Cooperative underwater acoustic signal time delay difference estimation method

Abstract

This invention relates to a collaborative underwater acoustic signal delay difference estimation method, belonging to the field of underwater acoustic positioning technology. This invention solves the problem that existing methods cannot achieve both low computational complexity and high delay difference estimation accuracy, and also cannot meet real-time requirements. Specifically, the invention involves: obtaining the arrival time of the received signal at each hydrophone and the corresponding copy correlation envelope peak value for each hydrophone based on the received signal received by hydrophones installed on an unmanned platform; calculating the delay difference between adjacent hydrophones based on the copy correlation method; further calculating the delay difference corresponding to the maximum and second-largest peaks of the generalized cross-correlation of adjacent hydrophones, as well as the standard cross-correlation envelope value; calculating the effective amplitude of the received signal at each hydrophone based on the calculated copy correlation envelope peak value and the standard cross-correlation envelope values ​​corresponding to the maximum and second-largest peaks of the generalized cross-correlation; and calculating the final delay difference for each group of adjacent hydrophone channels based on the effective amplitude of the received signal. This invention can be applied to the field of underwater acoustic positioning technology.

Description

Technical Field

[0001] This invention belongs to the field of underwater acoustic positioning technology, specifically relating to a method for estimating the time delay difference of cooperative underwater acoustic signals. Background Technology

[0002] With the development of science and technology, unmanned and intelligent operation has become the development trend of underwater acoustic systems. Autonomous detection systems for unmanned platforms are gradually evolving from single-platform independent operation to multi-platform collaborative operation. The literature (Review of the current development status and key technologies of underwater unmanned vehicle swarms [J]. Journal of Harbin Engineering University, 2020, 41(02):289-297.) introduces two main working modes of positioning between current underwater acoustic multi-platform collaborative systems: one is the parallel working mode, in which each unmanned platform is equipped with the same navigation equipment and obtains the position information of other platforms in the system through underwater acoustic communication; the other is the navigation mode, in which only the central unmanned platform (such as the unmanned surface vessel USV) is equipped with high-precision navigation equipment, and other unmanned platforms (such as unmanned underwater vehicles UUV) use acoustic means to calculate their position relative to the central platform to achieve their own accurate positioning. The parallel system has a simple structure, but each unmanned platform needs to be equipped with high-precision navigation equipment, which increases the manufacturing cost exponentially. The navigation system takes into account both positioning accuracy and manufacturing cost, and can be applied to different sea areas and environments, making it a key research direction for multi-unmanned platform collaborative operation.

[0003] Time delay difference typically refers to the difference in arrival times of target signals received by hydrophones at different spatial locations or between different time periods of a single hydrophone. For a navigation system, accurate positioning of the central platform by each peripheral platform is essential for the effective operation of the entire system. Positioning between multiple platforms requires obtaining the time delay difference parameters of sound waves arriving at different hydrophones; therefore, the accuracy of time delay estimation (TDE) directly affects the positioning accuracy between platforms. With the continuous development and improvement of signal processing methods, time delay estimation technology has rapidly developed in the field of underwater acoustics. The literature (Research on Precise Positioning Technology for Non-cooperative Targets [D]. Harbin Engineering University, 2006.) points out that commonly used time delay estimation methods can be divided into three categories: correlation method, phase spectrum method, and adaptive time delay estimation method.

[0004] The basic principle of correlation methods is to estimate time delay by detecting peak positions. Depending on the source of the reference signal, it can be divided into copy correlation and cross-correlation methods. Copy correlation first obtains the absolute arrival times of the target signal at each hydrophone, then calculates the difference between these absolute times to obtain the time delay. The advantage of copy correlation is its coherent gain, resulting in high stability of time delay estimation. However, when affected by non-stationary interference, multipath propagation, and radial motion between the target and the hydrophone, coherence decreases, and the accuracy of time delay estimation also declines. Cross-correlation methods are further divided into two types: standard cross-correlation and generalized cross-correlation. Generalized cross-correlation is based on standard cross-correlation; the difference lies in the principle that generalized cross-correlation pre-whitens and filters the input signal using a weighting function before correlation processing, which improves the accuracy of time delay estimation compared to standard cross-correlation. The specific implementation involves weighting the cross-power spectra of the two signals in the frequency domain. Different weighting functions correspond to different generalized cross-correlation delay estimation processors. The literature (The generalized correlation method for estimation of time delay[J].IEEE Trans. Acoustic, Speech, and Signal Processing, 1976, 24(8):320-327.) summarizes commonly used generalized cross-correlation processors, such as the Roth processor, the Smooth Coherent Transform (SCOT) processor, the Eckart filter processor, the maximum likelihood (ML) estimator, and the Hannan-Thomson (HT) processor. Compared with the copy correlation method, the cross-correlation method does not have a coherence gain, but its advantage lies in the fact that it can still achieve time delay estimation when the two signals have the same Doppler velocity, the channel multipath structure, or the copy correlation algorithm fails. Both the copy correlation method and the cross-correlation method estimate the time delay based on the statistical prior information of the known input signal and noise. This prior information is often difficult to obtain in practical applications. Overall, this type of algorithm can achieve time delay estimation under conditions of high noise and blurred signal waveforms, but its time delay estimation performance is poor under conditions of correlated noise, non-stationary environments, and time-varying parameters.

[0005] The phase spectrum method, also known as the cross-spectral method, essentially estimates time delay by calculating the slope of the cross-spectral phase spectrum. The standard phase spectrum method, based on the theorem that the correlation function and the power spectral density function of a signal are Fourier transforms of each other, converts the delay information represented by the correlation function in the time domain into the phase information of the signal's power spectrum in the frequency domain. The literature (Generalized Phase Spectral Delay Estimation [J]. Acta Acustica, 1985, (04): 201-215.) builds upon the standard phase spectrum method by using a phase weighting function to weight the phase function in the frequency domain, forming a generalized phase spectrum time delay estimation method, further improving the accuracy of time delay estimation. However, this type of method also requires statistical prior knowledge of the signal and noise, and is only suitable for application scenarios with no noise interference or uncorrelated Gaussian noise interference.

[0006] Adaptive time delay estimation methods can overcome the problem of insufficient prior information on signals and noise, meaning they can track parameters under dynamic or time-varying environments without relying on statistical prior information about the input signal and noise. However, in practical implementation, discretization introduces additional errors, and to ensure a certain level of accuracy, a high order of transverse filters is often required, significantly increasing the computational load. Two improved algorithms address this problem: one is the Direct Time Delay Estimation (EDTE) method and the variable step size method; the other is a direct time delay estimation method based on Lagrange interpolation. While these two methods improve the convergence speed and reduce computational load of adaptive algorithms, their time delay estimation accuracy is not high under low signal-to-noise ratio conditions. Furthermore, implementing adaptive algorithms in real-time systems remains challenging at high sampling frequencies. Overall, adaptive time delay estimation methods relax the requirements for statistical prior knowledge of signals and noise by sacrificing time delay estimation speed. These methods suffer from drawbacks such as high computational load and slow convergence speed, making their application in real-time positioning systems difficult.

[0007] In addition to the traditional methods mentioned above, various modern signal processing algorithms are often incorporated into time delay estimation methods. Examples include: time delay estimation methods based on higher-order statistics, time delay estimation methods based on time-frequency analysis, and time delay estimation methods based on fuzzy functions. Alternatively, traditional algorithms may be combined with new processing techniques or ideas: time delay estimation methods that combine line spectrum phase data time delay estimation using the signal phase matching principle with generalized correlation time delay estimation; time delay estimation by maximizing the cross-covariance function of narrowband signal quadratic sampling; time delay estimation methods based on genetic algorithms; and time delay estimation methods based on neural networks. These methods either improve the accuracy of time delay estimation in multipath and time-varying parameter contexts or accelerate the convergence speed of traditional algorithms. However, due to the large computational load, they cannot yet be applied to real-time systems.

[0008] In conclusion, it is essential to design a time delay estimation algorithm that can balance low computational complexity and high time delay difference estimation accuracy, while also meeting the real-time positioning requirements between platforms in unmanned multi-platform collaborative systems. Summary of the Invention

[0009] The purpose of this invention is to address the problem that existing methods cannot achieve both low computational complexity and high time delay difference estimation accuracy, and cannot meet real-time requirements. Therefore, a cooperative underwater acoustic signal time delay difference estimation method is proposed.

[0010] The technical solution adopted by the present invention to solve the above-mentioned technical problems is as follows:

[0011] A method for estimating the time delay difference of cooperative underwater acoustic signals, the method specifically includes the following steps:

[0012] Step 1: Initialize the non-centralized unmanned platform count l = 1;

[0013] Step 2: A receiving array is installed on the l-th unmanned platform. The receiving array consists of k hydrophones with different spatial positions. Based on the received signals of each hydrophone, the time when the received signal arrives at each hydrophone and the peak value of the copy correlation envelope corresponding to each hydrophone are obtained. The time delay difference between adjacent hydrophones is calculated based on the copy correlation method.

[0014] Next, calculate the time delay difference corresponding to the largest peak of the generalized cross-correlation between adjacent hydrophones, as well as the standard cross-correlation envelope value; and the time delay difference corresponding to the second largest peak of the generalized cross-correlation, as well as the standard cross-correlation envelope value.

[0015] The specific process of step two is as follows:

[0016] Step 2.1: The local signal transmitted by the central unmanned platform is s(t), and the signal received by the i-th hydrophone on the l-th unmanned platform is x. i (t);

[0017] After digitally sampling the signal received by the i-th hydrophone, the digital sampling result is passed sequentially through a bandpass filter and a quadrature filter to obtain x. i The analytic signal x′ corresponding to (t) i (t);

[0018] Similarly, the analytical signal corresponding to the signal received by each hydrophone is obtained;

[0019] The system pre-stores multiple local signals at different Doppler velocities and passes each local signal through an orthogonal filter to obtain the corresponding analytical signal.

[0020] Step 22: Analyze the signals x′ separately. i (t) Perform copy correlation calculation with the analytic signal corresponding to each local signal, obtain the highest copy correlation envelope peak value from the results of each copy correlation calculation, obtain the Doppler velocity of the local signal corresponding to the highest copy correlation envelope peak value, and record the number N of local signal points corresponding to the Doppler velocity. i Record the position x0 of the peak point of the acquired copy-related envelope, the position x1 of the left point adjacent to the peak point, and the position x2 of the right point adjacent to the peak point. At the same time, record the copy-related envelope amplitude y0 corresponding to position x0, the copy-related envelope amplitude y1 corresponding to position x1, and the copy-related envelope amplitude y2 corresponding to position x2.

[0021] Based on positions x0, x1, and x2, and the corresponding copy-correlation envelope amplitudes y0, y1, and y2, the time when the received signal arrives at the i-th hydrophone is obtained. The copy-correlation envelope peak value y corresponding to the i-th hydrophone i_copy ;

[0022] Similarly, the arrival time of the signal at each hydrophone and the corresponding copy correlation envelope peak value of each hydrophone are obtained respectively;

[0023] For adjacent hydrophones i and j, the time when the signal arrives at hydrophone i. The time when the signal arrives at the j-th hydrophone By subtracting the values, we obtain the time delay difference between the i-th hydrophone and the j-th hydrophone, estimated based on the copy correlation method.

[0024] Steps 2 and 3: Based on the analytical signals corresponding to the received signals of each hydrophone, calculate the points corresponding to the maximum and second largest peak values ​​of the generalized cross-correlation between adjacent hydrophones, and obtain the generalized cross-correlation peak value and standard cross-correlation envelope value corresponding to the point of the maximum peak value, and the generalized cross-correlation peak value and standard cross-correlation envelope value corresponding to the point of the second largest peak value.

[0025] Step 24: Calculate the time delay difference corresponding to the maximum and second largest peak values ​​of the generalized cross-correlation between adjacent hydrophones based on the locations corresponding to the maximum and second largest peak values ​​of the generalized cross-correlation between adjacent hydrophones and the corresponding peak values ​​of the generalized cross-correlation between the locations.

[0026] Step 3: Based on the peak value of the copy correlation envelope, the standard cross-correlation envelope value corresponding to the largest peak of the generalized cross-correlation, and the standard cross-correlation envelope value corresponding to the second largest peak of the generalized cross-correlation calculated in Step 2, calculate the effective amplitude of the received signal of each hydrophone.

[0027] Step 4: Based on the time delay difference calculated in Step 2 and the effective amplitude of the received signal calculated in Step 3, calculate the final time delay difference for each group of adjacent hydrophone channels.

[0028] Step 5: Determine if the count 1 has reached its maximum.

[0029] The process ends when the count reaches its maximum.

[0030] If the count has not reached the maximum, let l = l + 1 and return to step one.

[0031] Furthermore, the signal received by the i-th hydrophone is:

[0032] x i (t)=s(t-τ i )+n i(t) (1)

[0033] Where, τ i n represents the time delay of the transmitted local signal arriving at the i-th hydrophone. i It is the noise received by the i-th hydrophone.

[0034] Furthermore, after digitally sampling the signal received by the i-th hydrophone, the digital sampling result is sequentially passed through a bandpass filter and a quadrature filter to obtain x. i The analytic signal corresponding to (t); specifically:

[0035] The signal received by the i-th hydrophone is digitally sampled:

[0036] x i (m)=s(m-τ i )+n i (m) (2) where m represents the discrete time sequence of the sample, m = 1, 2, 3, ..., M, and M represents the length of the sampled signal;

[0037] After passing the digital sampling result through a bandpass filter, the result of the bandpass filtering is the real part of the analytic signal. Then, passing the bandpass filtering result through an orthogonal filter, the result of the orthogonal filtering is the imaginary part of the analytic signal.

[0038] Furthermore, the process of analyzing the signal x′ i (t) performs copy correlation calculations with the analytic signal corresponding to each local signal, specifically:

[0039] Let the analytic signal corresponding to any local signal be denoted as s′(t), x′ i The copy correlation calculation method between s(t) and s′(t) is as follows:

[0040] Reverse the time of s′(t) to obtain the time-reversed signal s′(-t), and then calculate x′. i Convolution of (t) and s′(-t):

[0041]

[0042] Where * represents convolution operation, The expression `ifft{·}` represents performing a Fourier transform on the part within the parentheses, and `x′` represents performing an inverse Fourier transform on the part within the parentheses. i (t)*s′(-t) represents the result of copy-related calculations.

[0043] Furthermore, the time when the received signal arrives at the i-th hydrophone is obtained. The specific process is as follows:

[0044] Based on the parabolic function, the copy-related envelope is interpolated, that is, the parabolic equation system of equation (5) is established:

[0045]

[0046] Where a, b, and c are the coefficients of the parabolic interpolation function to be solved;

[0047] The position of the peak point after interpolation for:

[0048]

[0049] according to Get the time when the signal arrives at the i-th hydrophone. Right now f s It is the sampling rate; and obtains the time. The corresponding copy-related envelope peak value.

[0050] Furthermore, the specific process of steps two and three is as follows:

[0051] Step 231: The analytical signal corresponding to the signal received by the j-th hydrophone is denoted as x′. j (t), after performing time-domain flipping on the signal received by the j-th hydrophone, the analytic signal corresponding to the time-domain flipped signal is denoted as x′. j2 ;

[0052] For analytic signal x′ i (t) Perform an N-point FFT transformation and denote the transformation result as X. i For the analytic signal x′ j (t) Perform an N-point FFT transformation and denote the transformation result as X. j For the analytic signal x′ j2 Perform an N-point FFT transformation and denote the transformation result as X. j * ;

[0053] Where N≥2 a , where 'a' is the minimum number of points in the analytic signal;

[0054] Step 232, X i and X j * After multiplying point by point, we obtain the point-by-point multiplication result A. We then perform an inverse Fourier transform on the point-by-point multiplication result A to obtain the standard cross-correlation result and record the position loc1 where the peak point of the standard cross-correlation envelope appears.

[0055] Step 233: Calculate the analytic signal x′ i The self-power spectrum X of (t) i 2Analyzing the signal x′ j The self-power spectrum X of (t) j 2 The result A from the point-by-point multiplication in steps two and three is... Perform point-by-point multiplication to obtain point-by-point multiplication result B; then perform inverse Fourier transform on point-by-point multiplication result B to obtain generalized cross-correlation result;

[0056] Steps two, three, and four: Define the distance between adjacent hydrophones as d, and the speed of sound in water as c. Then, based on physical reality, the time delay difference range is [-d / c, d / c]. Next, convert the actual time delay difference into the theoretical point range corresponding to the generalized cross-correlation peak value [loc]. _min ,loc _max ];

[0057] Based on the position loc1 of the peak point of the standard cross-correlation envelope and the theoretical number of points τ between two adjacent peaks of the standard cross-correlation, the theoretical point range [loc1-τ, loc1+τ] corresponding to the peak of the standard cross-correlation is obtained.

[0058] The interval [loc] _min ,loc _max The intersection of [loc1-τ,loc1+τ] and [corr] is denoted as [corr] _min ,corr _max ], within the range [corr _min ,corr _max Within [the range], retain the point x corresponding to the maximum peak value of the generalized cross-correlation. _max Point x _max The corresponding generalized cross-correlation peak value r _max and point x _max The corresponding standard cross-correlation envelope value y _max ;

[0059] Steps two, three, and five: [corr] _min ,corr _max The second largest peak of the generalized cross-correlation within the range is denoted as y_sec. Compare y_sec with a·y _max Size;

[0060] If y_sec>a·y _max Then, the point x corresponding to the second largest peak of the generalized cross-correlation is retained. _sec Point x _sec The corresponding second-largest peak value r of generalized cross-correlation _sec and point x _sec The corresponding standard cross-correlation envelope value y _sec ;

[0061] If y_sec≤a·y _maxThen the second largest peak point of the generalized cross-correlation x _sec The corresponding second-largest peak value of generalized cross-correlation and the standard cross-correlation envelope value are set to 0 and retained;

[0062] Steps 2, 3, and 6: Process the received signals of any two adjacent hydrophones using the methods described in steps 2, 3, 1 to 2, 3, and 5 respectively.

[0063] Furthermore, the specific process of step two and four is as follows:

[0064] Step 241: For adjacent hydrophones i and j, denote the position of the left point adjacent to the point of maximum generalized cross-correlation as x. _max-1 Let x be the position of the rightmost point adjacent to the point of maximum peak of generalized cross-correlation. _max+1 , will x _max-1 The corresponding generalized cross-correlation envelope value is denoted as r. _max-1 , will x _max+1 The corresponding generalized cross-correlation envelope value is denoted as r. _max+1 ;

[0065] Step 242: Based on the sampling rate f s Calculate x respectively _max-1 x _max and x _max+1 The corresponding time t _max-1 t _max and t _max+1 ;

[0066] Steps 2, 4, and 3: Design the waveform fitting function for equation (7):

[0067]

[0068] Where, ω r It is the oscillation frequency of the cross-correlation waveform.

[0069] Substituting the calculation result from step 2.4.2 into equation (7), we obtain the system of equations (8):

[0070]

[0071] According to the system of equations (8):

[0072]

[0073]

[0074] r _max+1 sinω r (t _max-1 -t _max ) = r _max sinωr (t _max-1 -t _max+1 )-r _max-1 sinω r (t _max -t _max+1 (11)

[0075] Because of t _max+1 -t _max =t _max -t _max-1 =Δt, divide both sides of equation (11) by sinω. r Δt yields the estimated frequency of the cross-correlation waveform oscillation:

[0076]

[0077] in, This represents the estimation result of the oscillation frequency of the cross-correlation waveform;

[0078] Step 244, when When the waveform fitting function of equation (7) reaches its maximum value, that is, when the waveform fitting function reaches its maximum value, it satisfies:

[0079]

[0080] in, This represents the peak time of the generalized cross-correlation between the i-th hydrophone and the j-th hydrophone, and

[0081] Steps two, four, and five: According to Calculate the time delay difference corresponding to the maximum peak value of the generalized cross-correlation between the i-th hydrophone and the j-th hydrophone. T represents the pulse width after Doppler compensation of the local signal;

[0082] Step 246: Similarly, using the methods from Step 241 to Step 245, calculate the time delay difference corresponding to the second largest peak of the generalized cross-correlation between the i-th hydrophone and the j-th hydrophone.

[0083] Furthermore, the specific process of step three is as follows:

[0084] Taking the i-th hydrophone as an example

[0085] The effective amplitude of the received signal from the i-th hydrophone is calculated as follows:

[0086] V = y / T (14) where V represents the effective amplitude of the signal, y represents the envelope value, and T represents the pulse width of the local signal after Doppler compensation;

[0087] Substituting the copy correlation envelope peak value of the i-th hydrophone into equation (14), we obtain the effective amplitude V corresponding to the copy correlation envelope peak value of the i-th hydrophone. i_copy ;

[0088] Substituting the standard cross-correlation envelope value corresponding to the maximum peak of the generalized cross-correlation of the i-th hydrophone into equation (14), we obtain the effective amplitude V corresponding to the maximum peak of the generalized cross-correlation of the i-th hydrophone. ij_gccmax ;

[0089] Substituting the standard cross-correlation envelope value corresponding to the second largest peak of the generalized cross-correlation of the i-th hydrophone into equation (14), we obtain the effective amplitude V corresponding to the second largest peak of the generalized cross-correlation of the i-th hydrophone. ij_gccsec .

[0090] Furthermore, the method for calculating the pulse width after Doppler compensation of the local signal is as follows:

[0091] N i =T×f s (15).

[0092] Furthermore, the specific process of step four is as follows:

[0093] Taking the i-th hydrophone and the j-th hydrophone as an example, the following steps are performed on the i-th hydrophone and the j-th hydrophone:

[0094] Step 4.1: Determine the time delay difference between the i-th hydrophone and the j-th hydrophone. Is it within range [corr] _min ,corr _max ]Inside;

[0095] like Within the range [corr] _min ,corr _max If the condition is within the specified range, proceed to step four two.

[0096] like Not in range [corr] _min ,corr _max Within ], the final time delay difference between the i-th hydrophone and the j-th hydrophone is

[0097] Step 42: Calculate the time delay difference With delay difference The difference is then used to calculate the delay difference. With delay difference The difference;

[0098] like and The difference is greater than or equal to the threshold b and and If the difference is greater than or equal to the threshold b, then proceed to step four three;

[0099] Otherwise, proceed to step four.

[0100] Step 4.3: Calculate the combined effective signal amplitude V of the i-th hydrophone and the j-th hydrophone. ij_copy :

[0101]

[0102] Among them, V j_copy It is the effective amplitude corresponding to the copy correlation envelope peak value of the j-th hydrophone;

[0103] V ij_gccmax and V ij_gccsec The larger one is denoted as V. ij_gcc Then compare V ij_copy With V ij_gcc Size;

[0104] If V ij_gcc ≤m·V ij_copy Then the final time delay difference between the i-th hydrophone and the j-th hydrophone is

[0105] If V ij_gcc >m·V ij_copy Then the final time delay difference between the i-th hydrophone and the j-th hydrophone is

[0106] The method for determining the value of V is as follows: if V ij_gccmax Greater than or equal to V ij_gccsec ,but If V ij_gccmax Less than V ij_gccsec ,but

[0107] Step 44: Calculation and And compare and Size;

[0108] like Less than or equal to The final time delay difference between the i-th hydrophone and the j-th hydrophone is:

[0109] like Greater than Then proceed to steps four and five;

[0110] Steps four and five: Determine the effective signal amplitude V corresponding to the location of the second largest peak of the generalized cross-correlation. ij_gccsecIs it 0?

[0111] If V ij_gccsec If the value is 0, proceed to step four or six;

[0112] If V ij_gccsec If the value is not 0, then the final time delay difference between the i-th hydrophone and the j-th hydrophone is:

[0113] Step 46: Compare V ij_gccmax With m·V ij_copy Size;

[0114] If V ij_gccmax ≤m·V ij_copy Then the final time delay difference between the i-th hydrophone and the j-th hydrophone is

[0115] If V ij_gccmax >m·V ij_copy Then the final time delay difference between the i-th hydrophone and the j-th hydrophone is

[0116] The beneficial effects of this invention are:

[0117] This invention fully utilizes the advantages of both copy correlation and cross-correlation methods. Under low signal-to-noise ratio conditions, it achieves robust time delay estimation by leveraging the high gain of copy correlation and the whitening background noise characteristics of generalized cross-correlation. Under Doppler and multipath effects, it achieves high-precision time delay difference estimation by using cross-correlation and interpolation methods. Moreover, it does not increase the computational complexity compared to existing correlation methods. Therefore, it can guarantee the estimation accuracy of time delay difference with low computational complexity and meet the real-time requirements of time delay estimation.

[0118] Multiple experiments have demonstrated that the time delay estimation method of this invention has high accuracy and stable performance. Even under Doppler effects and multipath channel conditions, it can still accurately and stably estimate the time delay difference parameter between different array elements. Attached Figure Description

[0119] Figure 1 This is a flowchart of copy-related operations implemented in the frequency domain;

[0120] Figure 2 This is a flowchart of the cross-correlation method implemented in the frequency domain;

[0121] Figure 3 This is a flowchart of the correlation peak processing method using cross-correlation;

[0122] Figure 4 This is a flowchart for calculating the effective amplitude of the received signal;

[0123] Figure 5 This is a flowchart for making a comprehensive judgment on the relevant legal results;

[0124] Figure 6 This is a comparison chart of the time delay difference estimation results using the method of this invention and the traditional correlation method on measured data. Detailed Implementation

[0125] Specific implementation method one: Combining Figure 1 This embodiment describes a method for estimating the time delay difference of cooperative underwater acoustic signals. The method specifically includes the following steps:

[0126] Step 1: Initialize the counter l=1 for non-central unmanned platforms (i.e., unmanned platforms located around the central unmanned platform, i.e., unmanned platforms in the navigation system that are not equipped with high-precision navigation equipment);

[0127] Step 2: A receiving array is installed on the l-th unmanned platform. The receiving array (in this invention, the receiving array can be a linear array, in which the first hydrophone and the second hydrophone are adjacent hydrophones, the second hydrophone and the third hydrophone are adjacent hydrophones, ..., the (k-1)-th hydrophone and the k-th hydrophone are adjacent hydrophones, and the k-th hydrophone and the first hydrophone are adjacent hydrophones, i.e., the number of adjacent hydrophone groups is equal to the number of hydrophones in the receiving array) consists of k hydrophones in different spatial positions. Based on the received signals of each hydrophone, the time when the received signal arrives at each hydrophone and the peak value of the copy correlation envelope corresponding to each hydrophone are obtained, and the time delay difference between adjacent hydrophones obtained based on the copy correlation method is calculated.

[0128] Next, calculate the time delay difference corresponding to the largest peak of the generalized cross-correlation between adjacent hydrophones, as well as the standard cross-correlation envelope value; and the time delay difference corresponding to the second largest peak of the generalized cross-correlation, as well as the standard cross-correlation envelope value.

[0129] The specific process of step two is as follows:

[0130] Step 2: The local signal transmitted by the central unmanned platform is s(t), and the signal received by the i-th hydrophone (i≤k) on the l-th unmanned platform is x. i (t);

[0131] After digitally sampling the signal received by the i-th hydrophone, the digital sampling result is passed sequentially through a bandpass filter and a quadrature filter to obtain x. i The analytic signal x′ corresponding to (t) i (t);

[0132] Similarly, the analytical signal corresponding to the signal received by each hydrophone is obtained;

[0133] The system pre-stores multiple local signals at different Doppler velocities and passes each local signal through an orthogonal filter to obtain the corresponding analytical signal.

[0134] Step 22: Analyze the signals x′ separately. i (t) Perform copy correlation calculation with the analytic signal corresponding to each local signal, obtain the highest copy correlation envelope peak value from the results of each copy correlation calculation, obtain the Doppler velocity of the local signal corresponding to the highest copy correlation envelope peak value, and record the number N of local signal points corresponding to the Doppler velocity. i (It should be noted that the Doppler velocities of hydrophones on the same platform are the same, that is, the number of local signal points corresponding to each hydrophone on the same platform is the same); and record the position x0 of the peak point of the acquired copy correlation envelope, the position x1 of the left point adjacent to the peak point, and the position x2 of the right point adjacent to the peak point. At the same time, record the copy correlation envelope amplitude y0 corresponding to position x0, the copy correlation envelope amplitude y1 corresponding to position x1, and the copy correlation envelope amplitude y2 corresponding to position x2.

[0135] Based on positions x0, x1, and x2, and the corresponding copy-correlation envelope amplitudes y0, y1, and y2, the time when the received signal arrives at the i-th hydrophone is obtained. The copy-correlation envelope peak value y corresponding to the i-th hydrophone i_copy (i.e., time) (corresponding envelope amplitude);

[0136] Similarly, the arrival time of the signal at each hydrophone and the corresponding copy correlation envelope peak value of each hydrophone are obtained respectively;

[0137] For the i-th hydrophone and the j-th hydrophone (i.e., the i-th hydrophone and the j-th hydrophone are adjacent hydrophones), the time when the signal arrives at the i-th hydrophone is determined. The time when the signal arrives at the j-th hydrophone By subtracting the values, we obtain the time delay difference between the i-th hydrophone and the j-th hydrophone, estimated based on the copy correlation method.

[0138] Steps 2 and 3: Based on the analytical signals corresponding to the received signals of each hydrophone, calculate the points corresponding to the maximum and second largest peak values ​​of the generalized cross-correlation between adjacent hydrophones, and obtain the generalized cross-correlation peak value and standard cross-correlation envelope value corresponding to the point of the maximum peak value, and the generalized cross-correlation peak value and standard cross-correlation envelope value corresponding to the point of the second largest peak value.

[0139] Step 24: Based on the locations corresponding to the maximum and second-largest peak values ​​of the generalized cross-correlation between adjacent hydrophones, and the corresponding peak values ​​of the generalized cross-correlation between the locations (i.e., the results of Step 23), calculate the time delay difference corresponding to the maximum and second-largest peak values ​​of the generalized cross-correlation between adjacent hydrophones.

[0140] Step 3: Based on the peak value of the copy correlation envelope, the standard cross-correlation envelope value corresponding to the largest peak of the generalized cross-correlation, and the standard cross-correlation envelope value corresponding to the second largest peak of the generalized cross-correlation calculated in Step 2, calculate the effective amplitude of the received signal of each hydrophone.

[0141] Step 4: Based on the time delay difference calculated in Step 2 and the effective amplitude of the received signal calculated in Step 3, calculate the final time delay difference for each group of adjacent hydrophone channels.

[0142] Step 5: Determine if the count 1 has reached its maximum.

[0143] If the count reaches its maximum, the process ends; the position of the non-central unmanned platform relative to the central platform is obtained based on the calculated final time delay difference.

[0144] If the count has not reached the maximum, let l = l + 1 and return to step one.

[0145] Specific Implementation Method Two: This implementation method differs from Specific Implementation Method One in that the signal received by the i-th hydrophone is:

[0146] x i (t)=s(t-τ i )+n i (t) (1)

[0147] Where, τ i n represents the time delay of the transmitted local signal arriving at the i-th hydrophone. i It is the noise received by the i-th hydrophone, and the noise is zero-mean Gaussian white noise that is uncorrelated with the local signal and spatially uncorrelated.

[0148] The other steps and parameters are the same as in Specific Implementation Method 1.

[0149] Specific Implementation Method Three: This implementation method differs from Specific Implementation Method One or Two in that, after digitally sampling the signal received by the i-th hydrophone, the digital sampling result is sequentially passed through a bandpass filter and a quadrature filter to obtain x. i The analytic signal corresponding to (t); specifically:

[0150] The signal received by the i-th hydrophone is digitally sampled:

[0151] x i (m)=s(m-τ i )+ni (m) (2) where m represents the discrete time sequence of the sample, m = 1, 2, 3, ..., M, and M represents the length of the sampled signal;

[0152] After passing the digital sampling result through a bandpass filter, the result of the bandpass filtering is the real part of the analytic signal. Then, passing the bandpass filtering result through an orthogonal filter, the result of the orthogonal filtering is the imaginary part of the analytic signal.

[0153] Other steps and parameters are the same as in specific implementation method one or two.

[0154] Passing the sampled signal through a bandpass filter preserves the signal within the band of interest while removing out-of-band noise. However, since the system may not accurately capture the original signal at the initial moment, an initial phase bias is introduced into the discrete original signal. This reduces the gain of the copy correlation processing and affects the accuracy of time delay estimation. Therefore, the bandpass-filtered sampled signal needs to be passed through an orthogonal filter to form an analytic signal to compensate for the phase bias caused by sampling.

[0155] Specific implementation method four: Combination Figure 1 This embodiment is described below. The difference between this embodiment and one of the specific embodiments one to three is that the step of analyzing the signal x′... i (t) performs copy correlation calculations with the analytic signal corresponding to each local signal, specifically:

[0156] Let the analytic signal corresponding to any local signal be denoted as s′(t), x′ i The copy correlation calculation method between s(t) and s′(t) is as follows:

[0157] Reverse the time of s′(t) to obtain the time-reversed signal s′(-t), and then calculate x′. i Convolution of (t) and s′(-t):

[0158]

[0159] Where * represents convolution operation, The expression `ifft{·}` represents performing a Fourier transform on the part within the parentheses, and `x′` represents performing an inverse Fourier transform on the part within the parentheses. i (t)*s′(-t) represents the result of the copy correlation calculation, that is, obtaining a copy correlation envelope.

[0160] The other steps and parameters are the same as those in one of the specific implementation methods one to three.

[0161] The calculation method related to time-domain copying is as follows:

[0162]

[0163] For the hydrophone receiving the signal, after passing through the copy correlator, at the signal trailing edge time τ... i A peak value appears at a certain location. By detecting the location of the peak value, the timing of the target signal in that hydrophone channel can be determined.

[0164] As can be seen, time-domain copy correlation requires multiplying the received signal and the local signal point by point, and then accumulating the results of the point-by-point multiplication, which consumes a lot of resources. In contrast, the frequency-domain copy correlation of this invention only requires performing the same number of fast Fourier transforms on the received signal and the time-reversed local signal respectively, then multiplying the transform results, and finally performing an inverse Fourier transform on the results to obtain the copy correlation result. Compared with the time-domain implementation, the frequency-domain implementation of this invention does not require multiple summations, greatly reducing computational complexity and making it more suitable as an algorithm for real-time processing systems.

[0165] Multiple local signals at different Doppler velocities are pre-stored locally. The same hydrophone channel signal and local signals at different Doppler velocities are then used in parallel for copy correlation calculation. The velocity of the local signal with the highest copy correlation envelope peak is the Doppler velocity between the current platforms. The copy correlation results of the local signal and the received signal at this velocity are retained. This avoids the problem that the Doppler phenomenon caused by the radial motion between platforms will lead to a decrease in correlation and affect the accuracy of time delay estimation.

[0166] Specific Implementation Method Five: This implementation method differs from Specific Implementation Methods One to Four in that it obtains the time when the received signal arrives at the i-th hydrophone. The specific process is as follows:

[0167] Based on the parabolic function, the copy-related envelope is interpolated, that is, the parabolic equation system of equation (5) is established:

[0168]

[0169] Where a, b, and c are the coefficients of the parabolic interpolation function to be solved;

[0170] The position of the peak point after interpolation for:

[0171]

[0172] according to Get the time when the signal arrives at the i-th hydrophone. Right now f s It is the sampling rate; and obtains the time. The corresponding copy-related envelope peak value.

[0173] The other steps and parameters are the same as those in one of the specific implementation methods one to four.

[0174] Specific Implementation Method Six: Combination Figure 2 and Figure 3 This embodiment is described below. The difference between this embodiment and specific embodiments one through five is that the specific process of steps two and three is as follows:

[0175] Step 231: The analytical signal corresponding to the signal received by the j-th hydrophone is denoted as x′. j (t), after performing time-domain flipping on the signal received by the j-th hydrophone, the analytic signal corresponding to the time-domain flipped signal is denoted as x′. j2 (i.e., for x′) j (t) Perform bandpass filtering, then time-domain flipping on the bandpass filtering result, and then orthogonal filtering on the time-domain flipped result. The result of orthogonal filtering is x′. j2 );

[0176] For analytic signal x′ i (t) Perform an N-point FFT transformation and denote the transformation result as X. i For the analytic signal x′ j (t) Perform an N-point FFT transformation and denote the transformation result as X. j For the analytic signal x′ j2 Perform an N-point FFT transformation and denote the transformation result as X. j * ;

[0177] Where N≥2 a , where 'a' is the minimum number of points in the analytic signal;

[0178] Step 232, X i and X j * After multiplying point by point, we obtain the point-by-point multiplication result A. We then perform an inverse Fourier transform on the point-by-point multiplication result A to obtain the standard cross-correlation result and record the position loc1 where the peak point of the standard cross-correlation envelope appears.

[0179] Step 233: Calculate the analytic signal x′ i The self-power spectrum X of (t) i 2 Analyzing the signal x′ j The self-power spectrum X of (t) j 2 The result A from the point-by-point multiplication in steps two and three is... Perform point-by-point multiplication to obtain point-by-point multiplication result B; then perform inverse Fourier transform on point-by-point multiplication result B to obtain generalized cross-correlation result;

[0180] Steps two, three, and four: Define the distance between adjacent hydrophones as d, and the speed of sound in water as c. Then, based on physical reality, the time delay difference range is [-d / c, d / c]. Next, convert the actual time delay difference into the theoretical point range corresponding to the generalized cross-correlation peak value [loc]. _min ,loc _max ];

[0181] Based on the position loc1 of the peak point of the standard cross-correlation envelope and the theoretical number of points τ between two adjacent peaks of the standard cross-correlation, the theoretical point range [loc1-τ, loc1+τ] corresponding to the peak of the standard cross-correlation is obtained.

[0182] The interval [loc] _min ,loc _max The intersection of [loc1-τ,loc1+τ] and [corr] is denoted as [corr] _min ,corr _max This intersection represents the range of peak locations in the generalized cross-correlation envelope search, within the range [corr]. _min ,corr _max Within [the range], retain the point x corresponding to the maximum peak value of the generalized cross-correlation. _max Point x _max The corresponding generalized cross-correlation peak value r _max and point x _max The corresponding standard cross-correlation envelope value y _max ;

[0183] Steps two, three, and five: In reality, due to the influence of multipath propagation and signal-to-noise ratio, there may be a relatively large secondary peak within the search range of the generalized cross-correlation envelope peak. At this point, it is impossible to determine whether the true delay difference corresponds to the maximum peak or the secondary peak. Therefore, the range [corr] is... _min ,corr _max The second largest peak of the generalized cross-correlation within the range is denoted as y_sec. Compare y_sec with a·y _max The size of 'a' (the value of 'a' is designed based on the actual delay calculation results);

[0184] If y_sec>a·y _max Then, the point x corresponding to the second largest peak of the generalized cross-correlation is retained. _sec Point x _sec The corresponding second-largest peak value r of generalized cross-correlation _sec and point x _sec The corresponding standard cross-correlation envelope value y _sec ;

[0185] If y_sec≤a·y _max Then the second largest peak point of the generalized cross-correlation x _secThe corresponding second-largest peak value of generalized cross-correlation and the standard cross-correlation envelope value are set to 0 and retained;

[0186] Steps 2, 3, and 6: Process the received signals of any two adjacent hydrophones using the methods described in steps 2, 3, 1 to 2, 3, and 5 respectively.

[0187] The other steps and parameters are the same as those in one of the specific implementation methods one to five.

[0188] Specific Implementation Method Seven: This implementation method differs from Specific Implementation Methods One to Six in that the specific process of step two to four is as follows:

[0189] Step 241: For adjacent hydrophones i and j, denote the position of the left point adjacent to the point of maximum generalized cross-correlation as x. _max-1 Let x be the position of the rightmost point adjacent to the point of maximum peak of generalized cross-correlation. _max+1 , will x _max-1 The corresponding generalized cross-correlation envelope value is denoted as r. _max-1 , will x _max+1 The corresponding generalized cross-correlation envelope value is denoted as r. _max+1 ;

[0190] Step 242: Based on the sampling rate f s Calculate x respectively _max-1 x _max and x _max+1 The corresponding time t _max-1 t _max and t _max+1 ;

[0191] With t _max-1 For example, the calculation formula is t _max-1 =x _max-1 ×f s ;

[0192] Steps 2, 4, and 3: Based on the characteristics of the cross-correlation envelope filling waveform, design the waveform fitting function of equation (7):

[0193]

[0194] Where, ω r It is the oscillation frequency of the cross-correlation waveform.

[0195] Substituting the calculation result from step 2.4.2 into equation (7), we obtain the system of equations (8):

[0196]

[0197] According to the system of equations (8):

[0198]

[0199]

[0200] r _max+1 sinω r (t _max-1 -t _max ) = r _max sinω r (t _max-1 -t _max+1 )-r _max-1 sinω r (t _max -t _max+1 (11)

[0201] Because of t _max+1 -t _max =t _max -t _max-1 =Δt (Δt is the sampling period), divide both sides of equation (11) by sinω. r Δt yields the estimated frequency of the cross-correlation waveform oscillation:

[0202]

[0203] in, This represents the estimation result of the oscillation frequency of the cross-correlation waveform;

[0204] Step 244: Given that the derivative of a continuous function is 0 at its peak, that is, when... When k is a non-negative integer, the waveform fitting function of equation (7) reaches its maximum value, that is, when the waveform fitting function reaches its maximum value, it satisfies:

[0205]

[0206] in, This represents the peak time of the generalized cross-correlation between the i-th hydrophone and the j-th hydrophone, and

[0207] Steps two, four, and five: According to Calculate the time delay difference corresponding to the maximum peak value of the generalized cross-correlation between the i-th hydrophone and the j-th hydrophone. T represents the pulse width after Doppler compensation of the local signal;

[0208] Step 246: Similarly, using the methods from Step 241 to Step 245, calculate the time delay difference corresponding to the second largest peak of the generalized cross-correlation between the i-th hydrophone and the j-th hydrophone.

[0209] The other steps and parameters are the same as those in one of the specific implementation methods one to six.

[0210] The method described in this embodiment is used to process the received signals of k groups of adjacent hydrophones respectively.

[0211] Specific implementation method eight: Combination Figure 4 This embodiment is described below. The difference between this embodiment and any one of specific embodiments one through seven is that the specific process of step three is as follows:

[0212] Taking the i-th hydrophone as an example

[0213] The effective amplitude of the received signal from the i-th hydrophone is calculated as follows:

[0214] V=y / T (14)

[0215] Where V represents the effective amplitude of the signal, y represents the envelope value, and T represents the pulse width of the local signal after Doppler compensation;

[0216] Substituting the copy correlation envelope peak value of the i-th hydrophone into equation (14) as y, we obtain the effective amplitude V corresponding to the copy correlation envelope peak value of the i-th hydrophone. i_copy ;

[0217] Substituting the standard cross-correlation envelope value corresponding to the maximum peak of the generalized cross-correlation of the i-th hydrophone into equation (14) as y, we obtain the effective amplitude V corresponding to the maximum peak of the generalized cross-correlation of the i-th hydrophone. ij_gccmax ;

[0218] Substituting the standard cross-correlation envelope value corresponding to the second largest peak of the generalized cross-correlation of the i-th hydrophone into equation (14) as y, we obtain the effective amplitude V corresponding to the second largest peak of the generalized cross-correlation of the i-th hydrophone. ij_gccsec .

[0219] The other steps and parameters are the same as those in any of the specific implementation methods one to seven.

[0220] Specific Implementation Method Nine: This implementation method differs from Specific Implementation Methods One through Eight in that the calculation method for the pulse width after Doppler compensation of the local signal is as follows:

[0221] N i =T×f s (15).

[0222] The other steps and parameters are the same as those in one of the specific implementation methods one to eight.

[0223] Specific Implementation Method Ten: Combining Figure 5 This embodiment is described below. The difference between this embodiment and any one of specific embodiments one through nine is that the specific process of step four is as follows:

[0224] Taking the i-th hydrophone and the j-th hydrophone as an example, the following steps are performed on the i-th hydrophone and the j-th hydrophone:

[0225] Step 4.1: Determine the time delay difference between the i-th hydrophone and the j-th hydrophone. Is it within range [corr] _min ,corr _max ]Inside;

[0226] like Within the range [corr] _min ,corr _max If the condition is within the specified range, proceed to step four two.

[0227] like Not in range [corr] _min ,corr _max Within ], the final time delay difference between the i-th hydrophone and the j-th hydrophone is

[0228] Step 42: Calculate the time delay difference With delay difference The difference is then used to calculate the delay difference. With delay difference The difference; b is related to the actual transmitted signal and channel conditions, and needs to be adjusted appropriately in practical applications;

[0229] like and The difference is greater than or equal to the threshold b and and If the difference is greater than or equal to the threshold b, then proceed to step four three;

[0230] Otherwise, proceed to step four.

[0231] Step 4.3: Calculate the combined effective signal amplitude V of the i-th hydrophone and the j-th hydrophone. ij_copy :

[0232]

[0233] Among them, V j_copy It is the effective amplitude corresponding to the copy correlation envelope peak value of the j-th hydrophone;

[0234] V ij_gccmax and V ij_gccsec The larger one is denoted as V. ij_gcc Then compare V ij_copy With V ij_gcc Size;

[0235] If V ij_gcc ≤m·V ij_copyWhere m is an empirical value that can be adjusted based on the actual estimated effect, and can be taken as 0.6 to 0.9. Then the final time delay difference between the i-th hydrophone and the j-th hydrophone is:

[0236] If V ij_gcc >m·V ij_copy Then the final time delay difference between the i-th hydrophone and the j-th hydrophone is

[0237] The method for determining the value of V is as follows: if V ij_gccmax Greater than or equal to V ij_gccsec ,but If V ij_gccmax Less than V ij_gccsec ,but

[0238] Step 44: Calculation and And compare and Size;

[0239] like Less than or equal to The final time delay difference between the i-th hydrophone and the j-th hydrophone is:

[0240] like Greater than Then proceed to steps four and five;

[0241] Steps four and five: Determine the effective signal amplitude V corresponding to the location of the second largest peak of the generalized cross-correlation. ij_gccsec Is it 0?

[0242] If V ij_gccsec If the value is 0, proceed to step four or six;

[0243] If V ij_gccsec If the value is not 0, then the final time delay difference between the i-th hydrophone and the j-th hydrophone is:

[0244] Step 46: Compare V ij_gccmax With m·V ij_copy Size;

[0245] If V ij_gccmax ≤m·V ij_copy Then the final time delay difference between the i-th hydrophone and the j-th hydrophone is

[0246] If V ij_gccmax >m·V ij_copy Then the final time delay difference between the i-th hydrophone and the j-th hydrophone is

[0247] The other steps and parameters are the same as those in any of the specific implementation methods one to nine.

[0248] The time delay difference between two adjacent hydrophone channels in each group is obtained using the method of this embodiment.

[0249] Simulation section

[0250] The time delay difference estimates obtained by the copy correlation method, cross-correlation method, and the method of this invention are compared, and the comparison results are as follows: Figure 6 As shown. Since the motion between the target and the receiving array is continuous within a short time, the time delay difference estimation result should also be continuous within the corresponding time range. Therefore, the performance comparison between the method of this invention and the traditional method can be transformed into a comparison of the continuity of the time delay difference estimation results of different methods. The coefficient of variation is the ratio of the standard deviation to the mean, and it is used to compare the degree of dispersion of different datasets or different levels of the same dataset. The larger the value of the coefficient of variation, the greater the relative dispersion of the data, that is, the worse the data continuity. Using this parameter as the performance index of the estimator, the... Figure 6 The dispersion of the time delay difference estimation results corresponding to the three methods is analyzed, as shown in Table 1:

[0251] Table 1. Performance Comparison of Three Delay Difference Estimation Methods

[0252] Delay difference estimation method Discrete coefficients Copy related 0.3645 Intercorrelation 0.3692 Method of the present invention 0.3162

[0253] The data in Table 1 shows that, for the same data, the estimation result obtained by the method of this invention has the smallest coefficient of variation. Therefore, the method of this invention can achieve robust, real-time, and high-precision time delay difference estimation.

[0254] The above examples of the present invention are merely illustrative of the computational model and process of the present invention, and are not intended to limit the implementation of the present invention. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is impossible to exhaustively list all possible implementations here. Any obvious variations or modifications derived from the technical solutions of the present invention are still within the scope of protection of the present invention.

Claims

1. A method for estimating the time delay difference of cooperative underwater acoustic signals, characterized in that, The method specifically includes the following steps: Step 1: Initialize the non-centralized unmanned platform count ; Step Two, in the... Each unmanned platform is equipped with a receiving array, which consists of receivers located at different spatial positions. The system consists of several hydrophones. Based on the received signals from each hydrophone, the system obtains the time when the received signal arrives at each hydrophone and the peak value of the copy correlation envelope corresponding to each hydrophone. The system also calculates the time delay difference between adjacent hydrophones based on the copy correlation method. Next, calculate the time delay difference corresponding to the largest peak of the generalized cross-correlation between adjacent hydrophones, as well as the standard cross-correlation envelope value; and the time delay difference corresponding to the second largest peak of the generalized cross-correlation, as well as the standard cross-correlation envelope value. The specific process of step two is as follows: Step 21: The local signal transmitted by the central unmanned platform is... , No. The first unmanned platform The signal received by the hydrophone is ; For the first After the signal received by each hydrophone is digitally sampled, the digital sampling result is passed sequentially through a bandpass filter and a quadrature filter to obtain... Corresponding analytical signal ; Similarly, the analytical signal corresponding to the signal received by each hydrophone is obtained; The system pre-stores multiple local signals at different Doppler velocities and passes each local signal through an orthogonal filter to obtain the corresponding analytical signal. Step 22: Separate the analytical signals Copy correlation calculations are performed on the analytic signal corresponding to each local signal. The highest copy correlation envelope peak value is obtained from the results of each copy correlation calculation. The Doppler velocity of the local signal corresponding to the highest copy correlation envelope peak value is obtained, and the number of points of the local signal corresponding to the Doppler velocity is recorded. And record the location of the peak points of the acquired copy-related envelope. The position of the left point adjacent to the peak point and the rightmost point adjacent to the peak point Record location at the same time Corresponding copy-related envelope amplitude ,Location Corresponding copy-related envelope amplitude and location Corresponding copy-related envelope amplitude ; According to location ,Location and location and location ,Location and location Corresponding copy-related envelope amplitude , and The received signal arrives at the first The moment of the hydrophone and the The copy-related envelope peak value corresponding to each hydrophone ; Similarly, the arrival time of the signal at each hydrophone and the corresponding copy correlation envelope peak value of each hydrophone are obtained respectively; For adjacent th The hydrophone and the first The first hydrophone will send the signal to the first The moment of the hydrophone With the signal arriving at the The moment of the hydrophone By subtracting, we obtain the first value estimated based on the copy correlation method. The hydrophone and the first The time delay difference of each hydrophone ; Steps 2 and 3: Based on the analytical signals corresponding to the received signals of each hydrophone, calculate the points corresponding to the maximum and second largest peak values ​​of the generalized cross-correlation between adjacent hydrophones, and obtain the generalized cross-correlation peak value and standard cross-correlation envelope value corresponding to the point of the maximum peak value, and the generalized cross-correlation peak value and standard cross-correlation envelope value corresponding to the point of the second largest peak value. Step 24: Calculate the time delay difference corresponding to the maximum and second largest peak values ​​of the generalized cross-correlation between adjacent hydrophones based on the locations corresponding to the maximum and second largest peak values ​​of the generalized cross-correlation between adjacent hydrophones and the corresponding peak values ​​of the generalized cross-correlation between the locations. Step 3: Based on the peak value of the copy correlation envelope, the standard cross-correlation envelope value corresponding to the largest peak of the generalized cross-correlation, and the standard cross-correlation envelope value corresponding to the second largest peak of the generalized cross-correlation calculated in Step 2, calculate the effective amplitude of the received signal of each hydrophone. Step 4: Based on the time delay difference calculated in Step 2 and the effective amplitude of the received signal calculated in Step 3, calculate the final time delay difference for each group of adjacent hydrophone channels. Step 5: Determine the count Has it reached its maximum? The process ends when the count reaches its maximum. If the count has not reached the maximum, then let Return to step one.

2. The method for estimating the time delay difference of cooperative underwater acoustic signals according to claim 1, characterized in that, The first The signal received by each hydrophone is: (1) in, The local signal representing the transmission arrived at the The delay of a hydrophone It is the first The noise received by the hydrophone.

3. The method for estimating the time delay difference of cooperative underwater acoustic signals according to claim 2, characterized in that, The first After the signal received by each hydrophone is digitally sampled, the digital sampling result is passed sequentially through a bandpass filter and a quadrature filter to obtain... The corresponding analytical signal; specifically: For the first The signals received by each hydrophone are digitally sampled: (2) in, Represents a discrete time series of samples. , This represents the length of the sampled signal; After passing the digital sampling result through a bandpass filter, the result of the bandpass filtering is the real part of the analytic signal. Then, passing the bandpass filtering result through an orthogonal filter, the result of the orthogonal filtering is the imaginary part of the analytic signal.

4. The method for estimating the time delay difference of cooperative underwater acoustic signals according to claim 3, characterized in that, The analytical signal The copy correlation calculation is performed with the analytic signal corresponding to each local signal, specifically as follows: Let the analytic signal corresponding to any local signal be denoted as . , and The copy-related calculation method is as follows: right Perform time reversal to obtain the time-reversed signal. , then calculate and Convolution: (3) in, Represents convolution operation. This indicates that a Fourier transform is performed on the part within the parentheses. This indicates that the part within the parentheses will undergo an inverse Fourier transform. This represents the result of copy-related calculations.

5. The method for estimating the time delay difference of cooperative underwater acoustic signals according to claim 4, characterized in that, The received signal reaches the first The moment of the hydrophone The specific process is as follows: Based on the parabolic function, the copy-related envelope is interpolated, that is, the parabolic equation system of equation (5) is established: (5) in, , and These are all coefficients of the parabolic interpolation function to be solved; The position of the peak point after interpolation for: (6) according to The signal arrived at the first The moment of the hydrophone ,Right now , It is the sampling rate; and obtains the time. The corresponding copy-related envelope peak value.

6. The method for estimating the time delay difference of cooperative underwater acoustic signals according to claim 5, characterized in that, The specific process of steps two and three is as follows: Step Two, Three One, The The analytical signal corresponding to the signal received by each hydrophone is denoted as . , for the After time-domain flipping of the signals received by the hydrophones, the analytic signal corresponding to the time-domain flipped signal is denoted as... ; For analytical signals conduct Point FFT transformation, the transformation result is denoted as ; for analytical signals conduct Point FFT transformation, the transformation result is denoted as ; for analytical signals conduct Point FFT transformation, the transformation result is denoted as ; in, , It is the minimum number of points in the analytic signal; Step 232, and After multiplying point by point, we obtain the point-by-point multiplication result A. We then perform an inverse Fourier transform on result A to obtain the standard cross-correlation result and record the positions of the peak points in the standard cross-correlation envelope. ; Steps 2 and 3: Calculate the analytical signal self-power spectrum Analyzing signals self-power spectrum The result A from the point-by-point multiplication in steps two and three is... Perform point-by-point multiplication to obtain point-by-point multiplication result B; then perform inverse Fourier transform on point-by-point multiplication result B to obtain generalized cross-correlation result; Steps two, three, and four: Define the spacing between adjacent hydrophones as... The speed of sound in water is Therefore, the time delay difference range based on physical reality is: Then, the actual time delay difference is converted into the theoretical point range corresponding to the peak value of the generalized cross-correlation. ; Based on the location of the peak point of the standard cross-correlation envelope Theoretical points between two adjacent peaks of the standard cross-correlation The theoretical point range corresponding to the peak value of standard cross-correlation was obtained. ; interval and The intersection is denoted as Within the scope Within, retain the points corresponding to the maximum peak value of the generalized cross-correlation. Points Corresponding generalized cross-correlation peak and locations Corresponding standard cross-correlation envelope value ; Steps two, three, and five: [Scope] The second largest peak of the generalized cross-correlation within is denoted as ,Compare and Size; like Then the points corresponding to the second largest peak of the generalized cross-correlation are retained. Points The corresponding second-largest peak value of generalized cross-correlation and locations Corresponding standard cross-correlation envelope value ; like Then the second largest peak point of the generalized cross-correlation will be... The corresponding second-largest peak value of generalized cross-correlation and the standard cross-correlation envelope value are set to 0 and retained; Steps 2, 3, and 6: Process the received signals of any two adjacent hydrophones using the methods described in steps 2, 3, 1 to 2, 3, and 5 respectively.

7. The method for estimating the time delay difference of cooperative underwater acoustic signals according to claim 6, characterized in that, The specific process of step two or four is as follows: Step 241: For adjacent... The hydrophone and the first For each hydrophone, the position of the leftmost point adjacent to the maximum peak of the generalized cross-correlation is denoted as... The position of the rightmost point adjacent to the point of maximum generalized cross-correlation is denoted as... ,Will The corresponding generalized cross-correlation envelope value is denoted as ,Will The corresponding generalized cross-correlation envelope value is denoted as ; Step 242: Based on the sampling rate Calculate separately , and corresponding time , and ; Steps 2, 4, and 3: Design the waveform fitting function for equation (7): (7) in, It is the oscillation frequency of the cross-correlation waveform. ; Substituting the calculation result from step 2.4.2 into equation (7), we obtain the system of equations (8): (8) According to the system of equations (8): (9) (10) (11) because Divide both sides of equation (11) by The estimation results of the cross-correlation waveform oscillation frequency are obtained: (12) in, This represents the estimation result of the oscillation frequency of the cross-correlation waveform; Step 244, when When the waveform fitting function of equation (7) reaches its maximum value, that is, when the waveform fitting function reaches its maximum value, it satisfies: (13) in, Representing the The hydrophone and the first The peak time of the generalized cross-correlation of each hydrophone, and ; Steps two, four, and five: According to Calculate the first The hydrophone and the first The time delay difference corresponding to the maximum peak value of the generalized cross-correlation of each hydrophone , , The pulse width after Doppler compensation of the local signal; Step 246: Similarly, use the methods from Step 241 to Step 245 to calculate the... The hydrophone and the first The time delay difference corresponding to the second largest peak of the generalized cross-correlation of each hydrophone .

8. The method for estimating the time delay difference of cooperative underwater acoustic signals according to claim 7, characterized in that, The specific process of step three is as follows: No. The method for calculating the effective amplitude of the received signal of a hydrophone is as follows: (14) in, Represents the effective amplitude of the signal. Represents the envelope value. The pulse width after Doppler compensation of the local signal; The first Substituting the copy-related envelope peak value of the first hydrophone into equation (14), we obtain the first... The effective amplitude corresponding to the copy-related envelope peak value of a hydrophone ; The first Substituting the standard cross-correlation envelope value corresponding to the maximum peak of the generalized cross-correlation of the first hydrophone into equation (14), we obtain the first... The effective amplitude corresponding to the maximum peak of the generalized cross-correlation of each hydrophone ; The first Substituting the standard cross-correlation envelope values ​​corresponding to the second largest peaks of the generalized cross-correlation of each hydrophone into equation (14), we obtain the first... The effective amplitude corresponding to the second largest peak of the generalized cross-correlation of each hydrophone .

9. The method for estimating the time delay difference of cooperative underwater acoustic signals according to claim 8, characterized in that, The method for calculating the pulse width after Doppler compensation of the local signal is as follows: (15)。 10. The method for estimating the time delay difference of cooperative underwater acoustic signals according to claim 9, characterized in that, The specific process of step four is as follows: For the adjacent first The hydrophone and the first Each hydrophone performs the following steps: Step 4.1, Determine the first... The hydrophone and the first The time delay difference of each hydrophone Is it within range? Inside; like In scope If inside, proceed to step four two; like Not in range Inside, then the first The hydrophone and the first The final time delay difference of each hydrophone is ; Step 42: Calculate the time delay difference With delay difference The difference is then used to calculate the delay difference. With delay difference The difference; like and The difference is greater than or equal to the threshold. and and The difference is greater than or equal to the threshold. Then proceed to step four three; Otherwise, proceed to step four. Step 4.3, Calculate the first... The hydrophone and the first The combined effective signal amplitude of the hydrophone : in, It is the first The effective amplitude corresponding to the copy-related envelope peak value of each hydrophone; Will and The larger one is denoted as Compare again and Size; like Then the first The hydrophone and the first The final time delay difference of each hydrophone is ; like Then the first The hydrophone and the first The final time delay difference of each hydrophone is ; The method for determining the value is: if Greater than or equal to ,but ,like Less than ,but ; Step 44: Calculation and and compare and Size; like Less than or equal to Then the first The hydrophone and the first The final time delay difference of each hydrophone is ; like Greater than Then proceed to steps four and five; Steps four and five: Determine the effective signal amplitude corresponding to the location of the second largest peak of the generalized cross-correlation. Is it 0? like If the value is 0, proceed to step four or six; like If it is not 0, then the first The hydrophone and the first The final time delay difference of each hydrophone is ; Step 46: Comparison and Size; like Then the first The hydrophone and the first The final time delay difference of each hydrophone is ; like Then the first The hydrophone and the first The final time delay difference of each hydrophone is .

Images