.

[0020] Furthermore, the expression of the first sound wave equation is:

[0021]

[0022] In the formula, c(z) represents the four - layer horizontal stratified sound speed linear model, represents the displacement potential at position r at time t, δ represents the Dirac function, S t represents the sound source function in the time domain, represents the Laplace operator, r 0 represents a certain determined position in the sound field.

[0023] Furthermore, the expression of the second sound wave equation is:

[0024]

[0025] In the formula, z represents the depth, z 0 represents the depth of a certain determined position in the sound field, k r represents the horizontal wave number, S f represents the sound source function in the frequency domain, and δ represents the Dirac function.

[0026] Furthermore, the target signal is acquired by a hydrophone array.

[0027] Compared with the prior art, the present invention has the following beneficial effects:

[0028] The present invention designs a four-layer medium model to simulate the shallow sea environment, then uses the full-wave simulation technology in the time domain to simulate the propagation of sound in the four-layer model environment, and finally realizes the sound source position estimation through the iterative Bayesian focusing algorithm, which can effectively improve the efficiency and accuracy of sound source localization in the shallow sea environment. Brief Description of the Drawings

[0029] Figure 1 It is a schematic flow chart of the present invention.

[0030] Figure 2 It is a schematic diagram of the four-layer sound speed stratification in the shallow sea environment.

[0031] Figure 3 It is a schematic flow chart of the iterative Bayesian focusing algorithm. Detailed Embodiment

[0032] The present invention will be described in detail below with reference to the drawings and specific embodiments. This embodiment is implemented on the premise of the technical solution of the present invention, and gives the detailed implementation manner and specific operation process, but the protection scope of the present invention is not limited to the following embodiments.

[0033] As Figure 1 shown, this embodiment provides a sound source localization method in a shallow sea environment, including the following steps:

[0034] Step S1: Obtain the sound pressure of the target signal;

[0035] Step S2: Obtain the shallow sea environment parameters, and calculate the initial sound speed according to the shallow sea environment parameters;

[0036] Step S3: Construct a four-layer horizontal stratified sound speed linear model according to the shallow sea environment parameters and the initial sound speed;

[0037] Step S4: Construct a first sound wave equation with sound speed variation according to the four-layer horizontal stratified sound speed linear model;

[0038] Step S5: Perform Hankel transform on the first sound wave equation to obtain a second sound wave equation related only to depth, and then obtain the numerical solution of the Green's function;

[0039] Step S6: Substitute the numerical solution of the Green's function into the iterative Bayesian focusing algorithm for calculation to obtain the Bayesian posterior mean, that is, the estimated position of the target signal.

[0040] The following is the specific expansion of the positioning method:

[0041] Step S1 is to obtain the data of the target signal. A hydrophone array and a data analysis center are arranged in the shallow sea area to be located. The hydrophone array is equipped with sensors that can measure the sound pressure of the signal, and each element transmits the data back to the data analysis center. The signal collected by the hydrophone array is a time-domain signal, which needs to be converted to the frequency domain for calculation in the positioning step. The classical Hanning window function is selected to truncate the time-domain signal, and then the truncated time-domain signal is converted to the frequency-domain signal through the fast Fourier transform. Each segment is a snapshot.

[0042] Thus, the observation data p of the hydrophone can be obtained.

[0043] Step S2 is to obtain the shallow sea environmental parameters. Sensors are arranged in the shallow sea area to be located to measure ocean parameters, including seawater salinity, seawater temperature, seawater density, and seawater depth. The calculation expression for the initial sound speed is:

[0044] c(z,T,S) = 1449.2 + 4.6T - 0.055T 2 + 0.00029T 3 +(1.34 - 0.01T)(S - 35)+0.016z

[0045] In the formula, S represents the seawater salinity, T represents the seawater temperature, and z represents the seawater depth.

[0046] Step S3 is to model a four-layer stratified sound speed linear model. As Figure 2 shown, in this embodiment, a four-layer medium model is introduced to simulate the shallow sea environment. This four-layer medium model can approximate the shallow sea environment, especially the shallow sea environment in summer. It is assumed that the sound speed is horizontally stratified; the vertical profile in the water column consists of two independent constant layers, and the sound speed changes linearly between the constant layers. In addition, a semi-infinite fluid bottom with a constant sound speed is also considered.

[0047] Then, according to the measured ocean parameters and the initial sound speed, the sound speed of each layer is linearly approximated, and its expression is:

[0048]

[0049] In the formula, c 1 represents the sound speed of the first horizontal layer, c 2 represents the sound speed of the third horizontal layer, c 3 represents the sound speed of the fourth horizontal layer, and z represents the depth.

[0050] Step S4 is to model a four-layer stratified medium sound propagation model. According to the sound speed of the four-layer stratified model, the first sound wave equation with sound speed variation can be obtained:

[0051]

[0052] In the formula, c(z) represents the linear model of sound speed for four-layer horizontal stratification, represents the displacement potential at position r at time t, δ represents the Dirac function, and S t represents the sound source function in the time domain, represents the Laplace operator.

[0053] Perform Fourier transform on both sides of the first sound wave equation to obtain the Helmholtz equation of the frequency-domain sound field :

[0054]

[0055] In the formula, k(z) represents the wave number related to depth.

[0056] Step S5 is to simplify the frequency-domain sound field by applying the Hankel transform.

[0057] The Hankel transform is:

[0058]

[0059] In the formula, the cylindrical coordinate system is selected, that is, r = (r, z, φ), r and φ represent the horizontal range and azimuth angle respectively, and z represents the depth.

[0060] Apply the Hankel transform to the Helmholtz equation to obtain the second sound wave equation that only depends on depth:

[0061]

[0062] In the formula, k r represents the horizontal wave number, and the Green's function H of the numerical solution can be obtained through this formula.

[0063] Step S6 is to substitute the observed data p of the hydrophone and the numerical solution H of the Green's function into the iterative Bayesian focusing algorithm for calculation to obtain the posterior mean of the sound source coefficient c, from which the estimated position of the target signal can be obtained. The specific process is as Figure 3 shown:

[0064] Step S61: Design an initial aperture function where

[0065] The prior distribution of the sound source coefficient can be obtained from obtained,

[0066] where

[0067] Step S62: Set k = 0. From and the Bayesian framework can obtain α 2 and the posterior mean where β 2 is the noise variance;

[0068] Step S63: Repeat steps S64 to S67 until convergence;

[0069] Step S64: Dok = k + 1

[0070] Step S65: Estimate the relative intensity as Set where 0 < ε 1 << 1;

[0071] Step S66: Update the aperture function to and construct a diagonal matrix whose j-th diagonal element is

[0072] Step S67: Obtain a new prior distribution from the updated aperture function and relative intensity, and thus calculate a new posterior mean

[0073] Step S68: Convergence criterion: When is less than a given threshold ε, 0 < ε < 1, stop the iteration;

[0074] Step S69: Use the posterior mean as an estimate of the target signal position.

[0075] In this step, through the joint estimation of the aperture function and the prior, the sound source distribution is gradually focused on the region of interest during the iteration process, promoting the sparsity of the sound source coefficients, and finally estimating the position of the sound source.

[0076] The above has described in detail the preferred specific embodiments of the present invention. It should be understood that those of ordinary skill in the art can make many modifications and variations according to the concept of the present invention without creative efforts. Therefore, all technical solutions that can be obtained by those skilled in the art in the technical field based on the concept of the present invention through logical analysis, reasoning or limited experiments on the basis of the prior art should be within the protection scope determined by the claims.

Claims

1. A method for sound source localization in a shallow sea environment, characterized in that, it includes the following steps: S1. Obtain the sound pressure of the target signal; S2. Obtain the shallow sea environment parameters, and calculate the initial sound speed according to the shallow sea environment parameters. The shallow sea environment parameters include seawater salinity, seawater temperature, seawater density and seawater depth. The calculation expression of the initial sound speed is: c(z, T, S) = 1449.2 + 4.6T - 0.055T 2 + 0.00029T 3 +(1.34 - 0.01T)(S - 35)+0.016z In the formula, S represents seawater salinity, T represents seawater temperature, and z represents seawater depth; S3. Construct a four-layer horizontally stratified sound speed linear model according to the shallow sea environment parameters and the initial sound speed. The expression of the four-layer horizontally stratified sound speed linear model is: where c 1 represents the sound velocity of the first horizontal layer, c 2 represents the sound velocity of the third horizontal layer, c 3 represents the sound velocity of the fourth horizontal layer; S4. Construct a first acoustic wave equation with sound speed variation according to the four-layer horizontally stratified sound speed linear model. The expression of the first acoustic wave equation is: where \(c(z)\) represents the linear model of sound speed for four horizontal layers, represents the displacement potential at position \(r\) at time \(t\), \(\delta\) represents the Dirac function, \(S\) t represents the sound source function in the time domain, represents the Laplace operator, \(r\) 0 represents a certain position in the sound field; S5. Perform Hankel transform on the first acoustic wave equation to obtain a second acoustic wave equation related only to depth, and then obtain the numerical solution of the Green's function. The expression of the second acoustic wave equation is: where z 0 represents the depth of a certain definite position in the sound field, k r represents the horizontal wave number, S f represents the sound source function in the frequency domain, δ represents the Dirac function, and k(z) represents the wave number related to the depth; S6. Substitute the numerical solution of the Green's function into the iterative Bayesian focusing algorithm for calculation to obtain the Bayesian posterior mean, that is, the estimated position of the target signal.

2. The method for sound source localization in a shallow sea environment according to claim 1, characterized in that, the target signal is obtained through a hydrophone array.

Images