A Method for Modal Waveform Separation of Single Hydrophone Signals Based on Modal Frequency Modulation Wavelet Transform

Abstract

This invention discloses a method for separating the modal waveforms of a single hydrophone signal based on modal frequency modulation wavelet transform. The method includes: acquiring a hydrophone signal; processing the hydrophone signal to obtain the time-domain signal of the sound source; using modal frequency modulation wavelet transform to obtain the time-frequency diagram of the sound source signal; selecting a region containing a desired mode in the time-frequency diagram and performing time-frequency domain filtering to obtain the time-frequency diagram of the selected mode; and using inverse modal frequency modulation wavelet transform to obtain the time-domain waveform of the selected mode. This method not only robustly separates the time-domain waveforms of each mode contained in the received signal from a single hydrophone, facilitating further parameter estimation, but also is insensitive to the selection of the mode range during time-frequency domain filtering, and is significantly less affected by changes in the time-frequency domain filtering range, making it more likely to obtain accurate modal waveforms and improving the accuracy of subsequent parameter estimation.

Description

Technical Field

[0001] This invention relates to the field of sound signal processing technology, and more specifically to a method for separating the modal waveforms of a single hydrophone signal based on modal frequency modulation wavelet transform. Background Technology

[0002] Single-hydrophone (sound transducer used underwater) signal processing technology has received considerable attention in recent years. With the increasing volume and complexity of human marine activities, the value of constructing a large number of acoustic observation base stations within the ocean is becoming increasingly apparent. Single-hydrophones offer significant cost advantages and are an ideal base station form for achieving multi-node, large-scale observation. Furthermore, their small size and relatively small data processing requirements make them well-suited for deployment on small underwater vehicles, expanding their application scenarios and increasing their value. This also complements existing fixed-point and drifting ocean observation methods. Therefore, single-hydrophone signal processing technology has developed rapidly in recent years, and its capabilities in target ranging, depth determination, direction finding (vector hydrophones), and ground acoustic parameter inversion are gradually gaining recognition.

[0003] Single-hydrophone signal processing technology must consider algorithm robustness. Seawater is a dispersive medium; when sound waves propagate in seawater, different modes and frequencies of sound waves propagate at different group velocities. Therefore, a simple signal will be distorted into a complex signal composed of multiple superimposed modes during propagation. The complexity of the sound propagation environment, such as seabed topography changes and internal waves, will exacerbate signal distortion, making it difficult to recover the original signal or information from the received sound signal. Developing environmentally tolerant signal processing methods has become an important goal in sound source localization or ground acoustic parameter inversion research.

[0004] Signal processing in the modal domain is a typical approach to improve algorithm robustness. Compared to directly using received sound pressure for parameter estimation, signal processing in the modal domain offers greater flexibility. It allows for the free selection and combination of modes, and by eliminating modal components heavily influenced by the environment, more robust parameter estimation can be achieved. Given the flexibility and robustness of signal processing in the modal space, mode separation has become an important research direction, representing the first step in modal domain signal processing.

[0005] Early modal separation mainly used conventional time-frequency analysis techniques, such as short-time Fourier transform or wavelet transform (POTTY GR, MILLER JH, LYNCH JF, et al. Tomographic inversion for sediment parameters in shallow water[J].The Journal of the Acoustical Society of America,2000,108(3):973-986.). Due to their low time-frequency resolution, they could only be used when the distance between the sound source and the receiver was large. Later, Le Touzé et al. introduced warping transform into modal separation, which has become the mainstream technique at present (LE). G, MARS J, LACOUME J L. Matched time-frequency representations and warping operator for modal filtering [C] / / 2006 14th European Signal Processing Conference. 2006: 1-5.). The role of warping transformation technology is only to change the step of separating modal waveforms directly in the time-frequency domain of the received signal to be performed in the warping domain of the received signal. When separating modal waveforms, conventional time-frequency analysis algorithms with low time-frequency resolution are still used (BONNEL J, THODE A, WRIGHT D, et al. Nonlinear time-warping made simple: A step-by-step tutorial on underwater acoustic modal separation with a single hydrophone [J]. The Journal of the Acoustical Society of America, 2020, 147(3): 1897-1926.). Although it improves the performance of separating modal waveforms by directly using conventional time-frequency analysis algorithms to a certain extent, it cannot completely avoid the influence of low time-frequency resolution. Summary of the Invention

[0006] The purpose of this invention is to overcome the shortcomings of the prior art and provide a method for separating the modal waveforms of a single hydrophone signal based on modal frequency modulation wavelet transform.

[0007] To achieve the above objectives, the present invention adopts the following technical solution:

[0008] The first aspect is a method for separating the modal waveforms of a single hydrophone signal based on modal frequency modulation wavelet transform, including:

[0009] Collect hydrophone signals;

[0010] The hydrophone signal is processed to obtain the time-domain signal of the sound source;

[0011] The time-frequency diagram of the sound source signal is obtained by using modal frequency modulation wavelet transform to obtain the time-frequency diagram of the sound source signal;

[0012] Select the region containing the desired mode in the time-frequency plot and perform time-frequency domain filtering to obtain the time-frequency plot of the selected mode;

[0013] The time-frequency diagram of the selected mode is used to obtain the time-domain waveform of that mode by inverse mode frequency modulation wavelet transform.

[0014] The further technical solution is as follows: the processing of the hydrophone signal to obtain the time-domain signal of the sound source includes:

[0015] The signal collected by the hydrophone is transformed from the time domain to the frequency domain by Fourier transform in order to obtain the measured sound pressure of the sound source.

[0016] The Green's function of the measurement field is obtained by removing the spectrum of the transmitted signal from the measured sound pressure of the sound source through deconvolution.

[0017] The Green's function of the measurement field is transformed from the frequency domain back to the time domain by Fourier transform to obtain the time domain signal of the sound source.

[0018] The further technical solution is as follows: obtaining the time-frequency diagram of the sound source signal by using modal frequency modulation wavelet transform on the time-domain signal of the sound source includes:

[0019] Wavelet decomposition is performed on the time-domain signal of the sound source using wavelet basis functions to obtain wavelet coefficients;

[0020] Modal decomposition is performed on the wavelet coefficients to obtain each modal component;

[0021] Perform frequency-modulated wavelet transform on each modal component to obtain the time-frequency diagram of each modal component;

[0022] The time-frequency diagrams of each modal component are superimposed to obtain the time-frequency diagram of the sound source signal.

[0023] The further technical solution is as follows: Selecting the region containing the desired mode in the time-frequency diagram and performing time-frequency domain filtering to obtain the time-frequency diagram of the selected mode includes:

[0024] Select the region containing the desired mode and record the time and frequency range of that region;

[0025] Use a filter to perform time-frequency domain filtering on this region;

[0026] Extract the filtered region to obtain the time-frequency plot of the selected mode.

[0027] Secondly, a single hydrophone signal mode waveform separation device based on modal frequency modulation wavelet transform includes an acquisition unit, a signal processing unit, a first transformation unit, a filtering unit, and a second transformation unit;

[0028] The acquisition unit is used to acquire hydrophone signals;

[0029] The signal processing unit is used to process the hydrophone signal to obtain the time-domain signal of the sound source;

[0030] The first transformation unit is used to obtain the time-frequency diagram of the sound source signal by using modal frequency modulation wavelet transform;

[0031] The filtering unit is used to select the region containing a desired mode in the time-frequency diagram and perform time-frequency domain filtering to obtain the time-frequency diagram of the selected mode.

[0032] The second transformation unit is used to obtain the time-domain waveform of the selected mode by using inverse mode frequency modulation wavelet transform.

[0033] The further technical solution is as follows: the signal processing unit includes a first transformation module, a filtering module, and a second transformation module;

[0034] The first transformation module is used to transform the signal collected by the hydrophone from the time domain to the frequency domain through Fourier transform in order to obtain the measured sound pressure of the sound source.

[0035] The filtering module is used to remove the transmitted signal spectrum from the measured sound pressure of the sound source by deconvolution in order to obtain the Green function of the measurement field.

[0036] The second transformation module is used to transform the Green's function of the measurement field from the frequency domain back to the time domain through Fourier transform, so as to obtain the time domain signal of the sound source.

[0037] The further technical solution is as follows: the first conversion unit includes a first decomposition module, a second decomposition module, a frequency modulation conversion module, and a superposition module;

[0038] The first decomposition module is used to perform wavelet decomposition on the time-domain signal of the sound source using wavelet basis functions to obtain wavelet coefficients;

[0039] The second decomposition module is used to perform modal decomposition on the wavelet coefficients to obtain each modal component;

[0040] The frequency modulation transform module is used to perform frequency modulation wavelet transform on each modal component to obtain the time-frequency diagram of each modal component;

[0041] The superposition module is used to superimpose the time-frequency diagrams of each modal component to obtain the time-frequency diagram of the sound source signal.

[0042] The further technical solution is as follows: the filtering unit includes a selection module, a filtering module, and an extraction module;

[0043] The selection module is used to select the region where a certain mode is located and record the time range and frequency range of the region.

[0044] The filtering module is used to perform time-frequency domain filtering on the region using a filter;

[0045] The extraction module is used to extract the filtered region to obtain the time-frequency diagram of the selected mode.

[0046] Thirdly, a computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the single hydrophone signal mode waveform separation method based on modal frequency modulation wavelet transform as described above.

[0047] Fourthly, a computer-readable storage medium stores a computer program, the computer program including program instructions that, when executed by a processor, cause the processor to perform the single hydrophone signal mode waveform separation method based on modal frequency modulation wavelet transform as described above.

[0048] The beneficial effects of this invention compared to existing technologies are as follows: This invention acquires hydrophone signals; processes the hydrophone signals to obtain the time-domain signal of the sound source; uses modal frequency modulation wavelet transform to obtain the time-frequency diagram of the sound source signal; selects the region containing a desired mode in the time-frequency diagram and performs time-frequency domain filtering to obtain the time-frequency diagram of the selected mode; uses inverse modal frequency modulation wavelet transform to obtain the time-frequency waveform of the selected mode. This not only robustly separates the time-domain waveforms of each mode contained in the single hydrophone received signal, facilitating further parameter estimation (sound source localization or ground acoustic parameter inversion), but also is insensitive to the selection of the mode range during time-frequency domain filtering, is significantly less affected by changes in the time-frequency domain filtering range, and is more likely to obtain accurate mode waveforms, thus improving the accuracy of subsequent parameter estimation.

[0049] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention, it can be implemented according to the contents of the specification. In order to make the above and other objectives, features and advantages of the present invention more obvious and understandable, preferred embodiments are described in detail below. Attached Figure Description

[0050] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0051] Figure 1 A flowchart of a single hydrophone signal modal waveform separation method based on modal frequency modulation wavelet transform provided for a specific embodiment of the present invention;

[0052] Figure 2 A schematic block diagram of a single hydrophone signal mode waveform separation device based on modal frequency modulation wavelet transform provided for a specific embodiment of the present invention;

[0053] Figure 3 A schematic block diagram of a computer device provided for a specific embodiment of the present invention;

[0054] Figure 4 A map showing seabed topographic changes provided for a specific embodiment of the present invention;

[0055] Figure 5 The explosion signal waveform, received signal waveform, and deconvolutioned received signal waveform diagrams are provided for specific embodiments of the present invention.

[0056] Figure 6 The short-time Fourier time-frequency diagrams of the received signal (left figure) and the deconvolutioned received signal (right figure) provided for a specific embodiment of the present invention;

[0057] Figure 7 The left image shows the resolution effect of MCT on the fourth mode, and the right image shows the resolution effect of EMCT on each mode, provided for specific embodiments of the present invention.

[0058] Figure 8 This is a schematic diagram illustrating the selection of a region containing a certain mode in a time-frequency graph and the subsequent time-frequency domain filtering, as provided in a specific embodiment of the present invention.

[0059] Figure 9 A schematic diagram of four cutoff frequency ranges provided for specific embodiments of the present invention;

[0060] Figure 10The waveform diagram of the fourth mode of the explosion signal obtained by separation is provided for a specific embodiment of the present invention. Detailed Implementation

[0061] The technical solutions of 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 some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0062] It should be understood that, when used in this specification and the appended claims, the terms "comprising" and "including" indicate the presence of the described features, integrals, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0063] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0064] It should also be further understood that the term "and / or" as used in this specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0065] This invention provides a method for separating the modal waveforms of a single hydrophone signal based on modal frequency modulation wavelet transform. The modal waveform features separated by this method can be used to estimate sound field parameters such as seabed density, seabed sound velocity, and sound source distance, which can then be used for sound source localization or ground acoustic parameter inversion studies.

[0066] like Figure 1 As shown, a method for separating the modal waveform of a single hydrophone signal based on modal frequency modulation wavelet transform includes the following steps: S10-S50.

[0067] S10, Collect hydrophone signals.

[0068] The sound signal from the underwater sound source target is received by a single hydrophone and converted into an electrical signal. After the electrical signal is filtered and amplified, it is recorded by a data acquisition instrument.

[0069] In this embodiment, the marine test site is located near (W70.524°, N40.499°), such as... Figure 4As shown, the seabed topography here is sloping. In the experiment, the receiving hydrophone was located 0.65 meters above the seabed, the sound source depth was approximately 12.5 meters, and the distance between the sound source and the receiving hydrophone was approximately 4.8 km. The sound speed profile in the experimental area changed linearly, with the sound speed at the sea surface being 1464 m / s and the sound speed at the seabed being 1465 m / s. Figure 5 The figures shown are the waveforms of the explosion signal, the received signal, and the received signal after deconvolution.

[0070] S20. Process the hydrophone signal to obtain the time-domain signal of the sound source.

[0071] In one embodiment, step S20 specifically includes the following steps: S201-S203.

[0072] S201. The signal collected by the hydrophone is transformed from the time domain to the frequency domain through Fourier transform in order to obtain the measured sound pressure of the sound source.

[0073] By transforming the signal collected by the hydrophone from the time domain to the frequency domain using Fourier transform, the sound pressure distribution of the sound source at different frequencies can be obtained, thereby achieving the purpose of measuring the sound pressure of the sound source.

[0074] In this embodiment, the received time-domain signal is transformed to the frequency domain using Fourier transform to obtain the measured sound pressure p at frequency f. mea (f); Ignoring noise, the measured sound pressure in an ocean waveguide can be written as:

[0075] p mea (f)=s(f)g mea (f), where s(f) represents the transmitted signal spectrum; g mea (f) represents the measured Green's function of the ocean waveguide.

[0076] S202. The transmitted signal spectrum is removed from the measured sound pressure of the sound source by deconvolution to obtain the Green function of the measurement field.

[0077] The purpose of deconvolution to obtain the Green's function of the measurement field is to deduce the spatial distribution of the sound source. Specifically, the Green's function of the measurement field obtained by deconvolution can be used to reconstruct the spatial sound field distribution image of the sound source.

[0078] In this embodiment, since the sound source location information is only contained in g mea In (f), the measured Green's function g of the ocean waveguide can be obtained. mea (f) as the measurement field. From p through deconvolution... mea (f) Remove the transmitted signal spectrum from the measured sound pressure level to obtain g mea The formula for (f) is as follows:

[0079] In the formula, gmea (f) denotes the Green's function of the measurement field, * denotes conjugate, |·| denotes taking the absolute value, and ò denotes a very small number, usually 0.01max|s(f)|, to prevent the denominator from being zero or close to zero, which would cause the Green's function g of the measurement field to be affected. mea (f) Abnormal results.

[0080] S203. The Green's function of the measurement field is transformed from the frequency domain back to the time domain by Fourier transform to obtain the time domain signal of the sound source.

[0081] Transforming the field back to the time domain using a Fourier transform essentially converts the field's Green's function from the frequency domain to the time domain. In the frequency domain, the Green's function is a complex number representing the response at different frequencies. In the time domain, however, the Green's function represents the response over a given time period. Therefore, converting the field's Green's function from the frequency domain to the time domain provides a more intuitive understanding of the field's response, including its duration and waveform.

[0082] In this embodiment, g is obtained mea After (f), it is transformed back to the time domain by Fourier transform and denoted as y(t).

[0083] In this embodiment, for steps S201-S203 as a whole, the explosion sound is generally regarded as an impulse signal, but as Figure 5 As shown, the actual explosion sound signal contains several bubble pulsations, each of which is equivalent to an impulse signal. The superposition of multiple impulse signals that are closely spaced in time makes the waveform of the received signal quite complex, which can interfere with the identification of various modes in the time-frequency domain. If the explosion waveform is known, the influence of bubble pulsations can be eliminated by deconvolution. Figure 5 The waveform of the explosion signal is in the time domain form of s(f), and the received signal waveform is p. mea The time-domain form of (f) is the waveform of the received signal after deconvolution, which is g. mea The time-domain form of (f) is y(t). Figure 6 As shown, the time-frequency diagram of the received signal after deconvolution exhibits a very regular pattern, relative to the arrival time t. r0 The signal is clearly discernible at approximately 0.055 seconds. The actual arrival time of the signal is approximately 3.2 seconds. The calculations below intentionally deviate from the true value of t0 to illustrate that this parameter has a relatively small impact on the time-frequency spectrum. However, in practical applications, if t0 is unknown, it is recommended to take a larger value; in this embodiment, t0 = 20 seconds. Taking the 4th mode as an example to demonstrate the resolution effect of MCT, a point (t, f) near the 4th mode is randomly selected in the time-frequency spectrum. According to the formula, the cutoff frequency is approximately f. cm =15Hz.

[0084] S30. Obtain the time-frequency diagram of the sound source signal by using modal frequency modulation wavelet transform.

[0085] The main purpose of obtaining the time-frequency diagram of the sound source signal by using modal frequency modulation wavelet transform is to provide a more intuitive understanding of the distribution of the sound source signal in time and frequency.

[0086] In one embodiment, step S30 specifically includes the following steps: S301-S304.

[0087] S301. Wavelet decomposition is performed on the time-domain signal of the sound source using wavelet basis functions to obtain wavelet coefficients.

[0088] S302. Perform mode decomposition on the wavelet coefficients to obtain each mode component.

[0089] S303. Perform frequency-modulated wavelet transform on each modal component to obtain the time-frequency diagram of each modal component.

[0090] S304. The time-frequency diagrams of each modal component are superimposed to obtain the time-frequency diagram of the sound source signal.

[0091] In this embodiment, for steps S301-S304, the modal frequency modulation wavelet transform is defined as follows:

[0092] In the formula, y represents the hydrophone received signal, w is the window function, and the rotation factor and frequency shift factor are respectively:

[0093] Where, Φ m and f m The phase and instantaneous frequency of the mode under ideal waveguide conditions are represented as follows:

[0094]

[0095] In the formula f cm t represents the cutoff frequency. r =r / c represents the time required for sound to travel a distance r in the signal water at the speed of sound c, with the moment the sound source emitted as a reference, t r This is the absolute arrival time of the signal from the receiver to the receiving point. However, in practical applications, since the time of sound emission from the sound source is unknown, the received signal only contains the relative arrival time t. r0 , and t r There exists an unknown time delay t0, i.e., t r =t r0 The original modal frequency modulated wavelet transform shown in equation +t0 requires manually specifying the cutoff frequency f of that mode. cm Since different modes have different cutoff frequencies, only a single mode can be processed at a time. In practical applications, the following extended form is recommended:

[0096] That is, the frequency range is divided into several frequency bands, and each frequency band interval f c +[-Δ f ,Δ f The (t,f) elements within the range share the same cutoff frequency f. cm (This cutoff frequency is obtained by solving equation f) m (t)=f c (Automatically specified), f c Represents the center frequency, 2Δ f This represents the bandwidth; the larger the bandwidth, the greater the 2Δ bandwidth. f The larger the value, the higher the computational efficiency; correspondingly, the difference between the calculated time-frequency diagram and the actual time-frequency diagram will gradually increase.

[0097] Implementing extended mode frequency modulated wavelet transform requires specifying two parameters, namely the relative arrival time t. r0 And the unknown delay t0, the specified rules are as follows:

[0098] t r0 The relative arrival time is one of the most important parameters. For time-domain waveforms, since the modal signal has a time range of t... r0 The waveform propagates to the receiving point at time t, and the amplitude of the received waveform will be at time t. r0 It increases rapidly after time t; for time-frequency graphs, the dispersion curve in the higher frequency range should closely follow t. r0 It occurs after a certain time. Generally, t can be determined based on the two characteristics mentioned above. r0 time.

[0099] t0: The unknown time delay has little impact on the MCT time-frequency diagram. Any possible and reasonable value can be used as an estimate of t0 in the experiment.

[0100] Figure 7 The left-middle figure shows the time-frequency plot of MCT. The fourth-order mode fringes are clearly visible, but the fringes of other modes cannot be displayed correctly. That is, MCT can only give the fringes of a single mode each time. Figure 7 The right-middle figure shows the time-frequency plot given by EMCT. It can be seen that the fringes of all modes are clearly distinguishable, and... Figure 6 Compared to the image on the right, the method presented in this paper produces significantly thinner stripes, resulting in a substantial improvement in resolution (stripe spacing).

[0101] S40. Select the region containing the desired mode in the time-frequency diagram and perform time-frequency domain filtering to obtain the time-frequency diagram of the selected mode.

[0102] Time-frequency domain filtering refers to selecting a region in a time-frequency graph, retaining the signal within the region, and setting the signal outside the region to zero, which is equivalent to weighting different regions in the time-frequency graph.

[0103] In one embodiment, step S40 specifically includes the following steps: S401-S403.

[0104] S401. Select the region containing the desired mode and record the time range and frequency range of that region.

[0105] S402. Use a filter to perform time-frequency domain filtering on this region.

[0106] S403. Extract the filtered region to obtain the time-frequency diagram of the selected mode.

[0107] In this embodiment, for steps S401-S403, as follows: Figure 8 As shown, during time-frequency domain filtering, a weighting matrix MASK(t,f) is pre-generated to weight each pixel in the time-frequency graph, retaining the target mode and setting the pixel values ​​of other modes to zero.

[0108] S50. Use inverse mode frequency modulation wavelet transform to obtain the time domain waveform of the selected mode from the time-frequency diagram of the selected mode.

[0109] In this embodiment, the inverse mode frequency modulated wavelet transform is defined as:

[0110] In the formula y IMCT (τ) is the reconstructed time-domain waveform. MASK(t,f) is the masking value of the time-frequency domain filter, which is generally between 0 and 1. For convenience, it is usually binarized to 0 or 1. Taking 0 means completely suppressing the spectral value at (t,f), and taking 1 means retaining the spectral value at (t,f). a represents the weighting, which can be any positive integer. It is recommended to take a = 1, that is, to reconstruct the waveform in the least squares sense.

[0111] Since the extended modal wavelet transform was used to calculate the time-frequency plot of the signal, the inverse extended modal wavelet transform should be used to reconstruct the time-domain waveform of the mode, specifically defined as:

[0112] In the formula, the positive integer N represents the total number of frequency bands. When all frequency points are divided into one bandwidth, i.e., N = 1, y IEMCT (τ) degenerates into y IMCT (τ).

[0113] In practical applications, modal filtering requires manually selecting the region containing the target mode. For the same target mode, different people may produce different segmentation results, with some segmenting the region too large and others too small. For example... Figure 9As shown, based on the cutoff frequency range, the white dotted line, white solid line, and white dashed line enclose three regions of different sizes, respectively, to examine the influence of the size of the divided regions on the modal waveform separation results. For comparison, in addition to the MCT and EMCT methods, time-frequency plots for two other existing methods are also presented here. Figure 9 Figure a in the middle and Figure 9 Figure b in the diagram shows that the inverse transforms of these two methods (denoted as ISTFT and IWT, respectively) can also be used for modal waveform separation.

[0114] The waveforms of the fourth-order mode obtained by the four methods are shown below. Figure 10 As shown, it is easy to see that within the theoretically optimal time-frequency domain filtering range [12.5, 16.5] Hz (shown by the white solid line), the waveforms separated by the four methods are generally similar in structure and amplitude. When the time-frequency domain filtering range shrinks, i.e., within the range of [13.5, 15.5] Hz (shown by the white dashed line), the waveforms separated by ISTFT and IWT show a significantly smaller amplitude compared to those within the range of [12.5, 16.5] Hz. Conversely, when the time-frequency domain filtering range expands, i.e., within the range of [11.5, 17.5] Hz (shown by the white dotted line), the waveforms separated by ISTFT and IWT show a significantly larger amplitude compared to those within the range of [12.5, 16.5] Hz. In contrast, when the time-frequency domain filtering range shrinks and expands, the waveforms separated by IMCT and IEMCT show some changes compared to the optimal time-frequency domain filtering range of [12.5, 16.5] Hz, but the waveform amplitude remains essentially the same. Figure 10 This indicates that using IMCT and IEMCT to separate modal waveforms is more robust than using ISTFT and IWT, significantly less affected by changes in the time-frequency domain filtering range, and more likely to yield accurate modal waveforms.

[0115] It should be noted that if there are multi-mode waveforms that need to be separated, steps S30 to S50 can be repeated.

[0116] This invention not only robustly separates the time-domain waveforms of each mode contained in the received signal of a single hydrophone, facilitating further parameter estimation (sound source localization or ground acoustic parameter inversion), but also is insensitive to the selection of mode range during time-frequency domain filtering, is significantly less affected by changes in the time-frequency domain filtering range, and is more likely to obtain accurate mode waveforms, which is beneficial to improving the accuracy of subsequent parameter estimation.

[0117] Figure 2This is a schematic block diagram of a single hydrophone signal modal waveform separation device based on modal frequency modulation wavelet transform provided in an embodiment of the present invention; corresponding to the above-mentioned single hydrophone signal modal waveform separation method based on modal frequency modulation wavelet transform, this embodiment of the present invention also provides a single hydrophone signal modal waveform separation device 100 based on modal frequency modulation wavelet transform.

[0118] like Figure 2 As shown, a single hydrophone signal mode waveform separation device 100 based on modal frequency modulation wavelet transform includes an acquisition unit 110, a signal processing unit 120, a first transformation unit 130, a filtering unit 140, and a second transformation unit 150. The acquisition unit 110 is used to acquire hydrophone signals. The signal processing unit 120 is used to process the hydrophone signals to obtain the time-domain signal of the sound source. The first transformation unit 130 is used to use modal frequency modulation wavelet transform to obtain the time-frequency diagram of the sound source signal. The filtering unit 140 is used to select the region containing a desired mode in the time-frequency diagram and perform time-frequency domain filtering to obtain the time-frequency diagram of the selected mode. The second transformation unit 150 is used to use inverse modal frequency modulation wavelet transform to obtain the time-domain waveform of the selected mode.

[0119] In one embodiment, the signal processing unit 120 includes a first transformation module, a filtering module, and a second transformation module. The first transformation module is used to transform the signal acquired by the hydrophone from the time domain to the frequency domain using Fourier transform to obtain the measured sound pressure level of the sound source. The filtering module is used to remove the transmitted signal spectrum from the measured sound pressure level of the sound source by deconvolution to obtain the Green's function of the measurement field. The second transformation module is used to transform the Green's function of the measurement field from the frequency domain back to the time domain using Fourier transform to obtain the time-domain signal of the sound source.

[0120] In one embodiment, the first transformation unit 130 includes a first decomposition module, a second decomposition module, a frequency modulation transformation module, and a superposition module. The first decomposition module is used to perform wavelet decomposition on the time-domain signal of the sound source using wavelet basis functions to obtain wavelet coefficients. The second decomposition module is used to perform modal decomposition on the wavelet coefficients to obtain each modal component. The frequency modulation transformation module is used to perform frequency-modulated wavelet transform on each modal component to obtain the time-frequency diagram of each modal component. The superposition module is used to superimpose the time-frequency diagrams of each modal component to obtain the time-frequency diagram of the sound source signal.

[0121] In one embodiment, the filtering unit 140 includes a selection module, a filtering module, and an extraction module. The selection module is used to select the region containing a desired mode and record the time and frequency range of that region. The filtering module is used to perform time-frequency domain filtering on that region using a filter. The extraction module is used to extract the filtered region to obtain the time-frequency diagram of the selected mode.

[0122] The aforementioned method for separating the modal waveforms of a single hydrophone signal based on modal frequency modulation wavelet transform can be implemented as a computer program, which can be used in applications such as... Figure 3 It runs on the computer device shown.

[0123] Please see Figure 3 , Figure 3 This is a schematic block diagram of a computer device provided in an embodiment of this application. The computer device 700 can be a server, wherein the server can be a standalone server or a server cluster composed of multiple servers.

[0124] like Figure 3 As shown, the computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the single hydrophone signal mode waveform separation method based on modal frequency modulation wavelet transform as described above.

[0125] The computer device 700 can be a terminal or a server. The computer device 700 includes a processor 720, a memory, and a network interface 750 connected via a system bus 710, wherein the memory may include a non-volatile storage medium 730 and internal memory 740.

[0126] The non-volatile storage medium 730 can store an operating system 731 and a computer program 732. When the computer program 732 is executed, it enables the processor 720 to execute any single hydrophone signal mode waveform separation method based on modal frequency modulation wavelet transform.

[0127] The processor 720 provides computing and control capabilities to support the operation of the entire computer device 700.

[0128] The internal memory 740 provides an environment for the operation of the computer program 732 in the non-volatile storage medium 730. When the computer program 732 is executed by the processor 720, the processor 720 can execute any single hydrophone signal mode waveform separation method based on modal frequency modulation wavelet transform.

[0129] This network interface 750 is used for network communication, such as sending assigned tasks. Those skilled in the art will understand that... Figure 3 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device 700 to which the present application is applied. The specific computer device 700 may include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements. The processor 720 is used to run program code stored in memory to implement the following steps:

[0130] A method for separating the modal waveforms of a single hydrophone signal based on modal frequency modulation wavelet transform includes:

[0131] Collect hydrophone signals;

[0132] The hydrophone signal is processed to obtain the time-domain signal of the sound source;

[0133] The time-frequency diagram of the sound source signal is obtained by using modal frequency modulation wavelet transform to obtain the time-frequency diagram of the sound source signal;

[0134] Select the region containing the desired mode in the time-frequency plot and perform time-frequency domain filtering to obtain the time-frequency plot of the selected mode;

[0135] The time-frequency diagram of the selected mode is used to obtain the time-domain waveform of that mode by inverse mode frequency modulation wavelet transform.

[0136] In one embodiment: the processing of the hydrophone signal to obtain the time-domain signal of the sound source includes:

[0137] The signal collected by the hydrophone is transformed from the time domain to the frequency domain by Fourier transform in order to obtain the measured sound pressure of the sound source.

[0138] The Green's function of the measurement field is obtained by removing the spectrum of the transmitted signal from the measured sound pressure of the sound source through deconvolution.

[0139] The Green's function of the measurement field is transformed from the frequency domain back to the time domain by Fourier transform to obtain the time domain signal of the sound source.

[0140] In one embodiment: obtaining the time-frequency diagram of the sound source signal by using modal frequency modulation wavelet transform includes:

[0141] Wavelet decomposition is performed on the time-domain signal of the sound source using wavelet basis functions to obtain wavelet coefficients;

[0142] Modal decomposition is performed on the wavelet coefficients to obtain each modal component;

[0143] Perform frequency-modulated wavelet transform on each modal component to obtain the time-frequency diagram of each modal component;

[0144] The time-frequency diagrams of each modal component are superimposed to obtain the time-frequency diagram of the sound source signal.

[0145] In one embodiment: the step of selecting the region containing a desired mode in the time-frequency diagram and performing time-frequency domain filtering to obtain the time-frequency diagram of the selected mode includes:

[0146] Select the region containing the desired mode and record the time and frequency range of that region;

[0147] Use a filter to perform time-frequency domain filtering on this region;

[0148] Extract the filtered region to obtain the time-frequency plot of the selected mode.

[0149] It should be understood that, in the embodiments of this application, the processor 720 may be a central processing unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0150] Those skilled in the art will understand that Figure 3 The structure of the computer device 700 shown does not constitute a limitation on the computer device 700, and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0151] In another embodiment of the present invention, a computer-readable storage medium is provided. This computer-readable storage medium may be a non-volatile computer-readable storage medium. The computer-readable storage medium stores a computer program, wherein when executed by a processor, the computer program implements the single hydrophone signal mode waveform separation method based on modal frequency modulation wavelet transform disclosed in the embodiments of the present invention.

[0152] Those skilled in the art will readily understand that, for the sake of convenience and brevity, the specific working processes of the devices, apparatuses, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this invention.

[0153] In the embodiments provided by this invention, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Units with the same function may be grouped into one unit. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. In addition, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interfaces, devices, or units, or it may be an electrical, mechanical, or other form of connection.

[0154] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of the embodiments of the present invention, depending on actual needs.

[0155] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0156] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), magnetic disks, or optical disks.

[0157] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for separating the modal waveforms of a single hydrophone signal based on modal frequency modulation wavelet transform, characterized in that, include: Collect hydrophone signals; The hydrophone signal is processed to obtain the time-domain signal of the sound source; The time-frequency diagram of the sound source signal is obtained by using modal frequency modulation wavelet transform to obtain the time-frequency diagram of the sound source signal; Select the region containing the desired mode in the time-frequency plot and perform time-frequency domain filtering to obtain the time-frequency plot of the selected mode; The time-frequency plot of the selected mode is used to obtain the time-domain waveform of that mode using inverse mode frequency modulation wavelet transform; The step of obtaining the time-frequency diagram of the sound source signal by using modal frequency modulation wavelet transform includes: Wavelet decomposition is performed on the time-domain signal of the sound source using wavelet basis functions to obtain wavelet coefficients; Modal decomposition is performed on the wavelet coefficients to obtain each modal component; Perform frequency-modulated wavelet transform on each modal component to obtain the time-frequency diagram of each modal component; The time-frequency diagrams of each modal component are superimposed to obtain the time-frequency diagram of the sound source signal.

2. The method for separating the modal waveforms of a single hydrophone signal based on modal frequency modulation wavelet transform according to claim 1, characterized in that, The process of processing the hydrophone signal to obtain the time-domain signal of the sound source includes: The signal collected by the hydrophone is transformed from the time domain to the frequency domain by Fourier transform in order to obtain the measured sound pressure of the sound source. The Green's function of the measurement field is obtained by removing the spectrum of the transmitted signal from the measured sound pressure of the sound source through deconvolution. The Green's function of the measurement field is transformed from the frequency domain back to the time domain by Fourier transform to obtain the time domain signal of the sound source.

3. The method for separating the modal waveforms of a single hydrophone signal based on modal frequency modulation wavelet transform according to claim 1, characterized in that, The step of selecting the region containing a desired mode in the time-frequency diagram and performing time-frequency domain filtering to obtain the time-frequency diagram of the selected mode includes: Select the region containing the desired mode and record the time and frequency range of that region; Use a filter to perform time-frequency domain filtering on this region; Extract the filtered region to obtain the time-frequency plot of the selected mode.

4. A single hydrophone signal mode waveform separation device based on modal frequency modulated wavelet transform, wherein, during operation, it executes the single hydrophone signal mode waveform separation method based on modal frequency modulated wavelet transform as described in any one of claims 1-3, characterized in that, It includes an acquisition unit, a signal processing unit, a first conversion unit, a filtering unit, and a second conversion unit; The acquisition unit is used to acquire hydrophone signals; The signal processing unit is used to process the hydrophone signal to obtain the time-domain signal of the sound source; The first transformation unit is used to obtain the time-frequency diagram of the sound source signal by using modal frequency modulation wavelet transform; The filtering unit is used to select the region containing a desired mode in the time-frequency diagram and perform time-frequency domain filtering to obtain the time-frequency diagram of the selected mode. The second transformation unit is used to obtain the time-domain waveform of the selected mode by using inverse mode frequency modulation wavelet transform.

5. The single hydrophone signal mode waveform separation device based on modal frequency modulation wavelet transform according to claim 4, characterized in that, The signal processing unit includes a first transformation module, a filtering module, and a second transformation module; The first transformation module is used to transform the signal collected by the hydrophone from the time domain to the frequency domain through Fourier transform in order to obtain the measured sound pressure of the sound source. The filtering module is used to remove the transmitted signal spectrum from the measured sound pressure of the sound source by deconvolution in order to obtain the Green function of the measurement field. The second transformation module is used to transform the Green's function of the measurement field from the frequency domain back to the time domain through Fourier transform, so as to obtain the time domain signal of the sound source.

6. The single hydrophone signal mode waveform separation device based on modal frequency modulation wavelet transform according to claim 4, characterized in that, The first transformation unit includes a first decomposition module, a second decomposition module, a frequency modulation transformation module, and a superposition module; The first decomposition module is used to perform wavelet decomposition on the time-domain signal of the sound source using wavelet basis functions to obtain wavelet coefficients; The second decomposition module is used to perform modal decomposition on the wavelet coefficients to obtain each modal component; The frequency modulation transform module is used to perform frequency modulation wavelet transform on each modal component to obtain the time-frequency diagram of each modal component; The superposition module is used to superimpose the time-frequency diagrams of each modal component to obtain the time-frequency diagram of the sound source signal.

7. The single hydrophone signal mode waveform separation device based on modal frequency modulation wavelet transform according to claim 4, characterized in that, The filtering unit includes a selection module, a filtering module, and an extraction module; The selection module is used to select the region where a certain mode is located and record the time range and frequency range of the region. The filtering module is used to perform time-frequency domain filtering on the region using a filter; The extraction module is used to extract the filtered region to obtain the time-frequency diagram of the selected mode.

8. A computer device, characterized in that, The method includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the single hydrophone signal mode waveform separation method based on modal frequency modulation wavelet transform as described in any one of claims 1 to 3.

9. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, which includes program instructions. When the program instructions are executed by a processor, the processor performs the single hydrophone signal mode waveform separation method based on modal frequency modulation wavelet transform as described in any one of claims 1 to 3.

Images