A method and system for remote sensing of a rough seabed interface based on broadband reverberation data

Abstract

The present application relates to the field of underwater acoustic physics and seabed rough interface measurement, and particularly relates to a seabed rough interface remote sensing method and system based on broadband reverberation data; the method comprises the following steps: obtaining a frequency characteristic curve of seabed reverberation intensity; estimating a frequency characteristic curve of ocean reverberation intensity waveguide attenuation term; obtaining a seabed rough interface spectrum function based on the frequency characteristic curve of seabed reverberation intensity and the frequency characteristic curve of ocean reverberation intensity waveguide attenuation term; and obtaining statistical characteristic parameters of seabed rough interface random fluctuation by using a roughness spectrum function model according to the obtained seabed rough interface spectrum function, so as to realize seabed rough interface remote sensing. The system comprises a reverberation intensity acquisition module, a reverberation intensity waveguide attenuation term acquisition module, a seabed rough interface spectrum function acquisition module and a statistical characteristic parameter acquisition module. The statistical characteristic parameter acquisition module comprises a frequency index acquisition unit and a related parameter acquisition unit. The present application realizes rapid remote sensing of seabed rough interface.

Description

Technical Field

[0001] This invention relates to the fields of underwater acoustic physics and seabed roughness interface measurement, and particularly to a remote sensing method and system for seabed roughness interfaces based on broadband reverberation data. Background Technology

[0002] The seabed boundary is a crucial constraint on sound propagation in marine waveguides. It comprises variations in seabed topography and the superposition of small-scale random undulations. Theoretically, with sufficiently high resolution, the seabed boundary can be completely described. Current ocean depth measurement methods provide reliable and relatively accurate results for large-scale topographic changes; however, they struggle to achieve rapid, large-area measurements of small-scale rough interfaces (on the centimeter scale), failing to meet the demands of underwater acoustics.

[0003] Small-scale rough interfaces, often on the order of centimeters, tend to scatter incident acoustic energy, increasing sound propagation loss. Therefore, obtaining information about small-scale seabed rough interfaces is crucial for underwater acoustics research. American scholars were among the first to conduct measurement studies on this type of interface. In the late 1980s, divers manually mapped a 1.8-meter profile of a perturbation-prone seabed interface. Subsequently, stereoscopic optical cameras were used to acquire localized random undulations of the seabed rough interface. Laser scanning equipment has also been used to measure these small-scale interfaces, but due to significant intensity attenuation of laser light underwater, measurements can only be performed near the seabed. These measurement methods are limited to localized measurements and cannot meet the application requirements of underwater acoustics. Domestic research on the measurement of such small-scale perturbation-prone interfaces has not yet been conducted.

[0004] Centimeter-scale seabed perturbations are the primary scattering source of the sound field and the main cause of seabed reverberation. According to the theory of full wave reverberation, centimeter-scale rough seabed interfaces are represented by a roughness spectrum. The intensity of seabed reverberation can be expressed as the product of the spectral function of the rough seabed interface and the reverberant waveguide attenuation term, which are independent of each other. Seabed reverberation data acquisition methods are simple and fast, and the reverberation data can reflect regionalized seabed rough interface results. Therefore, broadband seabed reverberation data can be used to achieve rapid estimation of the spectral function of the rough seabed interface.

[0005] In reverberation theory, the rough interface of seabed perturbations is described as a spatially distributed random process that satisfies statistical properties. In the 1980s, Jagoff and T. Jordan summarized a roughness spectral function model for random undulations of seabed rough interfaces based on measured data at sea. This model is described by three parameters: variance, spatial wavenumber, and spatial frequency exponent. Variance describes the height of the random undulations from the mean; spatial wavenumber describes the spatial correlation characteristics of the random undulations; and spatial frequency exponent describes the spatial frequency characteristics of the spectral function. Summary of the Invention

[0006] The purpose of this invention is to rapidly estimate large-area rough interfaces on the seabed at the centimeter scale. This invention leverages the advantage of broadband seabed reverberation data to quickly acquire and reflect information about large-area seabed rough interfaces. It obtains the curve of the seabed rough interface spectral function as a function of frequency, and then inversely retrieves the statistical characteristic parameters describing the seabed rough interface, achieving rapid remote sensing of seabed rough interfaces and providing technical support for rapid marine environmental assessment.

[0007] To achieve the above objectives, the present invention is implemented through the following technical solution.

[0008] This invention proposes a remote sensing method for seabed rough interfaces based on broadband reverberation data, the method comprising:

[0009] Obtain the frequency response curve of the seabed reverberation intensity;

[0010] Estimate the frequency response curve of the ocean reverberation intensity waveguide attenuation term;

[0011] Based on the frequency response curves of seabed reverberation intensity and ocean reverberation intensity waveguide attenuation term, the spectral function of seabed rough interface is obtained.

[0012] Based on the obtained seabed roughness interface spectral function, the statistical characteristic parameters of the random undulations of the seabed roughness interface are obtained using the roughness spectral function model, thereby realizing remote sensing of the seabed roughness interface.

[0013] As one improvement to the above technical solution, the acquisition of the frequency response curve of the seabed reverberation intensity specifically includes:

[0014] Deploy sound sources and transmit broadband sound signals to the seabed through these sources;

[0015] Deploy hydrophones in the water and collect seabed reverberation signals using the hydrophones;

[0016] The acquired reverberation signal is subjected to narrowband filtering and short-time energy averaging to obtain the reverberation intensity at multiple frequencies, and then the frequency response curve of the reverberation intensity is obtained.

[0017] As an improvement to the above technical solution, the frequency response curve of the estimated ocean reverberation intensity waveguide attenuation term specifically includes: using the placement positions of the sound source and hydrophone, the sound source emission frequency, the ocean sound velocity profile, and seabed acoustic parameters, the frequency response curve of the ocean reverberation intensity waveguide attenuation term corresponding to the reverberation time is estimated through numerical simulation.

[0018] As an improvement to the above technical solution, the seabed rough interface spectral function is obtained by removing the source level and reverberation intensity waveguide attenuation terms from the broadband seabed reverberation intensity, and its expression is:

[0019] SP(f) = RL(t,f) - SL(f) - G(t,f)

[0020] Where SP(f) is the spectral function of the seabed rough interface, which varies with frequency f; RL(t,f) represents the reverberation intensity at frequency f at time t; SL(f) represents the source level value at frequency f; and G(t,f) represents the waveguide attenuation term of the reverberation intensity at frequency f at time t.

[0021] As an improvement to the above technical solution, the process of removing the source level and reverberation intensity waveguide attenuation term from the broadband seabed reverberation intensity to extract the seabed rough interface spectral function further includes: averaging the seabed rough interface spectral function extracted at each time point.

[0022] As one of the improvements to the above technical solution, the frequency exponent of the seabed rough interface spectral function is directly reflected by the slope of the decibel value of the seabed rough interface spectral function as a function of frequency.

[0023] As one improvement to the above technical solution, the expression for the roughness spectrum function model W(κ) is:

[0024] W(κ)=σ 2 Lπ(1+(κL) 2 ) -(1+v)

[0025] Where, σ 2 L and L represent the variance and correlation length of the random undulations of the seabed rough interface, respectively, v is the frequency exponent of the spectral function of the seabed rough interface, and κ represents the spatial frequency of the random undulations of the seabed rough interface.

[0026] As an improvement to the above technical solution, the step of obtaining statistical characteristic parameters of random undulations of the seabed rough interface using a roughness spectrum function model based on the acquired seabed rough interface spectrum function includes:

[0027] Estimate the frequency exponent of the spectral function of the seabed rough interface;

[0028] The variance and correlation length of random undulations of the seabed rough interface are obtained using the estimated frequency exponent and roughness spectral function model.

[0029] This invention also proposes a remote sensing system for seabed rough interfaces based on broadband reverberation data, the system comprising:

[0030] The reverberation intensity acquisition module is used to acquire the frequency response curve of the seabed reverberation intensity.

[0031] The reverberation intensity waveguide attenuation term acquisition module is used to estimate the frequency response curve of the ocean reverberation intensity waveguide attenuation term;

[0032] The module for acquiring the spectral function of the seabed roughness interface is used to acquire the spectral function of the seabed roughness interface; and

[0033] The statistical characteristic parameter acquisition module is used to obtain the statistical characteristic parameters of the random undulations of the seabed rough interface using the acquired seabed rough interface spectral function and roughness spectral function model, thereby realizing remote sensing of the seabed rough interface.

[0034] As an improvement to the above technical solution, the statistical characteristic parameter acquisition module includes:

[0035] The frequency index acquisition unit is used to estimate the frequency index of the spectral function of the seabed roughness interface; and

[0036] The relevant parameter acquisition unit is used to obtain the variance and correlation length of random undulations of the seabed rough interface using the estimated frequency exponent and roughness spectral function model.

[0037] The present invention provides a remote sensing technique for detecting random undulations of seabed rough interfaces at the centimeter level using marine acoustic data. The technique is characterized by the following features:

[0038] (1) The remote sensing accuracy of seabed random undulation interfaces is in the centimeter range; the highest accuracy of existing ocean depth sounders in measuring seabed depth is in the meter range, which is greatly constrained by the preset sound velocity and is often difficult to be accurate to the centimeter level. This invention targets the centimeter-level random undulations of the seabed and considers their impact on sound scattering and reverberation sound field. It proposes a remote sensing technology for centimeter-level rough seabed interfaces based on seabed reverberation data, which can provide the random undulation characteristics of the seabed interface in the centimeter range, providing technical support for optimizing marine environmental parameters and marine acoustic environment forecasting theory;

[0039] (2) Significantly reduces time and labor costs; compared to traditional hand-drawing and optical measurement techniques, the remote sensing method for seabed rough interfaces using seabed reverberation data proposed in this invention can significantly save time and labor costs; reverberation data is relatively easy to acquire, requiring only a single sound source and a single-point receiving hydrophone; secondly, reverberation data can quickly reflect the reverberation data of a large area of ​​seabed rough interfaces, for example, 4s of reverberation data can reflect the random undulation characteristics of the seabed rough interface within a 3km scale range. Therefore, the centimeter-scale seabed rough interface remote sensing technology proposed in this invention can significantly reduce time and labor costs. Attached Figure Description

[0040] Figure 1 A flowchart illustrating the workflow of remote sensing technology for rough seabed interfaces;

[0041] Figure 2 Design drawings for a marine implementation plan for remote sensing of rough seabed surfaces;

[0042] Figure 3 The sound source level of a 25m depth-1000g equivalent acoustic blast;

[0043] Figure 4 Ocean reverberation time series collected by hydrophone;

[0044] Figure 5 The frequency response curve of the reverberation intensity at the seabed;

[0045] Figure 6 The frequency response curve of the attenuation term of the seabed reverberant waveguide;

[0046] Figure 7 Estimation of the spectral function for seabed rough interfaces;

[0047] Figure 8 For estimation of spectral parameters of seabed rough interfaces;

[0048] Figure 9 To estimate the 250Hz reverberation average intensity attenuation curve using the seabed rough interface parameters obtained by the method of this invention; Detailed Implementation

[0049] The technical solutions provided by the present invention will be further illustrated below with reference to the embodiments.

[0050] Example 1

[0051] like Figure 1 The diagram shown is a flowchart of a remote sensing method for seabed rough interfaces based on broadband reverberation data according to the present invention, which specifically includes the following steps:

[0052] (1) Acquisition of broadband submarine reverberation data

[0053] Broadband submarine reverberation data acquisition solution, such as Figure 2 As shown. A single-point sound source emits a broadband sound signal, and the signal source stage is as follows. Figure 3 As shown. The hydrophone is located underwater and is used to collect reverberation signals, such as... Figure 4 As shown, the acquired reverberation signal is subjected to narrowband filtering and short-time energy averaging to obtain the reverberation intensity at multiple frequencies, such as... Figure 5 As shown in the diagram. (For simplicity, the diagram is omitted.) Figure 5 The frequency response curves of the reverberation intensity at 4s and 5s are given only.

[0054] (2) Estimation of reverberation intensity waveguide attenuation term

[0055] use Figure 2 The experimental setup shown records the placement of the sound source and receiving hydrophone, the sound source emission frequency, the ocean sound velocity profile, and seabed acoustic parameters. Numerical simulation is used to estimate the ocean reverberation intensity waveguide attenuation term corresponding to the reverberation time, such as... Figure 6As shown in the figure, the frequency response curves of the reverberation intensity waveguide attenuation term at 4s and 5s are presented.

[0056] (3) Inversion of spectral function of seabed rough interface

[0057] The seabed reverberation intensity at each frequency point was obtained by short-time smoothing and averaging the measured broadband reverberation time-domain waveform in content (1). Figure 5 As shown), combined with the sound source level at the corresponding frequency point (such as... Figure 3 As shown), and the reverberation intensity waveguide attenuation term estimated in content (2) (as shown). Figure 6 As shown, by removing the source level and reverberation intensity waveguide attenuation terms from the reverberation data, the spectral function of the seabed rough interface can be obtained, i.e.

[0058] SP(f) = RL(t,f) - SL(f) - G(t,f)

[0059] In this equation, SP(f) is the spectral function of the seabed rough interface, which varies with frequency f; RL(t,f) represents the reverberation intensity at frequency f at time t; SL(f) represents the source level at frequency f; and G(t,f) represents the waveguide attenuation term of the reverberation intensity at frequency f at time t. Only the reverberation intensity and its waveguide attenuation term change with reverberation time in this equation; the source level and the spectral function of the seabed rough interface are both functions of frequency and are independent of reverberation time. Figure 7 The spectral function of the seabed rough interface extracted from broadband reverberation data is presented. Considering that the spectral function of the seabed rough interface is independent of time, the extraction results at each time point are averaged to reduce errors in the actual measurement process. Figure 7 The “.” in the text represents the extraction results at each time point, and “□” represents the average result at each time point.

[0060] (4) Statistical parameters of seabed rough interface

[0061] Based on marine measurement data, JAGoff and TH Jordan summarized a roughness spectral function model for the random undulations of the seabed rough interface, which is derived from the variance σ of the random undulations of the seabed rough interface. 2 The relevant length L and frequency exponent term v describe, i.e.

[0062] W(κ)=σ 2 Lπ(1+(κL) 2 ) -(1+v)

[0063] Here, κ is the spatial wavenumber of the rough seabed interface, which corresponds one-to-one with the sound source frequency.

[0064] from Figure 7 As can be seen, the spectral function of the seabed rough interface exhibits a linear relationship with the frequency exponent. Figure 7The vertical axis represents the decibel value of the spectral intensity of the seabed rough interface, and the horizontal axis represents the frequency in logarithmic form with base 10. The slope of the approximately linear change in the figure directly reflects the frequency index as -3.8008, i.e., -2(1+v) = -3.8008, thus calculating v = 0.9004. Based on the known frequency index, using the Goff-Jordan spectral function as a foundation, the spectral function curve of the seabed rough interface is matched to estimate the correlation length L = 11.556 m and variance σ of the seabed rough interface. 2 =0.325m 2 Thus, the statistical characteristics of the seabed rough interface are obtained, such as... Figure 8 As shown.

[0065] according to Figure 2 The experimental scheme shown was validated at sea. The experimental area was at a water depth of 31 m. The sound velocity profile was approximately at an isosonic level of 1488 m / s. The seabed acoustic parameters were a sound velocity of 1585 m / s and a density of 1.8 g / cm³. 3 The sound absorption coefficient is 0.3 dB / λ. The broadband sound source uses a 25m fixed-depth, 1000g equivalent TNT acoustic bomb, with a sound source level as follows: Figure 3 As shown. The receiving hydrophone was located 20m underwater, and the acquired reverberation time series is as follows. Figure 4 As shown. Narrowband filtering and short-time smoothing averaging are applied to the time-domain waveform to calculate the incoherent reverberation intensity, yielding the frequency response of the reverberation intensity as a function of frequency, as shown below. Figure 5 As shown. For ease of representation. Figure 5 The paper only provides the reverberation intensity frequency response curves at 4s and 5s. Based on the sound velocity profile, seabed acoustic parameters, and relative transmitter-receiver positions during the experiment, the frequency response of the seabed reverberation intensity waveguide attenuation term at the corresponding times is estimated, such as... Figure 6 As shown. From Figure 4 The source level and reverberation intensity waveguide attenuation terms are removed from the reverberation intensity data shown. The corresponding spectral intensity of the seabed rough interface at each time point is extracted, and the average over time is taken to obtain the frequency response curve of the seabed rough interface spectral function, as shown below. Figure 7 As indicated by the "□" in the diagram. Based on the Goff-Jordan spectral function model, firstly, according to the characteristic that the decibel value of the spectral function of the seabed rough interface changes linearly with the logarithm of frequency, the frequency exponent of the spectral function is estimated, yielding the spatial frequency exponent term v = 0.9004 for the seabed rough interface; secondly, based on this parameter determination, the correlation length L = 11.556 m and variance σ of the random undulations of the seabed rough interface are estimated. 2 =0.325m 2 Based on the inversion results, the attenuation curves of lower-frequency seabed reverberation intensity are predicted, such as... Figure 9 As shown, the forecast results are basically consistent with the measured results, indicating that the remote sensing results of the rough seabed interface at the centimeter level are accurate and reliable.

[0066] Example 2

[0067] This invention discloses a remote sensing system for seabed rough interfaces based on broadband reverberation data, comprising: a reverberation intensity acquisition module, a reverberation intensity waveguide attenuation term acquisition module, a seabed rough interface spectral function acquisition module, and a statistical characteristic parameter acquisition module. Among these,

[0068] The reverberation intensity acquisition module is used to acquire the frequency response curve of the seabed reverberation intensity.

[0069] The reverberation intensity waveguide attenuation term acquisition module is used to estimate the frequency response curve of the ocean reverberation intensity waveguide attenuation term;

[0070] The module for obtaining the spectral function of the seabed rough interface is used to obtain the spectral function of the seabed rough interface.

[0071] The statistical characteristic parameter acquisition module is used to obtain the statistical characteristic parameters of the random undulations of the seabed rough interface using the acquired seabed rough interface spectral function and roughness spectral function model, thereby realizing remote sensing of the seabed rough interface.

[0072] The statistical characteristic parameter acquisition module includes: a frequency index acquisition unit and a related parameter acquisition unit; wherein,

[0073] Frequency index acquisition unit, used to estimate the frequency index of the spectral function of the seabed rough interface;

[0074] The relevant parameter acquisition unit is used to obtain the variance and correlation length of random undulations of the seabed rough interface using the estimated frequency exponent and roughness spectral function model.

[0075] As can be seen from the above detailed description of the present invention, the present invention extracts the frequency characteristics of the roughness spectral function describing minute undulations at the centimeter scale from broadband seabed reverberation data, and further estimates the statistical characteristic parameters describing the rough interface based on these frequency characteristics. Simultaneously, the present invention leverages the advantage of rapid acquisition of broadband seabed reverberation data and the characteristic that reverberation data can reflect seabed rough interface information over a large scale, achieving the goal of rapid remote sensing of seabed rough interfaces at the centimeter scale, and providing technical support for rapid assessment of the marine environment.

[0076] 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 it. Although the present invention has been described in detail with reference to the embodiments, those skilled in the art should understand that modifications or equivalent substitutions to the technical solutions of the present invention do not depart from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A remote sensing method for seabed rough interfaces based on broadband reverberation data, the method comprising: Obtain the frequency response curve of the seabed reverberation intensity; Estimate the frequency response curve of the ocean reverberation intensity waveguide attenuation term; Based on the frequency response curves of seabed reverberation intensity and ocean reverberation intensity waveguide attenuation term, the spectral function of the seabed rough interface is obtained; the expression of the seabed rough interface spectral function is as follows: in, It is the spectral function of the seabed rough interface, which varies with frequency. f change; express Time, frequency The reverberation intensity; Representing frequency The sound source level value; express Time, frequency The reverberation intensity waveguide attenuation term; Estimate the frequency exponent of the spectral function of the seabed rough interface; use the estimated frequency exponent and roughness spectral function model to obtain the variance and correlation length of the random undulations of the seabed rough interface to achieve remote sensing of the seabed rough interface; The roughness spectrum function model The expression is: in, and Let V be the variance and correlation length of the random undulations of the seabed rough interface, respectively. The frequency exponent of the spectral function of the seabed rough interface. It represents the spatial frequency of random undulations on the rough seabed surface.

2. The remote sensing method for seabed rough interfaces based on broadband reverberation data according to claim 1, characterized in that, The frequency response curve for obtaining the seabed reverberation intensity specifically includes: Deploy sound sources and transmit broadband sound signals to the seabed through these sources; Deploy hydrophones in the water and collect seabed reverberation signals using the hydrophones; The acquired reverberation signal is subjected to narrowband filtering and short-time energy averaging to obtain the reverberation intensity at multiple frequencies, and then the frequency response curve of the reverberation intensity is obtained.

3. The remote sensing method for seabed rough interfaces based on broadband reverberation data according to claim 1, characterized in that, The frequency response curve of the estimated ocean reverberation intensity waveguide attenuation term specifically includes: using the placement positions of the sound source and hydrophone, the sound source emission frequency, the ocean sound velocity profile, and seabed acoustic parameters, the frequency response curve of the ocean reverberation intensity waveguide attenuation term corresponding to the reverberation time is estimated through numerical simulation.

4. The remote sensing method for seabed rough interfaces based on broadband reverberation data according to claim 1, characterized in that, The process of extracting the seabed rough interface spectral function by removing the source level and reverberation intensity waveguide attenuation terms from the broadband seabed reverberation intensity also includes averaging the extracted seabed rough interface spectral function at each time point.

5. The remote sensing method for seabed rough interfaces based on broadband reverberation data according to claim 1, characterized in that, The frequency exponent of the seabed rough interface spectral function is directly reflected by the slope of the decibel value of the seabed rough interface spectral function as a function of frequency.

6. A system based on the remote sensing method for seabed rough interfaces based on broadband reverberation data as described in claim 1, characterized in that, The system includes: The reverberation intensity acquisition module is used to acquire the frequency response curve of the seabed reverberation intensity. The reverberation intensity waveguide attenuation term acquisition module is used to estimate the frequency response curve of the ocean reverberation intensity waveguide attenuation term; The module for acquiring the spectral function of the seabed roughness interface is used to acquire the spectral function of the seabed roughness interface; and The statistical characteristic parameter acquisition module is used to obtain the statistical characteristic parameters of the random undulations of the seabed rough interface using the acquired seabed rough interface spectral function and roughness spectral function model, thereby realizing remote sensing of the seabed rough interface.

7. The remote sensing system for seabed rough interfaces based on broadband reverberation data according to claim 6, characterized in that, The statistical characteristic parameter acquisition module includes: The frequency index acquisition unit is used to estimate the frequency index of the spectral function of the seabed roughness interface; and The relevant parameter acquisition unit is used to obtain the variance and correlation length of random undulations of the seabed rough interface using the estimated frequency exponent and roughness spectral function model.

Images