Single-vector hydrophone orientation estimation method based on multi-joint understanding convolution projection CBF spatial spectrum

Abstract

The invention discloses a single-vector hydrophone orientation estimation method based on a multi-joint understanding convolution projection CBF spatial spectrum, and belongs to the technical field of underwater target detection. The method solves the problem that the existing method does not consider the sound energy information of the self-correlation part of each component in the sound field, so that the performance improvement of the existing method on weak target detection is limited. The method comprises the following steps of: firstly, carrying out multiple combined processing on sound pressure and vibration velocity channel receiving signals of the single-vector hydrophone; obtaining the maximum sound energy of the sound pressure and the combined vibration velocity, the sound energy of a sound pressure channel receiving signal and a vibration velocity x channel receiving signal, the sound energy of the sound pressure channel receiving signal and a vibration velocity y channel receiving signal, the sound energy of the vibration velocity x channel and the vibration velocity y channel, and the difference between the sound energy of the vibration velocity x channel and the sound energy of the sound pressure channel; and calculating a projection CBF spatial spectrum according to the calculated sound energy and the sound energy difference, and carrying out deconvolution on the projection CBF spatial spectrum based on a point spread function to obtain an underwater target orientation estimation result. The method can be applied to underwater target orientation estimation.

Description

Technical Field

[0001] The invention belongs to the technical field of underwater target detection, and in particular relates to a single-vector hydrophone azimuth estimation method based on multiple joint processing deconvolution projection CBF spatial spectrum. Background Art

[0002] Underwater target azimuth estimation is an important research topic in sonar technology. Usually, hydrophones are used to sense acoustic signals, and then signal processing is combined to extract useful information such as target azimuth. Vector hydrophones are composed of pressure hydrophones and velocity hydrophones. They can simultaneously and concurrently sense the sound pressure and particle velocity information of a certain point in the sound field. They have "8" receiving directivity, more comprehensive sound field information, and stronger noise suppression capabilities. Compared with pressure hydrophones, a single vector hydrophone can complete underwater target azimuth estimation, providing a new direction for underwater target azimuth estimation.

[0003] The characteristic of using a single vector hydrophone to estimate the direction of underwater targets is that the vector hydrophone can be used in non-cooperative systems or cooperative systems to provide target direction information for the system. The research results of Reference 1 (Hui Junying, Liu Hong, Yu Huabing, et al. A preliminary study on the joint information processing of sound pressure and vibration velocity and its physical basis [J]. Journal of Acoustics, 2000, 254.303-307;) and Reference 2 (Hawkes M, Nehorai A. Wideband source localization using a distributed acoustic vector-sensor array [J]. IEEE Trans. Signal Process., 2003, 51 (6): 1479-1491.) show that for the far-field sound field radiated by a finite-scale sound source, the sound pressure and vibration velocity are completely correlated; while for the isotropic noise field, the sound pressure and vibration velocity at the same point are uncorrelated, which theoretically reveals the anti-noise principle of the joint processing method of sound pressure and vibration velocity. Reference 3 (Sun Guiqing, Yang Desen, Zhang Lanyue, et al. Maximum likelihood ratio detection and maximum likelihood direction estimation based on vector hydrophone [J]. Journal of Acoustics, 2003, 1.66-72) obtained a spatial processing gain of 10-20 dB by using the sound intensity method through lake test data, further confirming that the joint processing method of sound pressure and vibration velocity has a strong noise suppression ability. Reference 4 (Bai Xingyu, Jiang Yu, Zhao Chunhui. Detection of the number of sources and direction estimation of acoustic vector arrays based on joint processing of sound pressure and vibration velocity [J]. Journal of Acoustics, 2008, 331.56-61) and Reference 5 (Shi S, Li Y, Yang D, et al., Sparse representation based direction-of-arrival estimation using circular acoustic vector sensor arrays [J]. Digit. Signal Prog., 2020, 99.) extended the idea of ​​sound intensity processing to different vector arrays, improving the detection performance of weak targets. However, the acoustic energy information of the autocorrelation part of each component in the sound field is still not taken into account. Therefore, the performance improvement of weak target detection is limited, and the performance of weak target detection needs to be further improved. Summary of the invention

[0004] The purpose of the present invention is to solve the problem that the existing methods do not consider the acoustic energy information of the autocorrelation part of each component in the sound field, resulting in limited improvement in the performance of weak target detection by the existing methods, and propose a single-vector hydrophone azimuth estimation method based on multiple joint processing deconvolution projection CBF spatial spectrum to improve the performance of weak target detection.

[0005] The technical solution adopted by the present invention to solve the above technical problems is: a single vector hydrophone azimuth estimation method based on multiple joint processing deconvolution projection CBF spatial spectrum, the method specifically comprises the following steps:

[0006] Step 1: The signals received by the single vector hydrophone at time t are recorded as p(t) and v x (t) and v y (t), where p(t) represents the received signal of the sound pressure channel, v x (t) represents the received signal of the vibration velocity x channel, v y (t) represents the received signal of the vibration velocity y channel;

[0007] Step 2: Perform multiple joint processing on the single vector hydrophone sound pressure channel and vibration velocity channel receiving signals to obtain the maximum sound energy I0 of the sound pressure and combined vibration velocity, the sound energy I0 of the sound pressure channel receiving signal and the vibration velocity x channel receiving signal px , the acoustic energy I of the signal received by the sound pressure channel and the signal received by the vibration velocity y channel py , the acoustic energy I of the vibration velocity x channel and the vibration velocity y channel xy , the difference between the acoustic energy of the vibration velocity x channel and the acoustic energy of the sound pressure channel I xx ;

[0008] According to I0, I px ,I py ,I xy and I xx Construct a high-order covariance matrix of a single-vector hydrophone, calculate the projected CBF weighted vector, and then calculate the projected CBF spatial spectrum based on the high-order covariance matrix and the projected CBF weighted vector;

[0009] Step 3: construct a point spread function according to the projection beam function, and deconvolve the projected CBF spatial spectrum based on the point spread function, and use the peak position in the azimuth spectrum obtained by deconvolution as the azimuth of the underwater target.

[0010] Furthermore, the received signal of the sound pressure channel, the received signal of the vibration velocity x channel, and the received signal of the vibration velocity y channel are respectively:

[0011] p(t)=s(t)+n p (t)

[0012] v x (t) = s(t)cosθ+n x (t) (1)

[0013] v y (t) = s(t) sinθ + n y (t)

[0014] Among them, s(t) is the target signal, θ is the incident direction of the target signal, t is the time, n p (t), n x (t) and n y (t) represents the ambient noise signals received by the sound pressure and vibration velocity x and y channels respectively.

[0015] Furthermore, in the step 2, multiple joint processing is performed on the single vector hydrophone sound pressure channel and the vibration velocity channel receiving signal to obtain the maximum sound energy I0 of the sound pressure and the combined vibration velocity, the sound energy I px , the acoustic energy I of the signal received by the sound pressure channel and the signal received by the vibration velocity y channel py , the acoustic energy I of the vibration velocity x channel and the vibration velocity y channel xy , the difference between the acoustic energy of the vibration velocity x channel and the acoustic energy of the sound pressure channel I xx ; The specific process is:

[0016] Step 21: The received signal v of the vibration speed x channel is x (t) and the received signal v of the vibration velocity y channel y (t) is projected to the φ direction, and then the combined vibration velocity signal v is obtained based on the projection result. c (t);

[0017] Then based on the sound pressure channel receiving signal and the combined vibration velocity signal v c (t) Maximum sound energy I0 for constructing sound pressure and combined vibration velocity;

[0018] Step 22: The sound energy of the sound pressure channel receiving signal and the vibration velocity x channel receiving signal, the sound energy of the sound pressure channel receiving signal and the vibration velocity y channel receiving signal, the sound energy of the vibration velocity x channel and the vibration velocity y channel, and the difference between the sound energy of the vibration velocity x channel and the sound energy of the sound pressure channel are respectively:

[0019]

[0020]

[0021]

[0022]

[0023] In the formula, I px It represents the sound energy of the received signal of the sound pressure channel and the received signal of the vibration velocity x channel, I py It represents the acoustic energy of the received signal of the sound pressure channel and the received signal of the vibration velocity y channel, I xy Represents the acoustic energy of the vibration velocity x channel and the vibration velocity y channel, I xx It represents the difference between the acoustic energy of the velocity x channel and the acoustic energy of the sound pressure channel, I x Indicates vibration velocity x channel sound energy, Ip represents the sound energy of the sound pressure channel, is the noise power of the sound pressure channel, is the noise power of the vibration channel, and

[0024] Furthermore, the combined vibration velocity signal v is obtained based on the projection result. c (t), specifically:

[0025]

[0026] Where φ is the projection angle.

[0027] Furthermore, the sound pressure channel receives the signal and the combined vibration velocity signal v c (t) The maximum sound energy I0 of the combined sound pressure and vibration velocity is constructed as follows:

[0028] Calculate the sound energy I of the combined sound pressure and vibration velocity pvc :

[0029]

[0030] In the formula, (·) H is the conjugate transpose operation; Indicates the power of the target signal;

[0031] And traverse the projection angle φ to obtain the maximum sound energy I0 of the combined sound pressure and vibration velocity:

[0032]

[0033] Further, according to I0, I px ,I py ,I xy and I xx Construct a high-order covariance matrix of a single-vector hydrophone. The specific process is:

[0034] The acoustic energy information I0, I px ,I py ,I xy and I xx Make a vector z:

[0035]

[0036] Where b(θ) is the steering vector matrix of the high-order covariance matrix, b(θ) = [1, cosθ, sinθ, cos(2θ), sin(2θ)] T , the superscript T represents the transpose of the matrix;

[0037] make Among them, u1 is the left singular vector of z, λ1 is the singular value, and β is the linear coefficient;

[0038] Then the high-order covariance matrix R of the vector hydrophone is:

[0039]

[0040] Here, E[·] represents expectation.

[0041] Furthermore, the projected CBF weighted vector is calculated, and then the projected CBF spatial spectrum is calculated based on the high-order covariance matrix and the projected CBF weighted vector; the specific process is:

[0042] Project the steering vector matrix of the high-order covariance matrix into the signal subspace to obtain the projected CBF weighted vector

[0043]

[0044] In the formula, is the scanning angle, U s represents the signal subspace;

[0045] Then the projected CBF spatial spectrum for:

[0046]

[0047] Here, |·| represents the value of the calculated determinant.

[0048] Furthermore, the specific process of step three is:

[0049] Step 31: Re-express formula (12) as:

[0050]

[0051] in, is the distribution function of the target signal power in the azimuth plane, * represents the convolution calculation symbol, is the projection beam function, is the point spread function of the projected CBF;

[0052] Step 32: Based on point spread function Deconvolute the projected CBF spatial spectrum to obtain a new azimuth spectrum

[0053] Furthermore, the distribution function of the target signal power in the angular plane is:

[0054]

[0055] Here, δ(·) represents the unit impulse function.

[0056] Furthermore, the specific process of step 32 is as follows:

[0057] Step 321: Initialize the number of iterations n = 1, and initialize the initial value of the target signal power distribution in the azimuth plane to Set the total number of iterations to M;

[0058] Step 322: Calculation

[0059]

[0060] In the formula, represents the target power distribution result obtained after the n+1th iteration, represents the target power distribution result obtained after the nth iteration;

[0061] Step 323: Determine whether n=M is ​​satisfied;

[0062] If not satisfied, set n=n+1 and return to step 322;

[0063] If it is satisfied, the target power distribution result obtained in the last iteration is used as the azimuth spectrum obtained by deconvolution.

[0064]

[0065] The beneficial effects of the present invention are:

[0066] The present invention proposes a single vector hydrophone azimuth estimation method based on multiple joint processing deconvolution projection CBF spatial spectrum. The present invention performs multiple joint processing on the sound pressure channel and the vibration velocity channel receiving signals of the single vector hydrophone to obtain the maximum sound energy of the sound pressure and the combined vibration velocity, the sound energy of the sound pressure channel receiving signal and the vibration velocity x channel receiving signal, the sound energy of the sound pressure channel receiving signal and the vibration velocity y channel receiving signal, the sound energy of the vibration velocity x channel and the vibration velocity y channel, and the difference between the sound energy of the vibration velocity x channel and the sound pressure channel. Then, the projected CBF spatial spectrum is calculated according to the calculated sound energy and the difference in sound energy, and the projected CBF spatial spectrum is deconvoluted based on the point spread function to obtain the underwater target azimuth estimation result. Compared with the traditional CBF and MVDR methods, the method of the present invention not only has strong robustness, but also has a narrow beam width and low side lobes, which can further improve the target resolution ability and azimuth estimation accuracy of the single vector hydrophone, further improve the performance of weak target detection, and is more suitable for actual complex underwater acoustic environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Figure 1 is a flow chart of the method of the present invention;

[0068] Figure 2(a) is a comparison of the spatial spectra of the CBF, MVDR, HCBF and DU-HCBF methods when the signal-to-noise ratio (SNR) is -5dB;

[0069] Figure 2(b) is a spatial spectrum comparison diagram of CBF, MVDR, HCBF and DU-HCBF methods when the signal-to-noise ratio (SNR) is -10dB;

[0070] Figure 2(c) is a spatial spectrum comparison diagram of CBF, MVDR, HCBF and DU-HCBF methods when the signal-to-noise ratio (SNR) is -15dB;

[0071] Figure 3 It is the comparison of the azimuth estimation accuracy of CBF, MVDR, HCBF and DU-HCBF methods;

[0072] Figure 4(a) is the anechoic pool data processing result of the CBF method;

[0073] Figure 4(b) is the anechoic pool data processing result of the HCBF method;

[0074] Figure 4(c) is the anechoic pool data processing result of the MVDR method;

[0075] Figure 4(d) shows the anechoic pool data processing result of the DU-HCBF method. DETAILED DESCRIPTION

[0076] Specific implementation method 1: Combination Figure 1 The present embodiment is described as follows. The present embodiment describes a single vector hydrophone azimuth estimation method based on multiple joint processing deconvolution projection CBF spatial spectrum, and the method specifically comprises the following steps:

[0077] Step 1: Taking the center of the vector hydrophone as the reference point, the known signals received by the single vector hydrophone at time t are recorded as p(t), v x (t) and v y (t), where p(t) represents the received signal of the sound pressure channel, v x (t) represents the received signal of the vibration velocity x channel, v y (t) represents the received signal of the vibration velocity y channel;

[0078] Step 2: Perform multiple joint processing on the single vector hydrophone sound pressure channel and vibration velocity channel receiving signals to obtain the maximum sound energy I0 of the sound pressure and combined vibration velocity, the sound energy I0 of the sound pressure channel receiving signal and the vibration velocity x channel receiving signal px , the acoustic energy I of the signal received by the sound pressure channel and the signal received by the vibration velocity y channel py , the acoustic energy I of the vibration velocity x channel and the vibration velocity y channel xy , the difference between the acoustic energy of the vibration velocity x channel and the acoustic energy of the sound pressure channel Ixx ;

[0079] According to I0, I px ,I py ,I xy and I xx Construct a high-order covariance matrix of a single-vector hydrophone, calculate the projected CBF weighted vector, and then calculate the projected CBF spatial spectrum based on the high-order covariance matrix and the projected CBF weighted vector;

[0080] Step 3: construct a point spread function according to the projection beam function, and deconvolve the projected CBF spatial spectrum based on the point spread function, and use the peak position in the azimuth spectrum obtained by deconvolution as the azimuth of the underwater target.

[0081] The present invention fully exploits the acoustic information and inherent relationship carried by the cross-correlation and autocorrelation of each channel of the vector hydrophone. Compared with the conventional beamforming (CBF) and minimum variance distortionless response (MVDR) methods in traditional methods, the method of the present invention not only has strong robustness, but also has a narrow beam width and low sidelobe, which can further improve the target resolution capability and azimuth estimation accuracy of the single vector hydrophone, and is more suitable for use in actual complex underwater acoustic environments.

[0082] The method of the present invention is specifically described for a target situation. It should be noted that the method of the present invention is also applicable to the situation of estimating the azimuth of multiple targets. When there are multiple targets, since the line spectrum components of the radiation noise of different targets are different, this characteristic can be used to perform Fourier transform on the signal received by the vector hydrophone, convert the received signal into the frequency domain, and obtain the frequency corresponding to each target signal based on the Fourier transform result. Then, the method of the present invention is used to process the signals at the corresponding frequencies respectively, so that the azimuth of each target can be accurately estimated.

[0083] Specific implementation method 2: This implementation method is different from the specific implementation method 1 in that the received signal of the sound pressure channel, the received signal of the vibration velocity x channel and the received signal of the vibration velocity y channel are respectively:

[0084]

[0085] Where s(t) is the target signal, θ is the incident direction of the target signal, t is time, t = 1, 2, ..., L, L is the number of snapshots of data collected during the test time, n p (t), n x (t) and n y (t) represents the ambient noise signals received by the sound pressure and vibration velocity x and y channels respectively.

[0086] The other steps and parameters are the same as those in the first embodiment.

[0087] Specific implementation method three: This implementation method is different from specific implementation method one or two in that, in step two, multiple joint processing is performed on the single vector hydrophone sound pressure channel and the vibration velocity channel receiving signal to obtain the maximum sound energy I0 of the sound pressure and the combined vibration velocity, the sound energy I0 of the sound pressure channel receiving signal and the vibration velocity x channel receiving signal px , the acoustic energy I of the signal received by the sound pressure channel and the signal received by the vibration velocity y channel py , the acoustic energy I of the vibration velocity x channel and the vibration velocity y channel xy , the difference between the acoustic energy of the vibration velocity x channel and the acoustic energy of the sound pressure channel I xx ; The specific process is:

[0088] Step 21: The received signal v of the vibration speed x channel is x (t) and the received signal v of the vibration velocity y channel y (t) is projected to the φ direction, and then the combined vibration velocity signal v is obtained based on the projection result. c (t);

[0089] Then based on the sound pressure channel receiving signal and the combined vibration velocity signal v c (t) Maximum sound energy I0 for constructing sound pressure and combined vibration velocity;

[0090] Step 22: Using the characteristics that the received signals of each channel of the single vector hydrophone are completely correlated and the received noise components are uncorrelated, the sound intensity of the received signal of the sound pressure channel and the received signal of the vibration velocity x channel, the sound intensity of the received signal of the sound pressure channel and the received signal of the vibration velocity y channel, the sound energy of the vibration velocity x channel and the sound energy of the vibration velocity y channel of the single vector hydrophone are constructed, which can better suppress isotropic noise;

[0091] The differences between the acoustic energy of the sound pressure channel receiving signal and the vibration velocity x channel receiving signal, the acoustic energy of the sound pressure channel receiving signal and the vibration velocity y channel receiving signal, the acoustic energy of the vibration velocity x channel and the vibration velocity y channel, and the acoustic energy of the vibration velocity x channel and the acoustic energy of the sound pressure channel are:

[0092]

[0093]

[0094]

[0095]

[0096] In the formula, I px It represents the sound energy of the received signal of the sound pressure channel and the received signal of the vibration velocity x channel, I pyIt represents the acoustic energy of the received signal of the sound pressure channel and the received signal of the vibration velocity y channel, I xy Represents the acoustic energy of the vibration velocity x channel and the vibration velocity y channel, I xx It represents the difference between the acoustic energy of the velocity x channel and the acoustic energy of the sound pressure channel, I x Indicates vibration velocity x channel sound energy, I p represents the sound energy of the sound pressure channel, is the noise power of the sound pressure channel, is the noise power of the vibration channel, and

[0097] The other steps and parameters are the same as those in the first or second embodiment.

[0098] In order to simplify the representation, the present invention omits the time parameter t in each acoustic energy calculation result and the subsequent calculation process, and executes the entire method of the present invention once for each data sampling moment, so as to obtain the target orientation corresponding to each data sampling moment.

[0099] Specific implementation method 4: This implementation method is different from any one of the specific implementation methods 1 to 3 in that the combined vibration velocity signal v is obtained based on the projection result. c (t), specifically:

[0100]

[0101] Where φ is the projection angle.

[0102] The other steps and parameters are the same as those in Specific Embodiments 1 to 3.

[0103] Specific implementation mode 5: This implementation mode is different from any one of the specific implementation modes 1 to 4 in that the signal received based on the sound pressure channel and the combined vibration velocity signal v c (t) The maximum sound energy I0 of the combined sound pressure and vibration velocity is constructed as follows:

[0104] Calculate the sound energy I of the combined sound pressure and vibration velocity pvc :

[0105]

[0106] In the formula, (·) H is the conjugate transpose operation; Indicates the power of the target signal;

[0107] According to the above formula, when φ=θ or φ=θ+180°, I pvc Maximum. And traverse the projection angle φ to obtain the maximum sound energy I0 of the combined sound pressure and vibration velocity:

[0108]

[0109] The other steps and parameters are the same as those in Specific Embodiments 1 to 4.

[0110] Specific implementation method 6: This implementation method is different from the specific implementation methods 1 to 5 in that the px ,I py ,I xy and I xx Construct a high-order covariance matrix of a single-vector hydrophone. The specific process is:

[0111] The acoustic energy information I0, I px ,I py ,I xy and I xx Make a vector z:

[0112]

[0113] Where b(θ) is the steering vector matrix of the high-order covariance matrix, b(θ) = [1, cosθ, sinθ, cos(2θ), sin(2θ)] T , the superscript T represents the transpose of the matrix;

[0114] make Among them, u1 is the left singular vector of z, λ1 is the singular value, and β is the linear coefficient;

[0115] Then the high-order covariance matrix R of the vector hydrophone is:

[0116]

[0117] Here, E[·] represents expectation.

[0118] The other steps and parameters are the same as those in Specific Implementation Methods 1 to 5.

[0119] Specific implementation method 7: This implementation method is different from any one of specific implementation methods 1 to 6 in that the projected CBF weighted vector is calculated, and then the projected CBF spatial spectrum is calculated based on the high-order covariance matrix and the projected CBF weighted vector; the specific process is:

[0120] Project the steering vector matrix of the high-order covariance matrix into the signal subspace to obtain the projected CBF weighted vector

[0121]

[0122] In the formula, is the scanning angle, U s represents the signal subspace;

[0123] Then the projected CBF spatial spectrum for:

[0124]

[0125] Here, |·| indicates the value of the calculated determinant.

[0126] The other steps and parameters are the same as those in Specific Embodiments 1 to 6.

[0127] Specific implementation eight: This implementation differs from specific implementations one to seven in that the specific process of step three is as follows:

[0128] Step 31: Re-express formula (12) as:

[0129]

[0130] in, is the distribution function of the target signal power in the azimuth plane, * represents the convolution calculation symbol, is the projection beam function, is the point spread function (PSF) of the projected CBF;

[0131] It can be seen from formula (13) that the projected CBF spatial spectrum can be expressed as the convolution of the distribution function of the target signal power in the azimuth plane and the square of the projection beam function;

[0132] For the sake of Deconvolution is performed, and the square of the projection beam function is The point spread function is used, and the spatial spectrum of the projected CBF method is deconvolved using the Richardson-Lucy (RL) iterative algorithm to obtain the output spatial spectrum of the deconvolved projected CBF algorithm.

[0133] Step 32: Based on point spread function Deconvolute the projected CBF spatial spectrum to obtain a new azimuth spectrum

[0134] The other steps and parameters are the same as those in Specific Embodiments 1 to 7.

[0135] Specific implementation method 9: This implementation method is different from any one of specific implementation methods 1 to 8 in that the distribution function of the target signal power in the angular plane is:

[0136]

[0137] Here, δ(·) represents the unit impulse function.

[0138] The other steps and parameters are the same as those in Specific Embodiments 1 to 8.

[0139] Specific implementation method 10: This implementation method is different from specific implementation methods 1 to 9 in that the specific process of step 32 is as follows:

[0140] Step 321: Initialize the number of iterations n = 1, and initialize the initial value of the target signal power distribution in the azimuth plane to Set the total number of iterations to M;

[0141] Step 322: Calculation

[0142]

[0143] In the formula, represents the target power distribution result obtained after the n+1th iteration, represents the target power distribution result obtained after the nth iteration;

[0144] Step 323: Determine whether n=M is ​​satisfied;

[0145] If not satisfied, set n=n+1 and return to step 322;

[0146] If it is satisfied, the target power distribution result obtained in the last iteration is used as the azimuth spectrum obtained by deconvolution. Right now

[0147] The other steps and parameters are the same as those in Specific Embodiments 1 to 9.

[0148] The method of the present invention can obtain the input point spread function according to the Poisson noise statistical standard. Deconvolution in the fuzzy spectrum obtains a clear target signal power distribution curve to improve the accuracy of target direction estimation.

[0149] Specific simulation verification example:

[0150] The simulation is used to verify the CBF spatial spectrum method based on multiple joint processing deconvolution projection (abbreviated as DU-HCBF) of the present invention, and the results are explained.

[0151] Simulation conditions: Assume that the two velocity channels of the vector hydrophone are along the positive directions of the x and y axes of the spatial rectangular coordinate system. Assume that the incident signal is a narrowband signal and is uncorrelated with each other. Assume that the background noise is isotropic, and the noise power received by the sound pressure channel and the two velocity channels are and SNR is defined as Among them, A sig is the signal amplitude.

[0152] Figures 2(a), 2(b) and 2(c) respectively show the changes in the spatial spectra of the CBF method, the MVDR method, the high-order cumulant CBF method (HCBF for short) and the DU-HCBF method under different signal-to-noise ratios, where the target incident angle is 110° and the signal frequency is 2kHz. It can be seen from Figures 2(a), 2(b) and 2(c) that as the SNR decreases, the background spectra of the CBF and MVDR methods become significantly higher, and the peak position of the spectrum gradually deviates from the true angle of the target; in contrast, HCBF and DU-HCBF always maintain lower side lobes under low SNR conditions, and can estimate the target direction with higher accuracy. Moreover, the DU-HCBF method of the present invention has a narrower main lobe than the HCBF method and has a higher target resolution capability.

[0153] Figure 3 The variation of the azimuth estimation accuracy of the CBF method, MVDR method, HCBF method and DU-HCBF method with SNR is given, where the target incident azimuth is 120°. Figure 3 It can be seen that with the increase of SNR, the root mean square error of the azimuth estimation of the above methods gradually decreases; the target estimation accuracy of the HCBF method and the DU-HCBF method is similar, and the azimuth estimation error of both methods is smaller than that of other methods.

[0154] Specific test verification example:

[0155] In order to further verify the CBF spatial spectrum method based on multiple joint processing deconvolution projection of the present invention, a single vector hydrophone azimuth estimation experiment was carried out in an anechoic pool, and the data was processed, analyzed and explained.

[0156] Figure 4(a), Figure 4(b), Figure 4(c) and Figure 4(d) respectively show the azimuth estimation results of CBF, HCBF, MVDR and DU-HCBF methods. It can be seen from Figure 4(a), Figure 4(b), Figure 4(c) and Figure 4(d) that compared with CBF, HCBF and MVDR methods, DU-HCBF method has narrower beam, lower background spectrum and sharper performance in azimuth spectrum.

[0157] The above calculation examples of the present invention are only used to explain the calculation model and calculation process of the present invention in detail, and are not intended to limit the implementation methods of the present invention. For ordinary technicians in the relevant field, other different forms of changes or modifications can be made based on the above description. It is impossible to list all the implementation methods here. All obvious changes or modifications derived from the technical solution of the present invention are still within the scope of protection of the present invention.

Claims

1. A single vector hydrophone azimuth estimation method based on multiple joint processing deconvolution projection CBF spatial spectrum, characterized in that: The method specifically comprises the following steps: Step 1: The signals received by the single vector hydrophone at time t are recorded as p(t) and v x (t) and v y (t), where p(t) represents the received signal of the sound pressure channel, v x (t) represents the received signal of the vibration velocity x channel, v y (t) represents the received signal of the vibration velocity y channel; Step 2: Perform multiple joint processing on the single vector hydrophone sound pressure channel and vibration velocity channel receiving signals to obtain the maximum sound energy I0 of the sound pressure and combined vibration velocity, the sound energy I0 of the sound pressure channel receiving signal and the vibration velocity x channel receiving signal px , the acoustic energy I of the signal received by the sound pressure channel and the signal received by the vibration velocity y channel py , the acoustic energy I of the vibration velocity x channel and the vibration velocity y channel xy , the difference between the acoustic energy of the vibration velocity x channel and the acoustic energy of the sound pressure channel I xx ; According to I0, I px ,I py ,I xy and I xx Construct a high-order covariance matrix of a single-vector hydrophone, calculate the projected CBF weighted vector, and then calculate the projected CBF spatial spectrum based on the high-order covariance matrix and the projected CBF weighted vector; Step 3: construct a point spread function according to the projection beam function, and deconvolve the projected CBF spatial spectrum based on the point spread function, and use the peak position in the azimuth spectrum obtained by deconvolution as the azimuth of the underwater target.

2. According to claim 1, a single vector hydrophone azimuth estimation method based on multiple joint processing deconvolution projection CBF spatial spectrum is characterized in that: The received signals of the sound pressure channel, the received signals of the vibration velocity x channel, and the received signals of the vibration velocity y channel are respectively: p(t)=s(t)+n p (t) v x (t)=s(t)cosθ+n x (t) (1) v y (t)=s(t)sinθ+n y (t) Among them, s(t) is the target signal, θ is the incident direction of the target signal, t is the time, n p (t), n x (t) and n y (t) represents the ambient noise signals received by the sound pressure and vibration velocity x and y channels respectively.

3. According to claim 2, a single vector hydrophone azimuth estimation method based on multiple joint processing deconvolution projection CBF spatial spectrum is characterized in that: In the step 2, multiple joint processing is performed on the single vector hydrophone sound pressure channel and the vibration velocity channel receiving signal to obtain the maximum sound energy I0 of the sound pressure and the combined vibration velocity, the sound energy I0 of the sound pressure channel receiving signal and the vibration velocity x channel receiving signal px , the acoustic energy I of the signal received by the sound pressure channel and the signal received by the vibration velocity y channel py , the acoustic energy I of the vibration velocity x channel and the vibration velocity y channel xy , the difference between the acoustic energy of the vibration velocity x channel and the acoustic energy of the sound pressure channel I xx ; The specific process is: Step 21: The received signal v of the vibration speed x channel is x (t) and the received signal v of the vibration velocity y channel y (t) is projected to the φ direction, and then the combined vibration velocity signal v is obtained based on the projection result. c (t); Then based on the sound pressure channel receiving signal and the combined vibration velocity signal v c (t) Maximum sound energy I0 for constructing sound pressure and combined vibration velocity; Step 22: The sound energy of the sound pressure channel receiving signal and the vibration velocity x channel receiving signal, the sound energy of the sound pressure channel receiving signal and the vibration velocity y channel receiving signal, the sound energy of the vibration velocity x channel and the vibration velocity y channel, and the difference between the sound energy of the vibration velocity x channel and the sound energy of the sound pressure channel are respectively: In the formula, I px It represents the sound energy of the received signal of the sound pressure channel and the received signal of the vibration velocity x channel, I py It represents the acoustic energy of the received signal of the sound pressure channel and the received signal of the vibration velocity y channel, I xy Represents the acoustic energy of the vibration velocity x channel and the vibration velocity y channel, I xx It represents the difference between the acoustic energy of the velocity x channel and the acoustic energy of the sound pressure channel, I x Indicates vibration velocity x channel sound energy, I p represents the sound energy of the sound pressure channel, is the noise power of the sound pressure channel, is the noise power of the vibration channel.

4. The single vector hydrophone azimuth estimation method based on multiple joint processing deconvolution projection CBF spatial spectrum according to claim 3 is characterized in that: The combined vibration velocity signal v is obtained based on the projection result. c (t), specifically: Where φ is the projection angle.

5. The single vector hydrophone azimuth estimation method based on multiple joint processing deconvolution projection CBF spatial spectrum according to claim 4 is characterized in that: The sound pressure channel receives the signal and combines the vibration velocity signal v c (t) The maximum sound energy I0 of the combined sound pressure and vibration velocity is constructed as follows: Calculate the sound energy I of the combined sound pressure and vibration velocity pvc : In the formula, (·) H is the conjugate transpose operation; Indicates the power of the target signal; And traverse the projection angle φ to obtain the maximum sound energy I0 of the combined sound pressure and vibration velocity:

6. A single vector hydrophone azimuth estimation method based on multiple joint processing deconvolution projection CBF spatial spectrum according to claim 5, characterized in that: According to I0, I px ,I py ,I xy and I xx Construct a high-order covariance matrix of a single-vector hydrophone. The specific process is: The acoustic energy information I0, I px ,I py ,I xy and I xx Make up a vector z: Where b(θ) is the steering vector matrix of the high-order covariance matrix, b(θ) = [1, cosθ, sinθ, cos(2θ), sin(2θ)] T , the superscript T represents the transpose of the matrix; make Among them, u1 is the left singular vector of z, λ1 is the singular value, and β is the linear coefficient; Then the high-order covariance matrix R of the vector hydrophone is: Here, E[·] represents expectation.

7. The single vector hydrophone azimuth estimation method based on multiple joint processing deconvolution projection CBF spatial spectrum according to claim 6 is characterized in that: The projected CBF weighted vector is calculated, and then the projected CBF spatial spectrum is calculated based on the high-order covariance matrix and the projected CBF weighted vector; the specific process is: Project the steering vector matrix of the high-order covariance matrix into the signal subspace to obtain the projected CBF weighted vector In the formula, is the scanning angle, U s represents the signal subspace; Then the projected CBF spatial spectrum for: Here, |·| represents the value of the calculated determinant.

8. The single vector hydrophone azimuth estimation method based on multiple joint processing deconvolution projection CBF spatial spectrum according to claim 7 is characterized in that: The specific process of step three is: Step 31: Re-express formula (12) as: in, is the distribution function of the target signal power in the azimuth plane, * represents the convolution calculation symbol, is the projection beam function, is the point spread function of the projected CBF; Step 32: Based on point spread function Deconvolute the projected CBF spatial spectrum to obtain a new azimuth spectrum 9. The single vector hydrophone azimuth estimation method based on multiple joint processing deconvolution projection CBF spatial spectrum according to claim 8, characterized in that: The distribution function of the target signal power in the angular plane is: Here, δ(·) represents the unit impulse function.

10. The single vector hydrophone azimuth estimation method based on multiple joint processing deconvolution projection CBF spatial spectrum according to claim 9, characterized in that: The specific process of step 32 is as follows: Step 321: Initialize the number of iterations n = 1, and initialize the initial value of the target signal power distribution in the azimuth plane to Set the total number of iterations to M; Step 322: Calculation In the formula, represents the target power distribution result obtained after the n+1th iteration, represents the target power distribution result obtained after the nth iteration; Step 323: Determine whether n=M is ​​satisfied; If not satisfied, set n=n+1 and return to step 322; If it is satisfied, the target power distribution result obtained in the last iteration is used as the azimuth spectrum obtained by deconvolution.

Images