Method for resolving shallow sea sound source depth based on vertical subarray cross-spectrum broadband phase fluctuations

Abstract

The present invention discloses a method for resolving the depth of a shallow - sea sound source based on the broadband phase fluctuation of the cross - spectrum of a vertical sub - array. The method comprises the following steps: extracting the broadband cross - spectrum of the complex sound pressure of the upper half - array and the lower half - array of the vertical array, and using the number of zero - crossing points of the broadband cross - spectrum phase fluctuation as a depth - dependent feature; the number of zero - crossing points above the threshold corresponds to a water - surface target, and vice versa for an underwater target. The advantage of this method is its low computational complexity, which is conducive to the engineering implementation of an underwater real - time processing system.

Description

Technical Field

[0001] The present invention belongs to the technical field of underwater acoustic positioning and recognition, and particularly relates to a method for shallow sea sound source depth resolution based on cross-spectrum broadband phase fluctuation of a vertical sub-array. Background Art

[0002] Bucker first proposed the concept of Matched Field Processing (MFP), which uses measured acoustic information to estimate the distance and depth of a sound source. However, MFP is very sensitive to changes in environmental parameters and requires a large array aperture. Since the modal functions of each order of normal modes change with depth, and a vertical line array (VLA) is used for the aperture in the vertical direction, sampling in the depth direction can be achieved. In view of this, some scholars proposed Matched Mode Processing (MMP) based on a vertical array to deal with the discrimination problem of surface and underwater targets. MMP first uses an acoustic field simulation program to calculate the modal functions corresponding to the depths of each array element according to environmental parameters and constructs a modal decomposition matrix. Then, the modal filtering is performed using the signals received by the vertical array and the modal decomposition matrix to calculate the modal intensities of each order excited by the sound source, and the discrimination of the target depth is achieved through the distribution of the modal intensities. It should be noted that, in the case of a finite array aperture and a finite number of array elements, the modal decomposition matrix is not orthogonally complete, which involves the selection of a modal inverse matrix.

[0003] Creamer and Premus considered that for a waveguide with a pressure release surface, all modal excitation functions are zero near the water surface, while for deeper depths, there is always a first-order normal mode that can reach the maximum value. Due to the influence of surface waves and internal waves, the depth of the target will randomly change with the undulation of the waves. For a target at the water surface, since the modal excitation function is near the zero crossing point and the derivative is the largest, the value of the modal excitation function changes violently; while for a target at a greater depth, since there is a certain order of mode that makes the modal excitation function at the maximum value and the derivative is zero, the change in the value of the modal excitation function is small. In view of this, Premus defined the modal scintillation coefficient to consider the fluctuation of the modal excitation function as a statistic for binary depth discrimination. When the waveguide environmental parameters are known, it is necessary to calculate the modal scintillation coefficients of each order of normal modes in the acoustic field simulation multiple times as the basis for depth discrimination, and the computational amount is large. In addition, based on the modal filtering theory, Premus et al. considered the differences between the trapped normal mode (RBR) and the free mode (SRBR) excited by surface and underwater targets in a negative gradient sound speed waveguide, and designed different subspace filters for target depth resolution. Premus also verified the feasibility of matched subspace resolution for horizontal arrays. In addition, scholars have proposed methods for target resolution based on horizontal arrays or using synthetic aperture methods to extract wavenumber features. Since this paper focuses on target resolution of vertical arrays, the problem of sound source target resolution of horizontal arrays will not be elaborated here.

[0004] The research on underwater target resolution in China started relatively late, and most of it is still in the stage of concept and principle verification. The typical research works are as follows: Yu Yun, Hui Junying et al. used the relationship between the reactive component of the complex acoustic intensity and the phase angle of the cross negative acoustic intensity of the sound pressure and vertical particle velocity of the double-receiving sensor normal modes to discriminate surface and underwater targets, improving the working frequency. However, the method is limited by the depth placement of the two receiving sensors and requires that there are 2 orders of normal modes that can propagate in the waveguide. Fang Shiliang et al. studied the method for estimating the distance and depth of a single hydrophone based on dispersion characteristics combined with time-frequency analysis. By using the multi-modal joint matching method, through the time-frequency filtering method of binary mask filtering to extract the required modes, by calculating the error between the actually extracted modal energy and the predicted modal energy, a cost function is established, and the depth of the sound source is determined by the modal energy matching method. The effectiveness of the method was verified by simulation in the shallow sea environment of the Pekeris waveguide model. Zhang Lu et al. used a vector vertical array to judge surface and underwater targets based on the positive and negative signs of the active component of the cross-spectrum function of sound pressure and particle velocity, and verified the method through numerical simulation and lake trial. Summary of the Invention

[0005] One of the objectives of the present invention is to provide a method for shallow sea sound source depth resolution based on the broadband phase fluctuation of the cross-spectrum of vertical sub-arrays, so as to solve the problems in the prior art of underwater target resolution, such as being limited by the depth deployment of receiving sensors and large computational complexity.

[0006] To achieve the above objective, the present invention provides the following technical solutions:

[0007] A method for shallow sea sound source depth resolution based on the broadband phase fluctuation of the cross-spectrum of vertical sub-arrays, the method comprising the following steps:

[0008] Extract the broadband cross-spectrum of the complex sound pressure of the upper half array and the lower half array of the vertical array, and use the number of zero-crossings of the broadband cross-spectrum phase fluctuation as the depth-dependent feature;

[0009] The number of zero-crossings above the threshold corresponds to the surface target, and vice versa for the underwater target.

[0010] Preferably, after performing a fast Fourier transform on the acoustic signals received by the N-element vertical line array, take the broadband cross-spectrum of the sum of the complex sound pressures received by the upper half array and the sum of the complex sound pressures received by the lower half array;

[0011] Take the phase of the broadband cross-spectrum to obtain the variation of the broadband cross-spectrum phase with frequency;

[0012] Take the number of zero-crossings of the broadband cross-spectrum phase and define it as the depth resolution feature quantity;

[0013] Use the depth resolution feature quantity to compare with the critical value of the acoustic field feature to determine whether the signal originates from a surface target or an underwater target.

[0014] Preferably, the method for obtaining the critical value of the acoustic field feature includes the following steps:

[0015] Based on the ocean environment information and the N-element vertical line array, use the Kraken model to obtain the broadband cross-spectrum of the replica field;

[0016] After obtaining the broadband phase fluctuation structure of the replica field based on the broadband cross-spectrum of the replica field, obtain the critical value of the acoustic field feature.

[0017] Preferably, the method for obtaining the critical value of the acoustic field feature is to obtain the number of zero-crossings of the broadband phase fluctuation structure of the replica field obtained by simulation at the critical depth as the critical value of the acoustic field feature.

[0018] Preferably, the broadband cross-spectrum of the sum of the complex sound pressures received by the upper half array and the sum of the complex sound pressures received by the lower half array is expressed as where B(Z s , ω) represents the broadband cross-spectrum, P up (ω) represents the sum of the complex sound pressures received by the upper half array, P down (ω) represents the sum of the complex sound pressures received by the lower half array, Z srepresents the depth of the sound source, ω represents the frequency, and p j p(j)(ω) represents the complex sound pressure received by the j-th array element.

[0019] Preferably, taking the phase of the broadband cross-spectrum is expressed as where represents the broadband cross-spectrum phase, and B(Z s , ω) represents the band cross-spectrum.

[0020] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0021] The advantages of this method are low computational complexity, easy engineering implementation for underwater real-time processing systems, and this sound source depth resolution method can be used for the development of shallow water underwater anti-submarine warning buoys. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 is the algorithm flowchart of the shallow water sound source depth resolution method based on the broadband phase fluctuation of the vertical subarray cross-spectrum of the present invention.

[0023] Figure 2 is the sound speed profile used in the simulation example, with a sea depth of 100 m and typical thermocline hydrological conditions.

[0024] Figure 3 are the curves of the cross-spectrum phase versus frequency corresponding to a target distance of 10 km and sound source depths of 5 m, 10 m, 50 m, and 70 m.

[0025] Figure 4 is the variation of the number of zero-crossings of the cross-spectrum phase fluctuation with the sound source distance and depth.

[0026] Figure 5 is the variation of the number of zero-crossings of the cross-spectrum phase corresponding to a water surface sound source with a depth of 5 m and an underwater sound source with a depth of 50 m with the distance in the simulation experiment. DETAILED DESCRIPTION OF THE INVENTION

[0027] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0028] A shallow water sound source depth resolution method based on the broadband phase fluctuation of the vertical subarray cross-spectrum of the present invention, the method comprising the following steps:

[0029] Extract the broadband cross-spectrum of the complex sound pressure of the upper half array and the lower half array of the vertical array, and use the number of zero-crossings of the broadband cross-spectrum phase fluctuation as the depth-dependent feature;

[0030] The number of zero-crossings above the threshold corresponds to surface targets, and vice versa for underwater targets.

[0031] In the present invention, the phase of the cross-spectrum of the broadband sound field received by the upper half of the vertical array and the broadband sound field received by the lower half of the vertical array has an obvious separability in the depth dimension in terms of the number of zero-crossings of its broadband fluctuations, and is independent of the sound source spectrum and the sound source distance. Therefore, this number of zero-crossings can be used as a depth-dependent feature to determine whether the target is underwater or on the surface by comparing it with a threshold.

[0032] Refer to Figure 1 , specifically, the method for shallow sea sound source depth resolution based on the broadband phase fluctuation of the cross-spectrum of vertical sub-arrays includes the following steps:

[0033] Step 1, after performing a fast Fourier transform on the acoustic signal received by the N-element vertical line array, take the sum P up (ω) of the complex acoustic pressures received by the upper half of the array and the sum P down (ω) of the complex acoustic pressures received by the lower half of the array to obtain the broadband cross-spectrum B(Z s , ω), expressed as:

[0034] where B(Z s , ω) represents the broadband cross-spectrum, P up (ω) represents the sum of the complex acoustic pressures received by the upper half of the array, P down (ω) represents the sum of the complex acoustic pressures received by the lower half of the array, Z s represents the depth of the sound source, ω represents the frequency, and p j (ω) represents the complex acoustic pressure received by the j-th array element.

[0035] Step 2, take the phase of the broadband cross-spectrum B(Z s , ω) to obtain the variation of the broadband cross-spectrum phase with frequency , that is, the broadband cross-spectrum phase fluctuation structure, expressed as:

[0036] where represents the broadband cross-spectrum phase, and B(Z s , ω) represents the frequency band cross-spectrum.

[0037] In Step 2 of the present invention, due to the downward refraction of the sound rays, the depth of the sound source corresponding to the surface target is very small, the proportion of the excited low-order modes is low, and the proportion of the high-order modes is high. At this time, the proportion of the fast-varying function of the phase with respect to frequency in the cross-spectrum coherence term is high, resulting in the phase of the cross-spectrum being a fast-varying function of frequency. Therefore, the cross-spectrum phase jumps quickly; conversely, for underwater targets, the depth of the sound source is large, the proportion of the excited low-order modes in the sound field increases, and the proportion of the corresponding high-order modes decreases after long-distance propagation. At this time, the proportion of the slow-varying function of the phase with respect to frequency in the cross-spectrum coherence term is high, resulting in the phase of the cross-spectrum being a slow-varying function of frequency. Therefore, the cross-spectrum phase jumps slowly.

[0038] Step 3: Obtain the number of zero-crossings of the broadband cross-spectrum phase and define it as the depth-resolution feature quantity Y test (Z s ).

[0039] Step 4: Obtain the critical value Y of the sound field characteristics lim .

[0040] Step 5: Use the depth-resolution feature quantity Y test (Z s ) and the critical value Y of the sound field characteristics lim to compare and determine whether the signal originates from a surface target or an underwater target: Y test (Z s ) > Y lim corresponds to the surface target H0; Y test (Z s ) < Y lim corresponds to the surface target H1.

[0041] In Step 4, the method for obtaining the critical value of the sound field characteristics includes the following steps:

[0042] Step 4.1: Based on the ocean environmental information and the N-element vertical line array, use the Kraken model to obtain the copy field broadband cross-spectrum;

[0043] Step 4.2: After obtaining the copy field broadband phase fluctuation structure based on the copy field broadband cross-spectrum, obtain the critical value of the sound field characteristics.

[0044] The method for obtaining the critical value of the sound field characteristics is to obtain the number of zero-crossings of the copy field broadband phase fluctuation structure obtained by simulation at the critical depth as the critical value of the sound field characteristics.

[0045] It should be noted that since it is necessary to determine whether the signal originates from a surface target or an underwater target, the critical depth value is usually set to 20 - 30 m underwater; since the distribution of the critical value of the sound field characteristics in the distance dimension is approximately distance-independent, the selection of the critical value of the sound field characteristics does not require the distance of the known target, and it can be assumed to be the medium distance within the preset distance range. For example, if the preset distance range is 0 - 20 km, the medium distance is 10 km.

[0046] Figure 2 The sound velocity profile used in the simulation experiment in Step 4 of the present invention is given, with a sea depth of 100 m and typical thermocline hydrological conditions.

[0047] Figure 3 The curves of the cross-spectrum phase versus frequency corresponding to the sound source depths of 5 m, 10 m, 50 m, and 70 m are given. It can be seen that the cross-spectrum phase corresponding to the shallow sound source (depths of 5 m and 10 m) jumps rapidly; the cross-spectrum phase corresponding to the deep sound source (depths of 50 m and 70 m) jumps slowly.

[0048] Figure 4 The number of zero-crossings of the cross-spectrum phase fluctuation with respect to the source distance and depth is given. It can be seen that when the source distance is greater than 2 km, there is an obvious separability between shallow sources and deep sources. The number of zero-crossings of the cross-spectrum phase corresponding to shallow sources is larger, while that corresponding to deep sources is smaller. In the range of source depth from 1 m to 30 m, as the source depth increases, the number of zero-crossings of the cross-spectrum phase gradually decreases, and the change rate is fast. When the source depth is greater than 30 m, as the source depth increases, the number of zero-crossings of the cross-spectrum phase does not change significantly. Moreover, the distribution of the number of zero-crossings of the cross-spectrum phase corresponding to shallow sources and deep sources is approximately independent of distance. Therefore, the number of zero-crossings of the cross-spectrum phase can be used as the basis for target depth resolution.

[0049] Figure 5 The variation of the number of zero-crossings of the cross-spectrum phase corresponding to a water surface source with a depth of 5 m and an underwater source with a depth of 50 m with respect to distance in the simulation experiment is given. It can be seen that in terms of the distribution of the number of zero-crossings, a clear distinction is made between the water surface source with a depth of 5 m and the underwater source with a depth of 50 m, that is, the resolution of water surface targets and underwater targets is achieved based on the number of zero-crossings of the cross-spectrum phase.

[0050] Aiming at the problem of target depth resolution of passive sound sources in shallow waters, this invention theoretically derives a depth-dependent feature for source depth resolution, defined as the phase of the cross-spectrum of the broadband sound field received by the upper half of a vertical array and the broadband sound field received by the lower half of the vertical array. The number of zero-crossings of its broadband fluctuation shows obvious separability in the depth dimension. Therefore, this number of zero-crossings is used as the depth-dependent feature, and corresponding source depth decision methods and decision criteria are proposed. The advantage of this method is its low computational complexity, which is easy to be implemented in the engineering of underwater real-time processing systems.

Claims

1. A method for resolving the depth of a shallow - sea sound source based on the broadband phase fluctuation of the cross - spectrum of vertical sub - arrays, characterized in that, The method includes the following steps: After performing a fast Fourier transform on the acoustic signals received by the N-element vertical line array, take the broadband cross-spectrum of the sum of the complex acoustic pressures received by the upper half array and the sum of the complex acoustic pressures received by the lower half array; Take the phase of the broadband cross-spectrum to obtain the variation of the broadband cross-spectrum phase with frequency; Take the number of zero-crossings of the broadband cross-spectrum phase and define it as the depth resolution feature quantity; Obtain the critical value of the acoustic field characteristics; Use the depth resolution feature quantity to compare with the critical value of the acoustic field characteristics to determine whether the signal originates from a surface target or an underwater target.

2. The method for resolving the depth of a shallow - sea sound source based on the broadband phase fluctuation of the cross - spectrum of vertical sub - arrays according to claim 1, characterized in that, The method for obtaining the critical value of the acoustic field characteristics includes the following steps: Based on the ocean environmental information and the N-element vertical line array, use the Kraken model to obtain the copy field broadband cross-spectrum; After obtaining the broadband phase fluctuation structure of the copy field based on the copy field broadband cross-spectrum, obtain the number of zero-crossings at the critical depth to obtain the critical value of the acoustic field characteristics.

3. The method for resolving the depth of a shallow - sea sound source based on the broadband phase fluctuation of the cross - spectrum of vertical sub - arrays according to claim 1, characterized in that, Take the sum P of the received complex sound pressures of the upper half array up (ω) and the sum P of the received complex sound pressures of the lower half array down (ω), and the broadband cross-spectrum B(Zs, ω) is expressed as: Among them, B(Zs, ω) represents the broadband cross-spectrum, and P up (ω) represents the sum of the received complex sound pressures of the upper half array, and P down (ω) represents the sum of the received complex sound pressures of the lower half array, Zs represents the depth of the sound source, ω represents the frequency, and p j (ω) represents the complex sound pressure received by the j-th array element.

4. The method for resolving the depth of a shallow - sea sound source based on the broadband phase fluctuation of the cross - spectrum of vertical sub - arrays according to claim 3, characterized in that, Taking the phase of the broadband cross-spectrum B(Zs, ω), the variation of the broadband cross-spectrum phase with frequency is obtained That is, the broadband cross-spectrum phase fluctuation structure, expressed as: Among them, represents the broadband cross-spectrum phase varying with frequency, and B(Zs, ω) represents the broadband cross-spectrum.

Images