A method and system for distinguishing the depth of deep-sea sound sources based on interference spectrum

Abstract

The present invention relates to the technical field of underwater acoustic engineering sonar, and in particular to a method and system for distinguishing the depth of deep-sea sound sources based on interference spectrum. The method comprises: step 1) receiving a broadband sound pressure signal of a target sound source by deploying a horizontal linear array on the seabed of a designated sea area, and obtaining environmental parameters of the sea area by means of an underwater acoustic device; step 2) converting the broadband sound pressure signal into sound intensity, and performing a two-dimensional Fourier transform to obtain an interference spectrum; step 3) performing subspace division on the interference spectrum, and calculating the energy proportion of the interference spectrum in different subspaces; step 4) according to the environmental parameters of step 1), simulating the received interference fringes when different grid virtual points are used as the sound source position by means of the sound field calculation model KRAKEN, and calculating the energy ratio between the subspaces on the interference spectrum, and obtaining a judgment threshold according to a given false alarm probability; step 5) comparing the energy proportion of the interference spectrum obtained in step 3) with the judgment threshold obtained in step 4), and distinguishing the depth type of the target sound source.

Description

Technical Field

[0001] The present invention relates to the technical field of underwater acoustic engineering sonar, and in particular to a method and system for distinguishing the depth of a deep-sea sound source based on an interference spectrum. Background Art

[0002] Underwater target positioning has been a hot topic in hydroacoustics in recent years. Current deep-sea sound source positioning methods mainly include matched field processing technology, multi-path arrival delay-based methods, and interference fringe-based methods. Among them, traditional matched field processing technology is more sensitive to ocean environmental parameters, and the calculation is large and time-consuming under deep-sea conditions, making it difficult to be widely used in deep-sea sound source positioning; the effect of the multi-path arrival delay-based method depends on the accuracy of the delay estimation, and the error is large in a low signal-to-noise ratio environment, and a small aperture array is required to improve the delay resolution.

[0003] In shallow waters, sound intensity presents a stable interference structure, based on which many related sound source localization methods have emerged. In the deep sea, the sound intensity also changes with distance and will show alternating light and dark fringes. This interference fringes is closely related to the location of the sound source and is widely used for deep sea sound source localization. For example, the method in reference [1] ("Passive ranging and depth measurement method of sound source in the shadow zone of a single hydrophone in the deep sea", published in the Journal of Acoustics, Vol. 43, No. 6, November 2018, starting with page 905) uses a single hydrophone to receive the signal and perform time-frequency analysis to extract two frequency domain interference fringe periods, which are then matched with the interference period calculated by simulation to estimate the depth and distance of the sound source. However, this type of method requires long-term observation of the continuously moving sound source to obtain obvious interference fringes. At the same time, the first frequency domain interference period is relatively large, and the sound source must meet the corresponding bandwidth requirements to observe this type of interference fringe from the received signal and extract the corresponding interference period. In fact, using a horizontal array for reception can obtain a stable and clear interference pattern for a long time, avoiding the problem of single hydrophone positioning. This method can be found in reference [2] ("Passive Source Depth Discrimination in Deep-Water", published in IEEE J. Sel. Topics Signal Process. Vol. 13, No. 1, March 2019, starting at page 185). This method uses a horizontal array to obtain interference patterns and perform a two-dimensional Fourier transform. The sound source depth is identified by calculating the modal energy ratio and comparing it with a given threshold. However, this type of method is applicable to complete deep-sea sound channels, while the South my country Sea is mainly composed of incomplete sound channels; at the same time, this method requires the horizontal array to be arranged on the sound channel axis, which is extremely costly and difficult to achieve in practical applications. Therefore, the article does not provide corresponding experimental verification. Summary of the invention

[0004] The purpose of the present invention is to overcome the defects of the prior art, obtain signal interference fringes by using a substrate horizontal array, and propose a method and system for distinguishing the depth of deep-sea sound sources based on interference spectra.

[0005] In order to achieve the above object, the present invention proposes a method for distinguishing the depth of deep-sea sound sources based on interference spectrum, comprising:

[0006] Step 1) receiving broadband sound pressure signals from target sound sources by deploying a horizontal linear array on the seabed of a designated sea area, and obtaining environmental parameters of the sea area by using hydroacoustic equipment;

[0007] Step 2) converting the received broadband sound pressure signal into sound intensity, and performing a two-dimensional Fourier transform on the sound intensity to obtain an interference spectrum;

[0008] Step 3) Divide the interference spectrum into subspaces according to the different types of interference spectrum peaks, and calculate the energy proportions of the interference spectrum in different subspaces;

[0009] Step 4) According to the environmental parameters obtained in step 1), the received interference fringes when different grid virtual points are used as the sound source position are simulated by the sound field calculation model KRAKEN, and the energy ratio between the subspaces on the interference spectrum is calculated, and the judgment threshold is obtained according to the given false alarm probability;

[0010] Step 5) compares the interference spectrum energy ratio obtained in step 3) with the judgment threshold obtained in step 4) to determine the depth type of the target sound source.

[0011] Preferably, the step 1) obtaining environmental parameters of the sea area by hydroacoustic equipment comprises:

[0012] The seawater sound velocity profile is obtained through the temperature, salinity and depth measuring instrument; the seawater depth and topographic information is obtained through the multi-beam system or nautical chart.

[0013] Preferably, the step 2) comprises:

[0014] The broadband sound pressure signal received by the horizontal array of length L is processed by modulus square to obtain the received sound intensity I(r,f), and then the sound intensity is transformed into the interference spectrum by two-dimensional Fourier transform.

[0015]

[0016] Among them, r0 and f0 represent the center distance of the horizontal linear array and the center frequency of the sound source respectively, x and y are the horizontal and vertical coordinates of the interference spectrum, representing the oscillation frequency of the interference fringes in the distance axis and frequency axis directions, r represents the distance between the sound source and the horizontal linear array, f represents the sound source frequency, B represents the sound source signal bandwidth, and i represents the imaginary part.

[0017] Preferably, the interference spectrum peak types in step 3) include: TM-TM, TM-BIM, TM-SIBIM, BIM-BIM, BIM-SIBIM and SIBIM-SIBIM.

[0018] Preferably, the step 3) comprises:

[0019] According to the different types of interference spectrum peaks and the characteristics of the incomplete sound channel, the interference spectrum subspace division boundaries y1(x), y2(x), y3(x) and y4(x) are obtained by the following formula:

[0020]

[0021]

[0022]

[0023]

[0024] Among them, c min and c bottom are the minimum value of the sound velocity profile and the sound velocity near the seabed, respectively. β1 and β2 are the waveguide invariants corresponding to different types of modes, respectively.

[0025] The interference spectrum is subspace-divided according to the interference spectrum subspace division boundary, wherein subspace D0 is divided by y1, y2 and y4, and subspace D1 is divided by y1, y2 and y3;

[0026] The interference spectrum energy ratio τ of subspaces D0 and D1 is obtained according to the following formula:

[0027]

[0028] Preferably, the step 4) comprises:

[0029] According to the environmental parameters obtained in step 1), the acoustic field calculation model KRAKEN is used to calculate and simulate the interference fringes when different grid virtual points are emitted, and the area where the sound source to be located may exist is grid-divided, and each grid point is a grid virtual point. According to the set grid point depth, distance range, depth grid step and distance grid step, as well as the set interference fringes center frequency, bandwidth and distance width calculated for each grid point, the energy ratio of different subspaces on the corresponding interference spectrum is calculated;

[0030] The ROC curve is drawn according to the simulation results, and the corresponding judgment threshold ν is obtained from the ROC curve according to the given false alarm probability.

[0031] Preferably, the step 5) comprises:

[0032] Comparing the interference spectrum energy ratio τ obtained in step 3) and the judgment threshold ν obtained in step 4), the following formula is obtained:

[0033] H0:0 <z s ≤z lim (if τ≤υ)

[0034] H1:z s >z lim (if τ>υ)

[0035] Among them, z s is the target sound source depth, z lim To determine the depth;

[0036] When the judgment result is H0, the target sound source depth is less than or equal to the judgment depth, and the target sound source is the sea surface sound source;

[0037] When the judgment result is H1, the target sound source depth is greater than the judgment depth, and the target sound source is an underwater sound source; thus, the deep-sea sound source depth discrimination is achieved.

[0038] On the other hand, the present invention proposes a deep-sea sound source depth determination system based on interference spectrum, comprising:

[0039] The receiving module is used to receive the broadband sound pressure signal of the target sound source through a horizontal linear array deployed on the seabed of the designated sea area, and obtain the environmental parameters of the sea area through the hydroacoustic equipment;

[0040] A conversion module, used for converting the received broadband sound pressure signal into sound intensity, and performing a two-dimensional Fourier transform on the sound intensity to obtain an interference spectrum;

[0041] An interference spectrum energy ratio calculation module is used to divide the interference spectrum into subspaces and calculate the interference spectrum energy ratios in different subspaces;

[0042] The judgment threshold acquisition module is used to simulate the receiving interference fringes when different grid virtual points are used as the sound source position through the sound field calculation model KRAKEN according to the environmental parameters obtained by the receiving module, and calculate the energy ratio between the subspaces on the interference spectrum, and obtain the judgment threshold according to the given false alarm probability;

[0043] The discrimination module is used to compare the interference spectrum energy ratio with the judgment threshold to discriminate the depth type of the target sound source.

[0044] Compared with the prior art, the advantages of the present invention are:

[0045] The method of the present invention utilizes the interference spectrum obtained after two-dimensional Fourier transformation of the interference pattern to realize the depth discrimination of deep-sea sound sources, and is suitable for the incomplete sound channel marine environment of the South China Sea; the method converts the interference structure from the stripe form to the spectrum peak form on the interference spectrum, and can observe the interference structure that cannot be observed due to bandwidth limitations; at the same time, the method adopts the base horizontal array to receive the signal to obtain the interference stripes, which reduces the difficulty of laying out the receiving array in practical application. In addition, the base horizontal array has a long available time and can realize the depth discrimination of the target within a certain time and space range. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 It is the sound velocity profile actually measured in the offshore experiment in the embodiment of the present invention, the seabed sediment parameters used in the simulation, and the phase velocity ranges of different types of simple normal wave modes.

[0047] FIG2( a ) is an interference spectrum of a 50 m sound source signal in an embodiment of the present invention;

[0048] FIG2( b ) is an interference spectrum of a 200 m sound source signal in an embodiment of the present invention;

[0049] Figure 3 are the phase slowness and group slowness of different types of simple normal wave modes in the embodiment of the present invention;

[0050] Figure 4 is a schematic diagram of the boundary of the interference spectrum subspace division in an embodiment of the present invention;

[0051] FIG5(a) is a result of subspace division of the interference spectrum of a 50m sound source signal in an embodiment of the present invention;

[0052] FIG5( b ) is a result of subspace division of the interference spectrum of the 200 m sound source signal in an embodiment of the present invention.

[0053] Figure 6 is a simulated receiver operating characteristic curve in an embodiment of the present invention;

[0054] Figure 7 is the sound source depth determination result of the offshore experiment in the embodiment of the present invention;

[0055] Figure 8 It is a flow chart of the method for distinguishing the depth of deep-sea sound sources based on interference spectrum of the present invention. DETAILED DESCRIPTION

[0056] The present invention proposes a method for distinguishing the depth of deep-sea sound sources based on interference spectrum, comprising:

[0057] Step 1) receiving broadband sound pressure signals from target sound sources by deploying a horizontal linear array on the seabed of a designated sea area, and obtaining environmental parameters of the sea area by using hydroacoustic equipment;

[0058] Step 2) converting the received broadband sound pressure signal into sound intensity, and performing a two-dimensional Fourier transform on the sound intensity to obtain an interference spectrum;

[0059] Step 3) Divide the interference spectrum into subspaces and calculate the energy proportion of the interference spectrum in different subspaces;

[0060] Step 4) According to the environmental parameters obtained in step 1), the received interference fringes when different grid virtual points are used as the sound source position are simulated by the sound field calculation model KRAKEN, and the energy ratio between the subspaces on the interference spectrum is calculated, and the judgment threshold is obtained according to the given false alarm probability;

[0061] Step 5) compares the interference spectrum energy ratio obtained in step 3) with the judgment threshold obtained in step 4) to determine the depth type of the target sound source.

[0062] The technical solution of the present invention is described in detail below with reference to the accompanying drawings and embodiments.

[0063] Example 1

[0064] A method for distinguishing the depth of a deep-sea sound source based on an interference spectrum is proposed in Example 1 of the present invention. First, a horizontal array is arranged on the bottom of the deep sea to receive a broadband sound pressure signal radiated by a target sound source; the original sound pressure signal received by the array is converted into sound intensity, and the sound intensity is subjected to a two-dimensional Fourier transform to obtain an interference spectrum; the interference spectrum is divided into subspaces, and the energy proportion of the modal combination in different subspaces is calculated; the environmental parameters are used for simulation calculation to obtain the simulation value of the energy proportion of the modal combination at different sound source depths, and the judgment threshold of the energy proportion is obtained under a given target false alarm probability; the energy ratio calculated by the experimental data is compared with the judgment threshold obtained by the simulation, and the depth of the deep-sea sound source can be distinguished according to the result. The method uses a base horizontal array to obtain interference fringes, has low system complexity, is easy to arrange and operate, and can achieve long-term target depth discrimination. At the same time, the subspace division boundary of the interference spectrum of the method is obtained based on the condition of a non-complete sound channel, so it is suitable for most of the South my country Sea.

[0065] The implementation process of the present invention is divided into the following steps:

[0066] Step 1: Deploy a horizontal array in the designated sea area. The array is located on the seabed and is used to receive the broadband sound pressure signal radiated by the target sound source and use other hydroacoustic equipment to obtain the environmental parameters of the sea area.

[0067] This embodiment selects a sea trial experiment in the South China Sea. The sea depth of the experimental sea area is 1760m. The horizontal array is placed at a depth of 1750m on the seabed. The array length is 1.2km, the array element interval is 15m, and the hydrophone sampling rate is 16kHz. The experimental sound source is a broadband explosion sound source of 200m and 50m. The experimental ship sails along the survey line and throws the explosion sound source at different distances. In the experiment, the seawater sound velocity profile is obtained through the temperature-salinity-depth meter (CTD), and the seawater depth and topography information is obtained through the multi-beam system or the nautical chart. The results are as follows: Figure 1 This sea area is a typical incomplete sound channel environment. Under the simple normal wave theory, the modes are divided into the following three types according to whether the modes interact with the seabed and the sea surface. The classification results are shown in Figure 1 Has been shown in.

[0068] Reversal Mode (TM)

[0069] ·Bottom Interaction Model (BIM)

[0070] Seabed Interaction-Sea Surface Interaction Mode (SIBIM)

[0071] Subsequently, based on the above modal theory, we will take two different sound source depth signals as examples (the horizontal distance of the sound source at a depth of 50m is 16.18km; the horizontal distance of the sound source at a depth of 200m is 16.81km) to demonstrate the complete sound source depth discrimination method.

[0072] Step 2: Convert the target sound pressure signal received by the horizontal array into sound intensity, and perform a two-dimensional Fourier transform on the sound intensity to obtain an interference spectrum.

[0073] The original sound pressure signal received by the horizontal array of length L is processed by modulus square to obtain the received sound intensity I(r,f), and then the sound intensity is transformed into the interference spectrum by two-dimensional Fourier transform.

[0074]

[0075] Where r0 and f0 represent the array center distance and the sound source center frequency, respectively. x and y are the horizontal and vertical coordinates of the interference spectrum, representing the oscillation frequency of the interference fringes in the distance axis and frequency axis directions. r represents the distance between the sound source and the horizontal linear array, f represents the sound source frequency, B represents the sound source signal bandwidth, and i represents the imaginary part. According to the above formula, the received sound intensity is transformed to obtain the interference spectrum, as shown in Figure 2. Since the interference spectrum is symmetrical about the origin center, Figure 2 only shows the right half of the interference spectrum (x ≥ 0), where Figure 2 (a) corresponds to the 50m sound source depth signal; Figure 2 (b) corresponds to the 200m sound source depth signal. The interference structure is converted from fringes to spectral peaks on the interference spectrum through two-dimensional Fourier transform. Since the modes are divided into three types in step 1, and the interference fringes are formed by the interaction of two modes, the spectral peak types corresponding to the interference fringes are also divided into the following types: TM-TM, TM-BIM, TM-SIBIM, BIM-BIM, BIM-SIBIM and SIBIM-SIBIM.

[0076] Step 3: Divide the interference spectrum into subspaces according to the different types of interference spectrum peaks, and calculate the energy proportion of the interference spectrum in different subspaces.

[0077] The interference spectrum is divided into subspaces. For incomplete channels, what needs to be paid attention to is the energy proportion of the spectrum peak containing TM or BIM. Therefore, the corresponding subspace division boundary is given by the following formula:

[0078]

[0079]

[0080]

[0081]

[0082] where c min and c bottom are the minimum value of the sound velocity profile and the sound velocity near the seabed, respectively. Figure 1 As shown in ; β is the waveguide invariant, which is the inverse of the ratio of the difference between the phase slowness of two modes and the difference between the group slowness. According to the measured environmental parameters in this embodiment, the phase slowness and group slowness of different modes are calculated using the acoustic field calculation model KRAKEN, as shown in Figure 3As shown, β is the opposite of the slope of the curve. When two modes belong to the same type, β can be approximately considered as a constant. Here, when the modes all belong to TM, β≈-4, and when the modes all belong to BIM or SIBIM, β≈1.5. β1 and β2 are the waveguide invariants corresponding to different types of modes, respectively. The sound field modes can be calculated using KRAKEN sound field software, and are calculated from the modal phase slowness and group slowness. For the subspace division boundary, y1 is the TM-TM boundary, so β1 is -4, y2, y3 and y4 are the BIM-SIBIM and SIBIM-SIBIM boundaries, so β2 is 1.5, thus two subspaces can be obtained, as shown Figure 4 As shown, subspace D0 is divided by y1, y2 and y4, represented by a blue solid line, and subspace D1 is divided by y1, y2 and y3, represented by an orange dotted line. The interference spectrum obtained in step 2 is divided into subspaces, and two subspaces are obtained. The interference spectrum energy ratio in the two subspaces is calculated:

[0083]

[0084] The specific division results are shown in Figure 5, where Figure 5(a) corresponds to the result of the sound source depth of 50m, and the calculated energy ratio is 0.566; Figure 5(b) corresponds to the result of the sound source depth of 200m, and the calculated energy ratio is 0.767.

[0085] Step 4: Use the sound field calculation model KRAKEN to simulate the received interference fringes when different grid virtual points are used as the sound source position, and calculate the energy ratio between the subspaces on the interference spectrum to obtain the judgment threshold under a given false alarm probability.

[0086] Using the experimentally measured environmental parameters, the acoustic field calculation model KRAKEN is used to perform calculations and simulate the interference fringes when different grid virtual points are emitted. The area where the sound source to be located may exist is grid-divided, and each grid point is the above-mentioned grid virtual point. The depth range of the simulated grid points is 0-300m, and the distance range is 0-50km; considering the calculation accuracy and speed, it is recommended to set the depth grid step to 1m and the distance grid step to 0.5km; the center frequency of the interference fringes calculated at each grid point is 355Hz, the bandwidth is 20Hz, and the distance width is 1.2km, and then the energy ratio of different subspaces on the interference spectrum is calculated according to step 3. In the embodiment, the discrimination depth is set to 65m, and the sound sources above and below the discrimination depth are considered to be sea surface sound sources and underwater sound sources. The probability of correctly discriminating the sea surface sound source is recorded as P D (detection probability), the probability of misidentifying an underwater sound source as a surface sound source is recorded as P FA (false alarm probability), and thus the receiver operating characteristic (ROC) curve can be drawn according to the simulation results, such as Figure 6As shown, when the false alarm probability is given (in this embodiment P FA =10%), the corresponding judgment threshold ν (0.664 in this embodiment) can be obtained from the ROC curve.

[0087] Step 5: Compare the interference spectrum subspace energy ratio and the judgment threshold to determine the depth type of the target sound source.

[0088] By comparing the subspace energy ratio τ obtained in step 3 and step 4 with the judgment threshold ν, there are two judgment results:

[0089] H0:0 <z s ≤z lim (if τ≤υ)

[0090] H1:z s >z lim (if τ>v)

[0091] where z s is the target sound source depth, z lim To determine the depth, the above two judgment results can be used to determine whether the target sound source belongs to the sea surface sound source or the underwater sound source, thereby realizing the sound source depth determination. According to the energy ratio and judgment threshold obtained in step 3 and step 4, it can be obtained that Figure 5(a) belongs to the H0 judgment result, so the sound source is a sea surface sound source (depth less than 65m); Figure 5(b) belongs to the H1 judgment result, so the sound source is an underwater sound source (depth greater than 65m).

[0092] The above steps are used to process the sound source signals at other distances and depths, and the final results are as follows: Figure 7 As shown in the figure, it can be seen that the results of different sound source depth discrimination are basically consistent with the actual depth. The individual error sources mainly include the undulation of the seabed topography and the influence of environmental noise. The overall sound source depth discrimination accuracy is 78%. The measured data verification shows that the method of the present invention can effectively realize the deep-sea sound source depth discrimination under incomplete sound channels. Figure 8 Shown is a flow chart of the method of the present invention.

[0093] Example 2

[0094] Embodiment 2 of the present invention proposes a deep-sea sound source depth determination system based on interference spectrum, which is implemented based on the method of embodiment 1 and includes:

[0095] The receiving module is used to receive the broadband sound pressure signal of the target sound source through a horizontal linear array deployed on the seabed of the designated sea area, and obtain the environmental parameters of the sea area through the hydroacoustic equipment;

[0096] A conversion module, used for converting the received broadband sound pressure signal into sound intensity, and performing a two-dimensional Fourier transform on the sound intensity to obtain an interference spectrum;

[0097] An interference spectrum energy ratio calculation module is used to divide the interference spectrum into subspaces and calculate the interference spectrum energy ratios in different subspaces;

[0098] The judgment threshold acquisition module is used to simulate the receiving interference fringes when different grid virtual points are used as the sound source position through the sound field calculation model KRAKEN according to the environmental parameters obtained by the receiving module, and calculate the energy ratio between the subspaces on the interference spectrum, and obtain the judgment threshold according to the given false alarm probability;

[0099] The discrimination module is used to compare the interference spectrum energy ratio with the judgment threshold to discriminate the depth type of the target sound source.

[0100] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention is described in detail with reference to the embodiments, it should be understood by those skilled in the art that any modification or equivalent replacement of the technical solutions of the present invention does not depart from the spirit and scope of the technical solutions of the present invention and should be included in the scope of the claims of the present invention.

Claims

1. A method for determining the depth of a deep-sea sound source based on an interference spectrum, comprising: Step 1) receiving broadband sound pressure signals from target sound sources by deploying a horizontal linear array on the seabed of a designated sea area, and obtaining environmental parameters of the sea area by using hydroacoustic equipment; Step 2) converting the received broadband sound pressure signal into sound intensity, and performing a two-dimensional Fourier transform on the sound intensity to obtain an interference spectrum; Step 3) Divide the interference spectrum into subspaces according to the different types of interference spectrum peaks, and calculate the energy proportions of the interference spectrum in different subspaces; Step 4) According to the environmental parameters obtained in step 1), the received interference fringes when different grid virtual points are used as the sound source position are simulated by the sound field calculation model KRAKEN, and the energy ratio between the subspaces on the interference spectrum is calculated, and the judgment threshold is obtained according to the given false alarm probability; Step 5) comparing the interference spectrum energy ratio obtained in step 3) with the judgment threshold obtained in step 4) to determine the depth type of the target sound source; The step 2) comprises: The broadband sound pressure signal received by the horizontal array of length L is processed by modulus square to obtain the received sound intensity I(r,f), and then the sound intensity is transformed into the interference spectrum by two-dimensional Fourier transform. Where r0 and f0 represent the center distance of the horizontal linear array and the center frequency of the sound source, respectively; x and y are the abscissa and ordinate of the interference spectrum, representing the oscillation frequency of the interference fringes in the distance axis and frequency axis directions; r represents the distance between the sound source and the horizontal linear array; f represents the sound source frequency; B represents the sound source signal bandwidth; and i represents the imaginary part; The interference spectrum peak types in step 3) include: TM-TM, TM-BIM, TM-SIBIM, BIM-BIM, BIM-SIBIM and SIBIM-SIBIM; The step 3) comprises: According to the different types of interference spectrum peaks and the characteristics of the incomplete sound channel, the interference spectrum subspace division boundaries y1(x), y2(x), y3(x) and y4(x) are obtained by the following formula: Among them, c min and c bottom are the minimum value of the sound velocity profile and the sound velocity near the seabed, respectively. β1 and β2 are the waveguide invariants corresponding to different types of modes, respectively. The interference spectrum is subspace-divided according to the interference spectrum subspace division boundary, wherein subspace D0 is divided by y1, y2 and y4, and subspace D1 is divided by y1, y2 and y3; The interference spectrum energy ratio τ of subspaces D0 and D1 is obtained according to the following formula:

2. The method for determining the depth of deep-sea sound sources based on interference spectrum according to claim 1, characterized in that: The step 1) obtaining the environmental parameters of the sea area by using a hydroacoustic device comprises: The seawater sound velocity profile is obtained through the temperature, salinity and depth measuring instrument; the seawater depth and topographic information is obtained through the multi-beam system or nautical chart.

3. The method for determining the depth of deep-sea sound sources based on interference spectrum according to claim 1, characterized in that: The step 4) comprises: According to the environmental parameters obtained in step 1), the acoustic field calculation model KRAKEN is used to calculate and simulate the interference fringes when different grid virtual points are emitted, and the area where the sound source to be located may exist is grid-divided, and each grid point is a grid virtual point. According to the set grid point depth, distance range, depth grid step and distance grid step, as well as the set interference fringes center frequency, bandwidth and distance width calculated for each grid point, the energy ratio of different subspaces on the corresponding interference spectrum is calculated; The ROC curve is drawn according to the simulation results, and the corresponding judgment threshold ν is obtained from the ROC curve according to the given false alarm probability.

4. The method for determining the depth of deep-sea sound sources based on interference spectrum according to claim 1, characterized in that: The step 5) comprises: Comparing the interference spectrum energy ratio τ obtained in step 3) and the judgment threshold ν obtained in step 4), the following formula is obtained: <h2 style=";text-align:left;direction:ltr">H0:0 <z <h2 style=";text-align:left;direction:ltr"> s <h2 style=";text-align:left;direction:ltr"> ≤z<h2 style=";text-align:left;direction:ltr"> lim <h2 style=";text-align:left;direction:ltr"> (ifτ≤v) H1:z s >from lim (ifτ>v) Among them, z s is the target sound source depth, z lim To determine the depth; When the judgment result is H0, the target sound source depth is less than or equal to the judgment depth, and the target sound source is the sea surface sound source; When the judgment result is H1, the target sound source depth is greater than the judgment depth, and the target sound source is an underwater sound source; thus, the deep-sea sound source depth discrimination is achieved.

5. A system based on the deep-sea sound source depth determination method based on interference spectrum according to claim 1, characterized in that: include: The receiving module is used to receive the broadband sound pressure signal of the target sound source through a horizontal linear array deployed on the seabed of the designated sea area, and obtain the environmental parameters of the sea area through the hydroacoustic equipment; A conversion module, used for converting the received broadband sound pressure signal into sound intensity, and performing a two-dimensional Fourier transform on the sound intensity to obtain an interference spectrum; An interference spectrum energy ratio calculation module is used to divide the interference spectrum into subspaces and calculate the interference spectrum energy ratios in different subspaces; The judgment threshold acquisition module is used to simulate the receiving interference fringes when different grid virtual points are used as the sound source position through the sound field calculation model KRAKEN according to the environmental parameters obtained by the receiving module, and calculate the energy ratio between the subspaces on the interference spectrum, and obtain the judgment threshold according to the given false alarm probability; and The discrimination module is used to compare the interference spectrum energy ratio with the judgment threshold to discriminate the depth type of the target sound source.

Images