A method for seabed depth measurement based on modified peak search

Abstract

The present invention provides a seabed depth measurement method based on modified peak search, which belongs to the technical field of multi-beam sounding. Aiming at the problem that the pseudo-peak interference in the MUSIC pseudo-spectrum obtained at the edge of the seabed using the small-aperture MUSIC algorithm in the multi-beam sounding system is strong, resulting in serious errors in the subsequent WMT algorithm's detection of the seabed edge terrain, an improvement is made to the seabed depth measurement method. The method assumes that the seabed terrain is continuous, selects the moment when the echo signal-to-noise ratio is the largest, and uses the MUSIC algorithm to perform DOA estimation to obtain the signal moment and angle reference value. Then, taking the reference value as the starting point, the TOA expected value is used to weight and search for peaks on the echo signal in the direction of decreasing and increasing angles, respectively, to achieve more accurate sounding. The present invention avoids the problem of large errors in edge beam depth estimation under small aperture in the multi-beam sounding system, and improves the accuracy of seabed depth measurement.

Description

Technical Field

[0001] The present invention belongs to the technical field of multi-beam sounding, and in particular relates to a seabed depth measurement method based on modified peak search. Background Art

[0002] Resolution is a key performance metric for multibeam echosounder systems, determining their ability to detect complex seafloor topography and small underwater targets. In early traditional multibeam echosounder systems, Direction of Arrival (DOA) estimation was often based on conventional beamforming (CBF) algorithms. In this scenario, the system's angular resolution was limited by the array's effective aperture. Using CBF algorithms with small-aperture arrays results in lower angular resolution. To address this issue, a series of super-resolution DOA estimation algorithms have been proposed, with the Multiple Signal Classification (MUSIC) algorithm being one of the most representative. It performs matrix decomposition on the covariance matrix of the receiving array to obtain signal and noise subspaces, then constructs a spatial spectrum function based on the orthogonality of the two subspaces to achieve high-resolution DOA estimation. In multibeam echosounder systems, the time-beam domain data obtained using the MUSIC algorithm loses phase information, necessitating the subsequent use of the classic amplitude-based echosounder algorithm, the Weighted Mean Time (WMT) algorithm, to obtain echosounder values. The WMT algorithm, with a fixed DOA, uses the amplitude information of range-beam domain data to estimate the time of arrival (TOA) of seafloor echoes arriving from all directions. This algorithm offers the advantages of simplicity and efficiency. However, due to the low signal-to-noise ratio of the edge beams of seafloor echoes, the performance of the MUSIC algorithm in multibeam bathymetry systems is degraded. The resulting MUSIC pseudo-spectrum contains strong pseudo-peak interference, leading to significant errors in the subsequent WMT algorithm's depth estimation results at the edge of the seafloor. Therefore, maintaining high angular resolution while maintaining accurate TOA estimation for edge beams while maintaining a small array size has become an important research topic. Summary of the Invention

[0003] In view of the deficiencies in the prior art, the present invention provides a seabed depth measurement method based on modified peak search, which has a simple implementation method, obvious improvement effect, and is easy to be applied in actual engineering.

[0004] To achieve the above object, the present invention adopts the following technical solutions:

[0005] A method for measuring seabed depth based on modified peak search, characterized by comprising the following steps:

[0006] S1: Use a multi-beam system to measure the echo signal of the seabed, select the moment with the largest signal-to-noise ratio in the echo signal as the time reference value, use the MUSIC algorithm to estimate the DOA of the time reference value, and obtain the angle reference value;

[0007] S2: Establish a range-beam domain matrix with angles as rows and distances as columns, where the elements represent the echo signal strength. The TOA value of the beam corresponding to the angle reference value is equal to the time reference value. The TOA value of the beam corresponding to the angle reference value is converted to the sound speed to obtain the corresponding seabed depth.

[0008] S3: Search the first direction based on the angle reference value:

[0009] S3.1: If the angle of the beam corresponding to the angle reference value is equal to the beam steering angle of the first beam in the echo signal, the search ends. Otherwise, starting from the angle reference value, the search proceeds toward the row with decreasing angles in the range-beam domain matrix. The expected TOA value of the next beam is calculated based on the TOA value of the beam corresponding to the angle reference value.

[0010] S3.2: Calculate the beam footprint of the next beam, centered around the expected TOA value of the first beam, and weight the amplitude sequence of the next beam within the beam footprint.

[0011] S3.3: Perform a peak search on the weighted amplitude sequence, obtain the TOA value of the next beam based on the peak value, and convert the TOA value of the next beam into the speed of sound to obtain the corresponding seabed depth;

[0012] S3.4: If the angle of the next beam is equal to the beam steering angle of the first beam, stop searching in the direction of the decreasing angle. Otherwise, use the angle of the next beam as the new angle reference value and continue searching in S3.1 until the search is complete.

[0013] S4: Search in the second direction according to the angle reference value:

[0014] S4.1: If the angle of the beam corresponding to the angle reference value is equal to the beam steering angle of the last beam in the echo signal, the search ends. Otherwise, starting from the angle reference value, the search proceeds toward the rows with increasing angles in the range-beam domain matrix. The expected TOA value of the next beam is calculated based on the TOA value of the beam corresponding to the angle reference value.

[0015] S4.2: Calculate the beam footprint of the next beam, centered around the expected TOA value of the first beam, and weight the amplitude sequence of the next beam within the beam footprint.

[0016] S4.3: Perform a peak search on the weighted amplitude sequence, obtain the TOA value of the next beam based on the peak value, and convert the TOA value of the next beam into the speed of sound to obtain the corresponding seabed depth;

[0017] S4.4: If the angle of the next beam is equal to the beam steering angle of the last beam, stop searching in the direction of the increasing angle. Otherwise, use the angle of the next beam as the new angle reference value and continue searching in S4.1 until the search is complete.

[0018] S5: Arrange the seabed depths corresponding to the first beam to the last beam to obtain the seabed depth within the beam coverage angle.

[0019] To optimize the above technical solutions, specific measures taken also include:

[0020] Furthermore, in S1, the DOA of the time reference value is estimated using the MUSIC algorithm to obtain the angle reference value, as follows:

[0021] The echo signal is received by a uniform linear array composed of M array elements with an array element spacing of d. The echo signals with N wavelengths of λ are received at different beam steering angles [θ1, θ2, ..., θ N ] is incident on a uniform linear array; the following spatial spectrum function is established using the MUSIC algorithm:

[0022]

[0023] Where, P MUSIC (t ref ,θ) represents the spatial spectrum function, t ref represents the reference value at the time, θ=θ1,θ2,...,θ N Indicates the angle, a(t ref , θ)=[1, exp(-j2πsinθd / λ),…, exp[-j2πsinθ(M-1)d / λ]] T The sampling time is t ref , j represents an imaginary number, U N The noise subspace obtained by matrix decomposition of the MUSIC algorithm, the superscript T represents transpose, and the superscript H represents conjugate;

[0024] Search the peak position of the spatial spectrum function to obtain the DOA estimate θ ref , and used as the angle reference value.

[0025] Furthermore, in S2, the distance-beam domain matrix is ​​established as follows:

[0026] After AD sampling, orthogonal demodulation and filtering extraction of the echo signal, the multi-channel raw data matrix S1 is obtained as follows:

[0027]

[0028] Where s ML Represents the multi-channel raw data corresponding to the Lth sampling sequence of the Mth array element, where M is the number of array elements and L is the number of sampling points;

[0029] Perform pulse compression operation on S1 to obtain the multi-channel pulse compression data matrix S2:

[0030]

[0031] Where s ML ′ represents the multi-channel pulse compression data corresponding to the Lth sampling sequence of the Mth array element;

[0032] S2 is divided into blocks along the sampling sequence direction, and the number of snapshots is K. Then the channel data matrix S before and after the sampling time l (K-1) / 2 is 2l for:

[0033]

[0034] Construct the MUSIC pseudo spectrum, normalize the spectrum function and multiply it by the data block S 2l The average amplitude of , we get:

[0035]

[0036] Where, P MUSIC (l, θ) is the MUSIC spectrum function at sampling time l, P WM_MUSIC (l,θ) is P MUSIC The result of weighted average of (l, θ), s′ m(k+l) is the matrix S 2l Channel data in;

[0037] Traverse the sampling moments to obtain the time-beam domain data matrix P WM_MUSIC , the time-beam domain data matrix is ​​a two-dimensional matrix about the sampling time l and the angle θ;

[0038] The time-beam domain data matrix is ​​converted into the distance-beam domain data matrix through sound speed conversion.

[0039] Furthermore, in S3.1 and S4.1, the TOA expected value of the next beam is calculated based on the TOA value of the beam corresponding to the angle reference value. The formula is as follows:

[0040]

[0041] Where, t exp represents the expected TOA value of the next beam, t ref Indicates the reference value at the moment, θ c is the beam steering angle corresponding to the desired beam.

[0042] Furthermore, in S3.2 and S4.2, the beam footprint of the next beam is calculated with the TOA expected value of the next beam as the center. The formula is as follows:

[0043]

[0044] Where, T foot represents the beam footprint width, H is the seabed depth, θ is the angle, Θ is the beam main lobe width, c is the speed of sound, and τ is the signal time width after pulse compression.

[0045] Furthermore, in S3.2 and S4.2, the amplitude sequence of the next beam is weighted within the beam footprint width. Specifically, within the beam footprint width, the signal closer to the expected TOA value is given a larger weight, and the signal farther away from the expected TOA value is given a smaller weight.

[0046] The beneficial effects of the present invention are as follows: It uses the MUSIC algorithm to estimate the time of maximum signal-to-noise ratio of the echo signal to obtain an angle reference value; it utilizes the correlation between the TOAs of adjacent beams on a flat seabed to perform a corrected peak search from the angle reference value in the directions of decreasing and increasing angles; and it uses the expected TOA value to weight the echo signal amplitude and perform a maximum search, ultimately obtaining an amplitude-corrected peak value and achieving more accurate TOA estimation. Essentially, the present invention improves upon the classic amplitude method (WMT), resolving the WMT algorithm's inaccurate TOA estimation for edge beams. It not only ensures higher angular resolution but also achieves more accurate seabed depth detection results compared to the WMT algorithm, yielding a more realistic seabed topography. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 The flowchart of a seabed depth measurement method based on modified peak search is shown in FIG.

[0048] Figure 2 This is the signal processing flow chart of the multi-beam echo sounder system using the MUSIC algorithm.

[0049] Figure 3a This is the range-beam domain distribution diagram based on the large-aperture CBF algorithm under flat terrain.

[0050] Figure 3b This is the range-beam domain distribution diagram based on the small-aperture MUSIC algorithm on flat terrain.

[0051] Figure 3c This is the range-beam domain distribution map finally obtained by using this method on flat terrain.

[0052] Figure 4a This is the range-beam domain distribution diagram based on the large-aperture CBF algorithm under relatively inclined terrain.

[0053] Figure 4b This is the range-beam domain distribution diagram based on the small-aperture MUSIC algorithm under relatively inclined terrain.

[0054] Figure 4c This is the range-beam domain distribution map finally obtained by using this method under relatively inclined terrain.

[0055] Figure 5 is a comparison of the depth measurement values ​​of the existing algorithm and this method under different terrains.

[0056] Figure 6 is a comparison of the depth measurement errors of existing algorithms and this method under different terrains. DETAILED DESCRIPTION

[0057] The present invention will now be described in further detail with reference to the accompanying drawings.

[0058] like Figure 1 As shown, this embodiment provides a seabed depth measurement method based on modified peak search, which specifically includes the following steps.

[0059] Step 1: Measure the echo signal from the seabed and select the moment when the echo signal-to-noise ratio is the largest as the reference time t ref , use the MUSIC algorithm to estimate the DOA at this moment and obtain the angle reference value θ ref .

[0060] Here we assume that there is a uniform linear array composed of M array elements with an array element spacing of d. Consider that there are N narrowband signals with wavelength λ and different beam steering angles [θ1, θ2, ..., θ N ] is incident on the array. Based on this, the spatial spectrum function of the MUSIC algorithm is defined as follows:

[0061]

[0062] Where, P USIC (t ref ,θ) represents the spatial spectrum function, t ref represents the reference value at the time, θ=θ1,θ2,...,θ N Indicates the angle, a(t ref , θ)=[1, exp(-j2πsinθd / λ),…, exp[-j2πsinθ(M-1)d / λ]] T The sampling time is tref , j represents an imaginary number, U N is the noise subspace obtained by matrix decomposition of the MUSIC algorithm. When there is a target signal at angle θ, a H (t ref ,θ)U N Approaching 0, searching for the peak position of the spatial spectrum function, the DOA estimation value θ can be obtained ref As an angular reference, the superscript T indicates transpose, and the superscript H indicates conjugate.

[0063] Step 2: First, use the small-aperture MUSIC algorithm to obtain a range-beam domain matrix. The rows of the matrix represent angles, the columns represent distances, and the values ​​in the matrix reflect the strength of the echo signal. The columns represent the distance direction. In reality, there are many distances from the ship to the seabed, and the corresponding distance to the seabed must be found among these distances.

[0064] Then, find the behavior angle reference value θ in the range-beam domain matrix ref The entire column of data is specified, and the sampling time obtained by converting the distance corresponding to the maximum value in the column of data into the speed of sound is t ref , that is, θ ref The TOA value of the corresponding beam number is t ref Based on this, in the distance-beam domain matrix, from θ ref Search in the two directions of decreasing and increasing angles respectively. Assuming the seabed is flat, considering the TOA of adjacent beams are correlated, the expected TOA value t of the next beam can be calculated based on the TOA result of the current beam. exp .

[0065] The signal processing flow of multi-beam bathymetric sonar using the MUSIC algorithm is as follows: Figure 2 As shown in FIG, after the received echo signal undergoes steps such as AD sampling, orthogonal demodulation, and filtering extraction, multi-channel raw data is obtained as the input of the signal processing flow.

[0066] In one transmit and receive cycle, the multi-channel raw data matrix S1 is in the following form:

[0067]

[0068] Where s ML Represents the multi-channel raw data corresponding to the Lth sampling sequence of the Mth array element, where M is the number of array elements and L is the number of sampling points.

[0069] By performing pulse compression on the sampling sequence of each array element, high resolution in the time dimension can be achieved. The resulting multi-channel pulse compression data matrix S2 is in the following form:

[0070]

[0071] Where s ML ′ represents the multi-channel pulse compression data corresponding to the Lth sampling sequence of the Mth array element.

[0072] S2 is divided into blocks along the sampling sequence direction, and the number of snapshots is K (K is an odd number). Then the channel data matrix S before and after the sampling time l (K-1) / 2 is 2l for:

[0073]

[0074] Construct the MUSIC pseudo spectrum, then normalize the spectrum function and multiply it by the data block S 2l The average amplitude is as follows:

[0075]

[0076] Where, P MUSIC (l, θ) is the MUSIC spectrum function at sampling time l, P WM_MUSIC (l,θ) is P MUSIC The result of weighted average of (l, θ), s′ m(k+l) is the matrix S 2l Each sampling moment has a MUSIC spectrum function.

[0077] By traversing the sampling moments, the time-beam domain data P can be obtained WM_MUSIC , which is a two-dimensional matrix about the sampling time l and angle θ, can be expressed as follows:

[0078]

[0079] Where s nl Denotes the beam steering angle as θ n , the data corresponding to the sampling time is l. Subsequently, the time-beam domain data is converted into distance-beam domain data through sound speed conversion, which can be used in step 2. Figure 3a 、 Figure 3b and Figure 3c These are the range beam domain distribution maps finally obtained based on the large-aperture CBF algorithm, the small-aperture MUSIC algorithm, and the method of this embodiment under flat terrain; Figure 4a 、 Figure 4b 、 Figure 4c These are the range beam domain distribution maps finally obtained based on the large-aperture CBF algorithm, the small-aperture MUSIC algorithm, and the method of this embodiment under relatively inclined terrain.

[0080] Next, the expected TOA value t of the next beam is obtained using the following formula: exp :

[0081]

[0082] Where, (t ref ,θ ref ) is the reference value, θ c is the beam steering angle corresponding to the desired beam.

[0083] Step 3: Take the expected value of TOA t exp As the center, select a certain range and weight the amplitude sequence of the beam. The selected range is the beam footprint width T of the beam foot , where H is the seabed depth, θ is the beam steering angle, Θ is the beam main lobe width, c is the speed of sound, and τ is the signal time width after pulse compression.

[0084]

[0085] Close to t exp The signal at the moment of t is given a greater weight, and the signal away from t exp The signal moment of the beam is given a smaller weight, so that the subsequent amplitude method tends to find the maximum value within the beam footprint of the beam when performing peak search.

[0086] Step 4: Use a fixed-length sliding window to smooth the signal amplitude, and then perform a peak search to obtain the TOA value t of the beam. c .

[0087] Step 5: If the angle θ corresponding to the beam c Satisfy θ c =θ1, where θ1 is the beam steering angle corresponding to the first beam, the search ends; otherwise, (t c ,θ c ) is the new reference value, and go to step 2. From the initial reference value toward the direction of the first beam, perform a modified peak search to achieve depth estimation of the left half. Similarly, modify the termination condition in step 5 to θ c =θ N , performing a modified peak search from the initial reference value toward the direction of the last beam, the depth estimation of the right half can be achieved.

[0088] Regarding the beam number, the number of beams and the angle of coverage are specified in the multi-beam system. For example, if the number of beams is 160 and the angle of coverage is 120°, then the beam numbers are 1 to 160, and the beam control angles corresponding to each beam are θ1 to θ 160Among beams 1 to 160, the central beam number is 80, and the angle corresponding to the central beam is specified to be 0°. Then, the angles corresponding to the beams on the left (beams 1 to 80) are -60° to 0°, and the angles corresponding to the beams on the right (beams 80 to 160) are 0° to 60°. Therefore, θ1 = -60°, θ 160 =60°.

[0089] Step 6: Following these steps, the seafloor depths corresponding to the first to last beams are arranged. The multibeam bathymetry system achieves seafloor depth measurement within the beam coverage angle using a small array aperture. This not only ensures high angular resolution but also yields more accurate seafloor depth measurements than the WMT algorithm, providing a more realistic representation of the seafloor topography.

[0090] The method of this embodiment was used to conduct an experiment to compare the depth measurement effects of the WMT algorithm based on large aperture CBF, the WMT algorithm based on small aperture MUSIC, and the method of this embodiment based on small aperture MUSIC under different terrains. The error percentages of the latter two algorithms relative to the first algorithm are as follows: Figure 5a 、 Figure 5b 、 Figure 6a 、 Figure 6b As shown in the figure. Under both terrain types, the WMT algorithm based on small-aperture MUSIC exhibits significant errors at edge angles. However, the method of this embodiment significantly reduces depth estimation errors at the edge of the seafloor, achieving bathymetric performance similar to that of the WMT algorithm using a large-aperture CBF at a small aperture. This comparison demonstrates that the method of this embodiment performs better than the other two algorithms, resolving the issue of strong pseudo-peak interference in the pseudo-spectra obtained by small-aperture MUSIC at the seafloor edge, which can lead to serious errors in WMT detection of seafloor edge terrain.

[0091] The above are merely preferred embodiments of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions based on the principles of the present invention are within the scope of protection of the present invention. It should be noted that for those skilled in the art, various improvements and modifications that do not depart from the principles of the present invention should be considered within the scope of protection of the present invention.

Claims

1. A method for measuring seabed depth based on modified peak search, characterized in that: The steps include: S1: Use a multi-beam system to measure the echo signal of the seabed, select the moment with the largest signal-to-noise ratio in the echo signal as the time reference value, use the MUSIC algorithm to estimate the DOA of the time reference value, and obtain the angle reference value; S2: Establish a range-beam domain matrix with angles as rows and distances as columns, where the elements represent the echo signal strength. The TOA value of the beam corresponding to the angle reference value is equal to the time reference value. The TOA value of the beam corresponding to the angle reference value is converted to the sound speed to obtain the corresponding seabed depth. S3: Search the first direction based on the angle reference value: S3.1: If the angle of the beam corresponding to the angle reference value is equal to the beam steering angle of the first beam in the echo signal, the search ends. Otherwise, starting from the angle reference value, the search proceeds toward the row with decreasing angles in the range-beam domain matrix. The expected TOA value of the next beam is calculated based on the TOA value of the beam corresponding to the angle reference value. S3.2: Calculate the beam footprint of the next beam, centered around the expected TOA value of the first beam, and weight the amplitude sequence of the next beam within the beam footprint. S3.3: Perform a peak search on the weighted amplitude sequence, obtain the TOA value of the next beam based on the peak value, and convert the TOA value of the next beam into the speed of sound to obtain the corresponding seabed depth; S3.4: If the angle of the next beam is equal to the beam steering angle of the first beam, stop searching in the direction of the decreasing angle. Otherwise, use the angle of the next beam as the new angle reference value and continue searching in S3.1 until the search is complete. S4: Search in the second direction according to the angle reference value: S4.1: If the angle of the beam corresponding to the angle reference value is equal to the beam steering angle of the last beam in the echo signal, the search ends. Otherwise, starting from the angle reference value, the search proceeds toward the rows with increasing angles in the range-beam domain matrix. The expected TOA value of the next beam is calculated based on the TOA value of the beam corresponding to the angle reference value. S4.2: Calculate the beam footprint of the next beam, centered around the expected TOA value of the first beam, and weight the amplitude sequence of the next beam within the beam footprint. S4.3: Perform a peak search on the weighted amplitude sequence, obtain the TOA value of the next beam based on the peak value, and convert the TOA value of the next beam into the speed of sound to obtain the corresponding seabed depth; S4.4: If the angle of the next beam is equal to the beam steering angle of the last beam, stop searching in the direction of the increasing angle. Otherwise, use the angle of the next beam as the new angle reference value and continue searching in S4.1 until the search is complete. S5: Arrange the seabed depths corresponding to the first beam to the last beam to obtain the seabed depth within the beam coverage angle.

2. The method for seabed depth measurement based on modified peak search according to claim 1, characterized in that: In S1, the MUSIC algorithm is used to estimate the DOA of the time reference value to obtain the angle reference value, as follows: The echo signal is received by a uniform linear array composed of M array elements with an array element spacing of d. The echo signals with N wavelengths of λ are received at different beam steering angles [θ1, θ2, ..., θ N ] is incident on a uniform linear array; the following spatial spectrum function is established using the MUSIC algorithm: Where, P MUSIC (t ref ,θ) represents the spatial spectrum function, t ref represents the reference value at the time, θ=θ1,θ2,...,θ N Indicates the angle, a(t ref , θ)=[1, exp(-j2πsinθd / λ),…, exp[-j2πsinθ(M-1)d / λ]] T The sampling time is t ref , j represents an imaginary number, U N The noise subspace obtained by matrix decomposition of the MUSIC algorithm, the superscript T represents transpose, and the superscript H represents conjugate; Search the peak position of the spatial spectrum function to obtain the DOA estimate θ ref , and used as the angle reference value.

3. The method for seabed depth measurement based on modified peak search according to claim 1, characterized in that: In S2, the range-beam domain matrix is ​​established as follows: After AD sampling, orthogonal demodulation and filtering extraction of the echo signal, the multi-channel raw data matrix S1 is obtained as follows: Where s ML Represents the multi-channel raw data corresponding to the Lth sampling sequence of the Mth array element, where M is the number of array elements and L is the number of sampling points; Perform pulse compression operation on S1 to obtain the multi-channel pulse compression data matrix S2: Where s ML ′ represents the multi-channel pulse compression data corresponding to the Lth sampling sequence of the Mth array element; S2 is divided into blocks along the sampling sequence direction, and the number of snapshots is K. Then the channel data matrix S before and after the sampling time l (K-1) / 2 is 2l for: Construct the MUSIC pseudo spectrum, normalize the spectrum function and multiply it by the data block S 2l The average amplitude of , we get: Where, P MUSIC (l, θ) is the MUSIC spectrum function at sampling time l, P WM_MUSIC (l,θ) is P MUSIC The result of weighted average of (l, θ), s′ m(k+l) is the matrix S 2l Channel data in; Traverse the sampling moments to obtain the time-beam domain data matrix P WM_MUSIC , the time-beam domain data matrix is ​​a two-dimensional matrix about the sampling time l and the angle θ; The time-beam domain data matrix is ​​converted into the distance-beam domain data matrix through sound speed conversion.

4. The method for seabed depth measurement based on modified peak search according to claim 1, characterized in that: In S3.1 and S4.1, the expected TOA value of the next beam is calculated based on the TOA value of the beam corresponding to the angle reference value. The formula is as follows: Where, t exp represents the expected TOA value of the next beam, t ref Indicates the reference value at the moment, θ c is the beam steering angle corresponding to the desired beam.

5. The method for seabed depth measurement based on modified peak search according to claim 1, characterized in that: In S3.2 and S4.2, the beam footprint of the next beam is calculated with the TOA expected value of the next beam as the center. The formula is as follows: Where, T foot represents the beam footprint width, H is the seabed depth, θ is the angle, Θ is the beam main lobe width, c is the speed of sound, and τ is the signal time width after pulse compression.

6. The method for seabed depth measurement based on modified peak search according to claim 1, characterized in that: In S3.2 and S4.2, the amplitude sequence of the next beam is weighted within the beam footprint width. Specifically, within the beam footprint width, the signal closer to the expected TOA value is given a larger weight, and the signal farther away from the expected TOA value is given a smaller weight.

Images