Time reversal processing method

Abstract

The invention provides a time reversal processing method for detecting an underwater moving target, which comprises the following steps of: transmitting a predetermined underwater sound signal s (t) at a transmitting platform for the first time, and receiving a first echo signal x (t) at a receiving platform; self-adaptively estimating H (.) according to the underwater acoustic signal s (t) transmitted for the first time and the echo signal x (t) for the first time, wherein H (.) represents underwater acoustic environment parameter information between an underwater acoustic signal transmitting platform and the underwater moving target, Doppler information of target movement and current channel response information; acquiring a data segment containing channel broadening and target impact response broadening from the first echo signal x (t), and performing time reversal processing to obtain an underwater acoustic signal simulating second active emission; calculating and simulating the underwater acoustic signal actively transmitted for the second time by using H (.), and carrying out convolution to obtain a corresponding second time echo signal; and processing the second echo signal to detect the underwater moving target.

Description

Technical Field

[0001] The present application relates to underwater acoustic signal processing technology, and in particular to a time reversal processing method and device suitable for detecting moving targets in water and decoupling from the underwater acoustic environment. Background Art

[0002] Given the weak domestic research foundation for the detection and identification of small mobile targets, it is very difficult to carry out research using existing technical routes and methods. In particular, the identification of small targets is even more difficult. It is urgent to conduct principle research on new technical routes and methods to lay the foundation for subsequent model development.

[0003] In 1995, Fink proposed a method of time reversal operator decomposition based on the time reversal mirror iterative algorithm for the case of two scatterers. This method performs eigenvalue decomposition on the time reversal operator and achieves selective focusing by processing the eigenvectors. Both theoretical and experimental results have proved that this method can achieve the ability to locate two scatterers in inhomogeneous media.

[0004] In 1996, Professor W.A. Kuperman of the United States, with the support of the Italian Saclant Anti-Submarine Center (NATO), conducted a sea trial in the Mediterranean Sea, proving the validity of the time reversal mirror principle and the physical phenomenon of its focusing under marine environmental conditions. In 1999, Kuperman et al. studied the iterative time reversal technology of underwater acoustics. For a single scatterer, iterative time reversal processing can enhance the spatial focusing performance of the reversal mirror. When there are multiple reflectors in the ocean, using an iterative algorithm, the time-reversed signal will be focused at the target corresponding to the maximum eigenvalue of the time reversal operator. In subsequent research, W.A. Kuperman's team conducted a large number of applications of underwater acoustic time reversal technology in reverberation suppression and communications.

[0005] In 2003, Professors David M. Pierson and David E. Aspnes of the United States, funded by the U.S. Navy, conducted research on the application of the time reversal method to buried mine detection and made significant progress. However, due to the sensitivity of the subject, no relevant content has been found in publicly published literature.

[0006] Zhejiang University conducted an experimental study on active time reversal detection under laboratory conditions, verifying the feasibility of iterative time reversal technology in seabed target detection, the feasibility of active time reversal methods for target detection, and the feasibility of time reversal processing for buried target detection. Harbin Engineering University conducted research on the use of active iterative time reversal technology to detect sunken and buried targets, and applied active time reversal technology to target intensity measurement, verifying the feasibility and noise resistance. Northwestern Polytechnical University conducted in-depth research on the suppression of reverberation and selective focusing of active time reversal technology. South China University of Technology verified the feasibility of iterative time reversal and time reversal operator decomposition methods for active sonar detection of seabed targets through laboratory simulation research.

[0007] It can be seen from the active time reversal detection research reports at home and abroad that at present, active time reversal technology is mainly used in the detection of sunken and buried targets, and the main methods are concentrated on Fink's iterative time reversal and time reversal operator decomposition methods and their improved methods. The common assumptions of these methods are:

[0008] ① The channel response is time-invariant. This is equivalent to requiring the target and the active sonar platform to be relatively static. The channel response is related to the relative position of the target and the active sonar platform. Even if the underwater acoustic environment parameters are time-invariant, if the target is moving, the changes in relative position and Doppler phenomenon make it difficult to guarantee the time-invariance requirement of the channel response.

[0009] ② The target is a point scattering source. If the scattering process of the target to the sound wave is regarded as a linear system, this assumption is equivalent to regarding the impulse response of the target as a delta function, without considering the impact of the target impulse response on the echo broadening phenomenon. In the traditional active time reversal process based on secondary emission, the impulse response of the target can bring about time reversal focusing gain.

[0010] ③The underwater acoustic environment parameters must be known.

[0011] These three assumptions limit the application of active time reversal technology in detecting moving targets in water. Further research is needed to establish an active time reversal detection model suitable for moving targets in water and decoupled from the underwater acoustic environment. Summary of the invention

[0012] The underwater acoustic time reversal phenomenon can achieve the refocusing of the scattered sound waves. Its physical laws have been deeply understood and mastered, but it requires that the detection platform and the target remain in no relative motion, which affects the application of underwater acoustic time reversal technology in underwater acoustic small target detection. The purpose of this application is to study the change law of the time reversal phenomenon with the relative motion of the target when the detection platform and the small target are in relative motion, and to explore the method of achieving sound wave refocusing through spatial position and Doppler compensation, so as to achieve the detection of small underwater targets. And remove the dependence of time reversal technology on environmental parameters, broaden the application scope of underwater acoustic time reversal technology, improve the tolerance of time reversal technology to environmental conditions, and use time reversal technology to increase the active detection distance of small targets.

[0013] According to a first aspect of the present application, a first time reversal processing method for detecting a moving target in water is provided, comprising: transmitting a predetermined hydroacoustic signal s(t) for the first time on a transmitting platform, and receiving a first echo signal x(t) on a receiving platform; adaptively estimating H(·) based on the hydroacoustic signal s(t) transmitted for the first time and the first echo signal x(t), wherein H(·) represents the hydroacoustic environment parameter information between the hydroacoustic signal transmitting platform and the moving target in water, the Doppler information of the target motion, and the current channel response information; obtaining a data segment including channel broadening and target impulse response broadening from the first echo signal x(t) for time reversal processing to obtain a hydroacoustic signal simulating a second active transmission; calculating and convolving the hydroacoustic signal simulating the second active transmission with H(·) to obtain a corresponding second echo signal; processing the second echo signal to detect the moving target in water; wherein H(·) is estimated by the following steps: the first echo signal x(t) is expressed as:

[0014]

[0015] Where n is the discrete time point number, s(n) represents the first transmitted underwater acoustic signal in discrete form, m is the time delay discrete point number of H, M is the maximum time delay point number of H(·), and v(n) is the sum of noise and reverberation contributions. The first step is to roughly select possible H points, including using s(n) and the received signal as a correlation function, assuming that the signal is uncorrelated with noise or reverberation, and performing normalization processing, and we can get r(m) = H. e (m), m = 0, 1, 2..., M-1, where the subscript e represents normalization processing; the second step is to select H e ; Including selecting K possible H e Combine two by two to build: ss k1,k2 (n) = H e (m k1 )s(nm k1 )+H e (m k2 )s(nmk2 ), where k1 and k2 are two different labels of the rough selected points, m k1 and m k2 are the corresponding delay points respectively; ss intercepted from the echo signal x(n) k1,k2 (n) The corresponding data segment x k1,k2 (n), taking the correlation coefficient between the two as the objective function, optimize the search H e (m k1 ) and H e (m k2 ), to maximize the objective function, that is: C max (k1,k2)=coef[x k1,k2 (n),ss k1,k2 (n)]; C based on all pairwise combinations of records max (k1, k2), preferably select L corresponding delay points with larger correlation coefficients as H e (·) selected points; the third step is to use all L delay points to further globally optimize L H e (·) value; including optimization C max (k1,k2,…,kl)=coef[x k1,k2,…,kl (n),ss k1,k2,…,kl (n)] to obtain the corresponding H at L delay points e The optimized estimate of .

[0016] According to the second time reversal processing method for detecting moving targets in water according to the first aspect of the present application, the third step uses all L time delay points to further globally optimize L H e The value also includes L H e (·)Optimize value grouping.

[0017] According to the third time reversal processing method for detecting moving targets in water according to the first aspect of the present application, the first echo signal is resampled to obtain multiple resampling channels corresponding to multiple Doppler channels, and the above steps one to three are performed for each Doppler channel.

[0018] According to the fourth time reversal processing method for detecting a moving target in water according to the first aspect of the present application, the first echo signal h1(t) and h3(t) are the channel impulse responses of the acoustic wave from the transmitting platform to the target and from the target to the receiving platform, respectively. t represents time, and h2(t) represents the impulse response of the target scattering process. Let have

[0019] According to the fifth time reversal processing method for detecting moving targets in water according to the first aspect of the present application, a spatiotemporal two-dimensional filter is set on the receiving platform to process the echo signal.

[0020] According to the second aspect of the present application, a first time reversal processing method for detecting a moving target in water comprises: transmitting a predetermined hydroacoustic signal ss(t) for the first time on a transmitting platform, and receiving a first echo signal x(t) on a receiving platform; wherein ss(t)=s(t)+s1(tT-T1), s(t) represents the first half of the transmitted signal, s(-t)=s(t), s1(tT-T1) represents the second half of the transmitted signal, the second half of the transmitted signal is the time reversal of the first half, T is an interval set to reduce the superposition of the two parts in the echo, and T is much smaller than the time interval between sending the transmitted signal and receiving the echo signal, and T1 is the width of s(t) in time;

[0021] H(-t) is adaptively estimated based on the first transmitted underwater acoustic signal ss(t) and the first echo signal x(t), where v(t) is the sum of interference and noise, where the second half x1(t) of the first echo signal x(t) corresponding to the second half s1 of the transmitted signal is taken and time-inversely processed to obtain Where v1 corresponds to the sum of interference and noise in the second half of the echo signal; correlating s(t) with x1(n,-t), we can get: y1(τ)=H(-τ)+vs

[0022] (τ), where vs(τ) =<s(t),v1(-t+τ)> , <·,·> represents correlation, n is the discrete time point number, y1(τ) is the estimation of H(-τ) using the echo signal, and τ is the pulse width; take the first half of the echo signal x(n,t), denoted as y(t), and convolve it with y1(τ) to obtain: as a simulated echo signal; and processing the simulated echo signal to detect the underwater moving target.

[0023] According to the second aspect of the present application, the computing device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, one of the methods according to the first aspect or the second aspect of the present application is implemented.

[0024] The positive technical effect of this application is that it uses time reversal technology in a breakthrough way in the field of underwater acoustic small target detection, realizes the focusing of underwater acoustic detection signals, successfully achieves higher processing gain under low signal-to-noise ratio conditions, and expands small target detection technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present application. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0026] Figure 1 It is a flow chart of detecting underwater moving targets by using time reversal technology according to an embodiment of the present application.

[0027] Figure 2 Schematic diagram of signal waveform of adaptive time inversion process.

[0028] Figure 3A It is the correlation coefficient of the signal when no time inversion processing is performed in the underwater acoustic experiment.

[0029] Figure 3B It is the correlation coefficient after time reversal processing in underwater acoustic experiment. DETAILED DESCRIPTION

[0030] The following is a clear and complete description of the technical solutions in the embodiments of the present application in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.

[0031] Based on the analysis of the traditional active time-reversal process of underwater acoustics based on secondary emission, this application provides a solution to the problem, then establishes a corresponding active time-reversal detection model that is independent of prior underwater acoustic environment information and underwater moving target parameters, develops corresponding algorithms, and realizes active time-reversal detection of underwater moving targets.

[0032] The traditional active time-reversal process of the combined transmitter and receiver includes two underwater acoustic signal transmission processes. The refocusing achieved by the time-reversed underwater acoustic signal enhances the detection effect of the target.

[0033] The first time the preset signal s(t) is transmitted, the received echo signal is:

[0034]

[0035] In formula (4-1), h1(t) and h3(t) are the channel impulse response of the sound wave from the transmitting platform to the target, and the channel impulse response from the target to the receiving platform, respectively, and t represents time. For the moving transmitting platform, the moving receiving platform and the moving target, the influence of the Doppler phenomenon is also included; considering the target scattering process as a linear system, h2(t) is its impulse response, which also includes the influence of the Doppler factor.

[0036] When the launch platform, receiving platform and target are all stationary, and the underwater acoustic environment parameters are constant, h1(t) = h3(t). However, when one of the launch platform, receiving platform and target is in motion, considering the influence of the Doppler phenomenon, h1(t) = h3(t) is difficult to hold.

[0037] In addition to the influence of the Doppler factor, the physical meaning of h2(t) can be understood as follows: when a sound wave is incident on a target with a certain spatial scale, the backscattered sound waves generated at different positions of the target will have a certain time difference, amplitude and phase distribution when they arrive at the receiving position. If the target is regarded as a point scattering source, h2(t) has the form of a delta function.

[0038] make We can get:

[0039]

[0040] The echo signal x(t) waveform is time-reversed (to obtain x(-t)) and transmitted for the second time. The echo signal received for the second transmitted signal x(-t) is:

[0041]

[0042] The peak value of the echo signal is:

[0043] y max (t)=s(-t)·∫H 2 (τ)dτ max (4-4)

[0044] Where τ is the pulse width.

[0045] The speed of underwater targets is relatively slow. It is assumed that no observable Doppler changes, channel impulse responses, and changes in target impulse responses will occur on the time scale of the transmitted signal and its echo broadening.

[0046] According to the embodiment of the present application, H(·) is adaptively estimated in real time based on the first transmitted signal and the received echo signal. H(·) contains underwater acoustic environment parameter information, Doppler information of target motion and current channel response information. It is equivalent to a black box, which can describe this information numerically without separating the effects of various factors.

[0047] Then, the data segment containing the channel widening and the target impulse response widening is taken out from the echo signal, and after time reversal, the computer simulates the second active signal transmission process without actually transmitting the signal. During the second active signal transmission process, the transmitted signal is the signal obtained by time reversing the echo signal of the first transmission signal. Then, using the characteristic that the underwater acoustic environment does not change rapidly, the computer calculates the echo signal obtained by the second active transmission signal through the previously estimated H(·). Since the second echo signal can achieve refocusing, the target can be effectively detected from the second echo signal. Furthermore, a good performance space-time two-dimensional filter is designed at the receiving end. On the one hand, it has the ability to correct the estimation error of H(·), and on the other hand, it has a strong reverberation suppression ability. Finally, an active time reversal sonar with good moving target detection performance is realized.

[0048] Optionally, during the second active signal transmission process simulated by the computer, the estimated H(·) is substituted into formula (4-2) to calculate the second echo signal generated by the second active signal transmission.

[0049] Based on the obtained estimate of H(·), the computer simulation of the second launch process has been separated from the influence of the underwater acoustic environment and the target motion parameters. In this sense, the active time-reversal process has been decoupled from the underwater acoustic environment and the target motion.

[0050] Figure 1 A flow chart of detecting underwater moving targets by time reversal technology according to an embodiment of the application is shown.

[0051] The transmitting platform transmits the underwater acoustic signal, and the receiving platform receives the echo signal. The echo signal contains the signal reflected by the target, as well as the interference and noise of the underwater acoustic environment.

[0052] The transmitting platform transmits the hydroacoustic signal s(t) for the first time and receives the first echo signal x(t) accordingly. The transmitted hydroacoustic signal is significantly different from the ambient noise, so the echo signal corresponding to the transmitted hydroacoustic signal can be easily identified from the echo signal.

[0053] Using the first transmitted underwater acoustic signal s(t) and the first echo signal x(t), H(·) is estimated as a comprehensive model that describes the influence of the underwater acoustic environment, moving targets and the current channel.

[0054] The first echo signal x(t) is time-reversed to obtain x(-t), and x(-t) is convolved with H(·) to simulate the second echo signal corresponding to the time-reversed signal x(-t). The underwater acoustic signal is refocused by time reversal, so that the second echo signal has a higher signal processing gain. Therefore, the detection capability of the moving target in the water from the second echo signal is stronger. A variety of existing methods for detecting targets from underwater acoustic echo signals can be used, which do not belong to the technical improvement of the present application.

[0055] According to an embodiment of the present application, two methods for adaptively estimating H(·) in real time from a first echo signal of a first transmitted signal are also provided.

[0056] Estimation method 1: Estimation using the correlation of the first echo signal

[0057] The resampling of the echo signal waveform or the transmission signal waveform can describe the Doppler phenomenon of the moving target. By setting multiple different resampling channels, it is equivalent to setting multiple different Doppler channels, and Doppler scanning can be performed. The following discussion is carried out on a given Doppler channel, and the Doppler channel number is omitted.

[0058] The echo signal can be expressed as:

[0059]

[0060] Where n is the discrete time point number, m is the time delay discrete point number of H(·), M is the maximum time delay point number of H(·), and v(n) is the sum of noise and reverberation contributions.

[0061] The first step is to roughly select possible H(·) points. Use s(n) and the received signal as correlation functions, assume that the signal is unrelated to noise or reverberation, and perform normalization, and we can get

[0062] r(m)=H e (m) m=0,1,2...,M-1 (4-6)

[0063] The subscript e represents normalization processing.

[0064] In fact, there may be some correlation between reverberation and signal, and sometimes the correlation is even good.

[0065] Step 2: Select H e (·). Select the K possible H e (·) Combine two by two to construct:

[0066] ss k1,k2 (n) = H e (m k1 )s(nmk1 )+H e (m k2 )s(nm k2 ) (4-7)

[0067] Among them, k1 and k2 are two different labels of the rough selected points, m k1 and m k2 are the corresponding delay points. ss intercepted from the echo signal x(n) k1,k2 (n) The corresponding data segment x k1,k2 (n), taking the correlation coefficient between the two as the objective function, optimize the search H e (m k1 ) and H e (m k2 ), so as to maximize the objective function, that is:

[0068] C max (k1,k2)=coef[x k1,k2 (n),ss k1,k2 (n)] (4-8)

[0069] Only when H e (m k1 ) and H e (m k2 ) is consistent with the corresponding value contained in x(n), the correlation coefficient between the two can be maximized.

[0070] C based on all pairwise combinations of records max (k1, k2), preferably select L corresponding delay points with larger correlation coefficients as H e (·)'s selected points.

[0071] The third step is to use all L delay points to further globally optimize L H e (·) value.

[0072] C max (k1,k2,…,kl)=coef[x k1,k2,…,kl (n),ss k1,k2,…,kl (n)](4-9)

[0073] After optimizing equation (4-9), the corresponding H at L delay points can be obtained. e The optimized estimate of (·).

[0074] Optimize the estimate H eThe process of (·) is based on the time correlation of the echo signal. If the reverberation also has a long time correlation, the estimation result should include the influence of the reverberation. For example, if the time correlation of the seabed reverberation is long, it may have a greater impact on the estimation result. When performing global optimization according to formula (4-9), if x(n) does not contain a selected H e (·), the global optimization process will make the H e The value of (·) is close to 0.

[0075] If the overall optimization calculation is too large, L H e (·)Optimize value grouping.

[0076] Estimation method 2: Estimation through signal waveform design and algorithm

[0077] The waveform of the first transmitted signal is designed so that the two parts of the transmitted signal waveform are time-inverted waveforms. Using this feature of the transmitted signal, an algorithm is used to adaptively estimate H(·), and finally active time reversal focusing (ATR) is achieved using the signal itself.

[0078] like Figure 2 , the transmitted signal waveform is:

[0079] ss(t)=s(t)+s1(tT-T1) (4-10)

[0080] Among them, s(-t) = s(t), s(t) represents the first half of the transmitted signal, s1(tT-T1) represents the second half of the transmitted signal, and the second half of the transmitted signal is the time reversal of the first half. T is the interval set to minimize the superposition of the two parts of the echo, and T is much smaller than the time interval between the transmitted signal and the echo signal.

[0081] The first step is to estimate H(-t), assuming the echo signal is:

[0082]

[0083] Among them, ss(t) is given by (4-10), and v(t) is the sum of interference (reverberation) and noise. Take the second half of the echo signal (corresponding to the second half of the transmitted signal. Due to the existence of interval T, the second half of the echo signal x1(t) can be easily identified) and perform time inversion processing:

[0084]

[0085] Where v1 corresponds to the sum of interference (reverberation) and noise in the second half of the echo signal.

[0086] By correlating s(t) with x1(n,-t), we can obtain:

[0087] y1(τ)=H(-τ)+vs(τ) (4-13)

[0088] Among them, <·, ·> represents correlation, vs(τ)=<s(t),v1(-t+τ)> .

[0089] From formula (4-13), we can see that y1(τ) is the estimate of H(-τ) using the echo signal.

[0090] The second step is time-reverse focusing.

[0091] Take the first half of x(n,t), recorded as y(t), and convolve it with y1(τ):

[0092]

[0093] Take τ = 0, then the signal waveform after time inversion focusing is:

[0094] ztr(t)=s(t)·∫H 2 (τ)dτ+vvs(n,n′,t) (4-15)

[0095] in, Let τ = 0, and n and n′ are the numbers of discrete time points.

[0096] The above process uses specially designed signal waveforms and the echo data of the detection pulse itself to adaptively achieve time reversal focusing (ATR), and focuses all the scattered signal components in the entire detection system without requiring any system parameters. The process of achieving time reversal focusing is actually the process of estimating H(·), and then superimposing the sum of interference (reverberation) and noise to complete the echo estimation.

[0097] Experimental data verification

[0098] At a distance of 40km and a sound source level of 180db, using the main correlation peak of the Rake channel as the reference point, collecting the energy of each multipath signal basically does not bring any gain in signal-to-noise ratio. The main reason is that the multipath correlation peak is too low compared to the main peak. In order to verify the time-reversal processing gain under low signal-to-noise ratio conditions, we find a lower multipath peak as a reference point, and then perform time-reversal processing together with other multipath correlation peaks, as follows Figure 3A and 3B shown. Figure 3A is the correlation coefficient of the signal without time inversion processing. Figure 3B is the correlation coefficient after time-reversal processing.

[0099] contrast Figure 3A and Figure 3BIt can be seen that through time reversal processing, the correlation coefficient increased from 0.05439 to 0.09244, nearly doubling, and according to power estimation, it increased by about 4-5db. It also shows that time reversal processing has a higher processing gain at a lower signal-to-noise ratio. The experimental data obtained a higher processing gain after time reversal processing. In underwater acoustic signal processing, a higher processing gain can achieve better processing results. In the field of small target detection, due to the natural problems of small targets such as small size and weak echo signals, time reversal technology improves the detection capability of small targets by increasing the signal processing gain.

[0100] This application has made a breakthrough in the use of time reversal technology in the field of underwater acoustic small target detection, achieved the focusing of underwater acoustic detection signals, successfully achieved higher processing gain under low signal-to-noise ratio conditions, and expanded small target detection technology.

[0101] The information processing device running the implementation of the present application includes, for example, a memory, one or more processors, one or more presentation components, I / O components, and a power supply directly or indirectly coupled to the bus. The bus may represent one or more buses (such as an address bus, a data bus, or a combination thereof).

[0102] Although the preferred embodiments of the present application have been described, those skilled in the art may make additional changes and modifications to these embodiments once they are aware of the basic creative concepts. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present application. Obviously, those skilled in the art may make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A time reversal processing method for detecting a moving target in water, comprising: The transmitting platform transmits a predetermined hydroacoustic signal s(t) for the first time, and the receiving platform receives a first echo signal x(t); Adaptively estimate H(·) according to the first transmitted hydroacoustic signal s(t) and the first echo signal x(t), wherein H(·) represents the hydroacoustic environment parameter information between the hydroacoustic signal transmitting platform and the underwater moving target, the Doppler information of the target motion and the current channel response information; obtain the data segment containing the channel broadening and the target impulse response broadening from the first echo signal x(t) and perform time inversion processing to obtain the hydroacoustic signal simulating the second active transmission; Use H(·) to calculate and simulate the second active emission of the underwater acoustic signal to convolve and obtain the corresponding second echo signal; Processing the second echo signal to detect the underwater moving target; where H(·) is estimated by the following steps: The first echo signal x(t) is expressed as: Where n is the discrete time point number, s(n) represents the first emitted underwater acoustic signal in discrete form, m is the time delay discrete point number of H, M is the maximum time delay point number of H(·), and v(n) is the sum of noise and reverberation contributions; The first step is to roughly select possible H points, including using s(n) and the received signal as a correlation function, assuming that the signal is uncorrelated with noise or reverberation, and performing normalization, and we can get r(m)=H e (m), m = 0, 1, 2..., M-1, where the subscript e represents normalization; Step 2: Select H e ; Including selecting K possible H e Combine two by two to construct: ss k1,k2 (n)=H e (m k1 )s(n-m k1 )+H e (m k2 )s(n-m k2 ), Among them, k1 and k2 are two different labels of the rough selected points, m k1 and m k2 are the corresponding delay points respectively; ss intercepted from the echo signal x(n) k1，k2 (n) The corresponding data segment x k1，k2 (n), taking the correlation coefficient between the two as the objective function, optimize the search H e (m k1 ) and H e (m k2 ), to maximize the objective function, that is: C max (k1, k2) = coef[x k1，k2 (n), ss k1，k2 (n)]; C based on all pairwise combinations of records max (k1, k2), preferably select L corresponding delay points with larger correlation coefficients as H e (·)'s selected points; The third step is to use all L delay points to further globally optimize L H e (·) value; including optimization C max (k1, k2, ..., k1) = coef[x k1，k2，...，kl (n), ss k1，k2，...，kl (n)] to obtain the corresponding H at L delay points e The optimized estimate of .

2. The method according to claim 1, wherein The third step uses all L delay points to further globally optimize L H e The value also includes L H e (·)Optimize value grouping.

3. The method according to claim 1 or 2, wherein The first echo signal resampling process is performed to obtain multiple resampling channels corresponding to multiple Doppler channels, and the above steps 1 to 3 are performed on each Doppler channel.

4. The method according to claim 3, wherein First echo signal h1(t) and h3(t) are the channel impulse responses of the acoustic wave from the transmitting platform to the target and from the target to the receiving platform, respectively. t represents time, and h2(t) represents the impulse response of the target scattering process. Let have 5. The method according to claim 4, wherein A spatiotemporal two-dimensional filter is set on the receiving platform to process the echo signal.

6. A time reversal processing method for detecting a moving target in water, comprising: The transmitting platform transmits a predetermined hydroacoustic signal ss(t) for the first time, and the receiving platform receives a first echo signal x(t); in ss(t)=s(t)+s1(tT-T1), s(t) represents the first half of the transmitted signal, s(-t)=s(t), s1(tT-T1) represents the second half of the transmitted signal, the second half of the transmitted signal is the time reversal of the first half, T is the interval set to reduce the superposition of the two parts in the echo, and T is much smaller than the time interval between sending the transmitted signal and receiving the echo signal, T1 is the width of s(t) in time; H(-t) is adaptively estimated based on the first transmitted underwater acoustic signal ss(t) and the first echo signal x(t), where v(t) is the sum of interference and noise, where the second half x1(t) of the first echo signal x(t) corresponding to the second half s1 of the transmitted signal is taken and time-inversely processed to obtain Where v1 corresponds to the sum of interference and noise in the second half of the echo signal; correlating s(t) with x1(n, -t), we can get: y1(τ)=H(-τ)+vs(τ), where vs(τ)=<s(t),v1(-t+τ)> , <·,·> represents correlation, n is the discrete time point number, y1(τ) is the estimation of H(-τ) using the echo signal, and τ is the pulse width; Take the first half of the echo signal x(n, t), record it as y(t), and convolve it with y1(τ) to get: As a simulated echo signal; The simulated echo signal is processed to detect the underwater moving target.

7. A computing device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method according to any one of claims 1 to 6 when executing the program.

Images