Simulation method of seabed reverberation of broadband signals based on variable fractional delay filter

Abstract

The present invention discloses a method for simulating seabed reverberation of broadband signals based on a variable decimal time delay filter, which relates to the technical field of underwater acoustic engineering and solves the problem that in the prior art, only scattering intensity is used to describe the roughness of the seabed in the seabed reverberation modeling, and the undulating seabed surface is not considered to cause the sound path delay difference. The point scattering model based on the superposition of scattering unit echoes comprehensively considers factors such as the scattering intensity of the seabed scattering unit, four types of primary seabed scattering and the transmission signal, and simulates the obtained broadband reverberation time series. The obtained reverberation time series is subjected to frequency statistics, and the simulation results show that the probability density function of the instantaneous value of the reverberation and the envelope obeys Gaussian and Rayleigh distributions. The present invention has high simulation accuracy and good practicality.

Description

Technical Field

[0001] The invention relates to the technical field of underwater acoustic engineering, and in particular to a method for simulating seabed reverberation of broadband signals based on a variable fractional delay filter. Background Art

[0002] Since the target echo of broadband signal carries a large amount of target information and has weak reverberation background correlation, it is conducive to target detection, parameter estimation and target feature extraction. Therefore, broadband signal has become the main transmitting signal in active sonar. Conducting theoretical analysis of broadband signal reverberation is of great significance to the research of active sonar anti-reverberation technology.

[0003] To suppress reverberation, we must first obtain reverberation data. Usually, we need to conduct water tank tests, lake tests, and offshore tests to obtain reverberation data. In actual engineering applications, various conditions often limit the possibility of conducting a large number of lake and offshore tests to obtain reverberation signals. In addition, the test data often have various features fused together, making them difficult to separate. Therefore, it is necessary to complete the reverberation modeling of broadband signals to meet the reverberation data requirements of the new reverberation suppression algorithm.

[0004] In the previous seabed reverberation modeling based on the ray model, the scattering intensity was mainly used to describe the roughness of the seabed, which could not reflect the undulating characteristics of the seabed medium. The actual roughness of the seabed is manifested in two aspects: first, the seabed bottom affects the scattering intensity, and second, the incident signal on the undulating seabed surface will cause a time delay in the sound path. Under narrowband conditions, since the phase and time delay show a linear relationship, an additional phase is added to replace the time delay, and the value range of the additional phase reflects the roughness of the seabed interface. Since the frequency of the broadband signal is no longer constant, the time delay and phase of the broadband signal are no longer a linear relationship. The present invention uses a time delay filter to introduce a time delay difference to describe the undulation of the seabed surface. Taking into account the undulation of the seabed scattering unit, four types of primary seabed scattering and transmission signals, the physical characteristics of the reverberation such as spatial correlation and space-time characteristics can be naturally reflected, and the requirements of the new reverberation suppression algorithm for reverberation data can be met. Summary of the invention

[0005] The present invention provides a method for simulating seabed reverberation of broadband signals based on a variable fractional delay filter, in order to solve the problem that in the prior art, only scattering intensity is used to describe the roughness of the seabed in seabed reverberation modeling, and the problem of sound path delay difference caused by the undulating seabed surface is not considered.

[0006] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0007] A method for simulating seabed reverberation of broadband signals based on a variable fractional delay filter specifically comprises the following steps:

[0008] Step 1: Set the transmission signal, ocean and seabed roughness parameters;

[0009] Step 2: Arrange reverberation scattering units on the seabed space, adopt a point scattering model, and set the coordinates of the scattering units;

[0010] Step 3: Obtain the optimal weight of the FARROW structure filter based on the acoustic path difference of the seabed scattering unit, and store the obtained filter parameters;

[0011] Step 4: Set the scattering intensity of each scattering unit and the sound path delay difference caused by the undulating seabed surface;

[0012] Step 5: adjusting the FARROW structure filter parameters according to the sound path delay difference of each scattering unit, and inputting the preset broadband signal into the adjusted FARROW structure filter for filtering to generate a reverberation sequence;

[0013] Step 6: Superimpose the reverberation sequences generated by all units in the scattering area to obtain the total reverberation simulation waveform.

[0014] Preferably, in step one, the pulse width and bandwidth of the transmitting signal are set as required to determine the transmitting sound source depth, the receiving sound source depth and the sea depth.

[0015] Preferably, in step 2, the distribution range of the seabed scattering units is first determined, and the reverberation scattering units are arranged within this range.

[0016] Preferably, in step 3, the specific steps of finding the optimal weight of the FARROW structure filter are:

[0017] Step 31: The objective function of fitting the frequency characteristics of the digital delay filter is

[0018] minimax W(ω)|H(e jω )-H d (e jω )|

[0019] Among them, H(e jω ) is the Fourier transform of the ideal infinite length h(n), H d (e jω ) is the target frequency characteristic, W(ω) is the frequency weighting function;

[0020] Step 32: Assume that the range of the fractional delay d is [-0.5, 0.5], and the order of the polynomial fitting is p, then the delay filter is:

[0021]

[0022] Among them, C in (n) is the fitting coefficient of the fractional delay filter, N is the number of taps of the filter, and hd (n) is the filter coefficient of the time delay d, then the filter transfer function H d (e jω )for

[0023]

[0024] in,

[0025] Preferably, in step 4, the specific steps of setting the scattering intensity of each scattering unit and the acoustic path delay difference caused by the undulating seabed surface are:

[0026] Step 41: setting the scattering coefficient A(i) of the i-th scattering unit according to the seabed scattering medium, where A(i) obeys a Gaussian distribution with a mean value of μ;

[0027] Step 42: Set the delay difference Δτ of the i-th scattering unit according to the fluctuation of the seabed surface i , Δτ i It obeys the uniform distribution of (τ1, τ2), where τ1 and τ2 are pre-set values.

[0028] Preferably, in step 5, the formula for generating the reverberation sequence is:

[0029] Step 51: The reverberation sequence of the i-th scattering unit is:

[0030]

[0031] Where A(i) is the scattering coefficient of the i-th scattering unit, t is and R is are the time and distance of the sound wave propagating to the i-th scattering unit, t im and R im are the propagation time and distance of the signal received by the receiving hydrophone from the i-th scattering unit, dA is the area of ​​the scattering unit, the incident angle of the sound wave is θ, and the output angle is

[0032] Step 52: Adjust the input delay d = Δτ of the FARROW structure filter i , the input signal of the ith scattering element is filtered by the FARROW structure filter to output the reverberation time series of the ith scattering element, and then the filtered time series is multiplied by the weight

[0033] Preferably, in step six, the reverberation signal received by the hydrophone from the rough surface of the seabed is

[0034]

[0035] The beneficial effects of the present invention are as follows: the present invention provides a method for simulating seabed reverberation of broadband signals based on a variable time-delay filter, which solves the problem that in traditional seabed reverberation modeling, only scattering intensity is used to describe the roughness of the seabed, and the undulating seabed surface is not considered to cause the sound path delay difference. The present invention is based on a point scattering model of superimposed echoes of scattering units, and comprehensively considers factors such as the scattering intensity of the seabed scattering unit, four types of primary seabed scattering, and the transmitted signal, to simulate and obtain a broadband reverberation time series. The obtained reverberation time series is subjected to frequency statistics, and the simulation results show that the probability density function of the instantaneous value and envelope of the reverberation obeys Gaussian and Rayleigh distributions. The new method can meet the requirements of the new reverberation suppression algorithm for reverberation data. It has high simulation accuracy and good practicality. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 It is the main flow chart of the broadband signal seabed reverberation simulation based on the variable fractional delay filter of the present invention;

[0037] Figure 2 Schematic diagram of four types of primary seafloor scattering models;

[0038] Figure 3 It is a variable fractional delay FIR filter of FARROW structure;

[0039] Figure 4 is the amplitude-frequency response curve of the FARROW structure filter;

[0040] Figure 5 It is the phase-frequency response curve of the FARROW structure filter;

[0041] Figure 6 It is the waveform of the total reverberation signal;

[0042] Figure 7 It is the probability density curve of the instantaneous value of the simulated reverberation sequence;

[0043] Figure 8 It is the probability density curve of the simulated reverberation sequence envelope. DETAILED DESCRIPTION

[0044] The following is a clear and complete description of the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work belong to the protection scope of the present invention.

[0045] See also Figure 1 In an embodiment of the present invention, a method for simulating seabed reverberation of a broadband signal based on a variable delay filter is provided, comprising the following steps:

[0046] Step 1: Set the parameters of the transmitting signal, ocean, and seabed roughness. In this example, the basic parameters are set as follows: the pulse width T and bandwidth B of the transmitting signal, the sound source depth h1, the receiving hydrophone depth h2, and the sea depth H. Assuming that the transmitting sound source and the scattered sound source are expanded as spherical waves, the sound wave incident angle θ and reflection angle of each scattering unit are solved based on the four types of seabed primary scattering models. The distance that the sound wave travels from the sound source to the scattering unit is R is , the distance from the scattered echo to the receiving hydrophone is R is .like Figure 2 As shown:

[0047] The incident angle of the first type of sound ray is The reflection angle is

[0048] The incident angle of the second type of sound ray is The reflection angle is

[0049] The incident angle of the third type of sound ray is The reflection angle is

[0050] The incident angle of the fourth type of sound is The reflection angle is

[0051] Step 2: Arrange reverberation scattering elements in the seabed space, use the point scattering model, and set the coordinates of the scattering elements. This example determines the distribution range of the seabed scattering elements based on the length of the broadband reverberation time series, and distributes the reverberation scattering elements within this range. Based on the spherical coordinates, evenly distribute the scattering elements in radius and horizontal angle.

[0052] Step 3: Obtain the optimal weight of the FARROW structure filter based on the acoustic path difference of the seabed scattering unit, and store the obtained filter parameters;

[0053] Step 31: The objective function of fitting the frequency characteristics of the digital delay filter is minimax W(ω)|H(e jω )-H d (e jω )|; where H(e jω ) is the Fourier transform of the ideal infinite length h(n), H d (e jω ) is the target frequency characteristic, and W(ω) is the frequency weighting function.

[0054] Step 32: Assume that the range of the fractional delay d is [-0.5, 0.5] and the order of the polynomial fitting is p, then the delay filter is

[0055]

[0056] Among them, C in (n) is the fitting coefficient of the fractional delay filter, N is the number of taps of the filter, and h d (n) is the filter coefficient for the time delay d. Then the filter transfer function H d (e jω )for

[0057]

[0058] in,

[0059] Step 4: Set the scattering intensity of each scattering unit and the sound path delay difference caused by the undulating seabed surface;

[0060] Step 41: The scattering coefficient A(i) of the i-th scattering unit is set according to the seabed scattering medium, and A(i) obeys a Gaussian distribution with a mean value of μ.

[0061] Step 42: Set the delay difference Δτ of the i-th scattering unit according to the fluctuation of the seabed surface i , Δτ i It obeys the uniform distribution of (τ1, τ2), where τ1 and τ2 are pre-set values.

[0062] Step 5 is to adjust the parameters of the FARROW structure filter according to the time delay path difference of each scattering unit, and input the preset broadband signal into the adjusted FARROW structure filter for filtering to generate a reverberation sequence;

[0063] Step 51: The reverberation sequence of the i-th scattering unit is:

[0064]

[0065] Where A(i) is the scattering coefficient of the i-th scattering unit, t is and R is are the time and distance of the sound wave propagating to the i-th scattering unit, respectively, where v is the speed of sound waves in water. im and R im are the propagation time and distance of the signal from the i-th scattering unit received by the receiving hydrophone, respectively. dA is the area of ​​the scattering element.

[0066] Step 52: Adjust the input delay d = Δτ of the FARROW structure filter i , the input signal of the ith scattering element is filtered by the FARROW structure filter to output the reverberation time series of the ith scattering element, and then the filtered time series is multiplied by the weight

[0067] Step 6: Superimpose the reverberation generated by all units in the scattering area to obtain the total reverberation signal. The specific steps of this example are:

[0068] Update the reverberation time sequence of each scattering unit, and superimpose the time before and after each scattering unit reaches the hydrophone to obtain the total reverberation sequence. The formula is

[0069]

[0070] Embodiment 1:

[0071] Step 1: Set the parameters such as the transmission signal, ocean and seabed roughness;

[0072] A linear frequency modulation signal with a duration of 200ms and a bandwidth range of 600Hz to 1400Hz is selected as the transmitting signal; the ocean simulation conditions are: the shallow sea depth is 200m, the water depth of the transmitting sound source is 10m, the depth of the receiving transducer is 50m, and the water sound speed is 1500m / s.

[0073] Step 2: Randomly arrange reverberation scattering elements in the seabed space;

[0074] Step 3: Obtain the optimal weight of the FARROW structure filter based on the acoustic path delay difference of the seabed scattering unit, and store the obtained filter parameters;

[0075] Figure 3 For a FARROW structure filter, the order of the finite impulse response filter is set to 15, the order of the polynomial fitting is set to 5, and the passband cutoff normalized frequency of the low-pass filter is set to 0.6.

[0076] Figure 4 , 5 These are the amplitude-frequency characteristic curve and phase-frequency characteristic curve of the filter obtained for different delay values.

[0077] Step 4: Set the scattering intensity of each scattering unit and the delay difference caused by the undulating seabed surface, and adjust the FARROW structure filter parameters. The specific steps of this example are:

[0078] A(i) follows a Gaussian distribution with mean μ=0, Δτ i Obeying the uniform distribution of (τ1, τ2), τ1 = -0.2 and τ2 = 0.2.

[0079] Step five is to input the preset broadband signal into the FARROW structure filter for filtering to generate a reverberation sequence.

[0080] Step six is ​​to sum the reverberations generated by all units in the scattering area to obtain a total reverberation simulation waveform;

[0081] Figure 6This is the total reverberation waveform obtained by adding four types of primary seabed scattering reverberations. Figure 7 and Figure 8 The probability density functions of the instantaneous value and envelope of the total reverberation are shown in Figure 2. The frequency statistics of the seabed reverberation time series obtained by simulation based on the variable fractional delay filter are performed. The instantaneous value and reverberation obtained by averaging a large amount of data obey the Gaussian distribution and Rayleigh distribution, which are consistent with the theoretical value.

[0082] In the description of this specification, specific features, structures, materials or characteristics may be combined in an appropriate manner in any one or more embodiments or examples.

[0083] Of course, the present invention may have many other embodiments. Without departing from the spirit and essence of the present invention, technicians familiar with the field may make various corresponding changes and modifications based on the present invention, but these changes and modifications should all fall within the scope of protection of the claims of the present invention.

Claims

1. A method for simulating the seabed reverberation of a broadband signal based on a variable fractional delay filter, characterized in that: The specific steps include: Step 1: Set the transmission signal, ocean and seabed roughness parameters; Step 2: Arrange reverberation scattering units on the seabed space, adopt a point scattering model, and set the coordinates of the scattering units; Step 3: Obtain the optimal weight of the FARROW structure filter based on the acoustic path difference of the seabed scattering unit, and store the obtained filter parameters; Step 4: Set the scattering intensity of each scattering unit and the sound path delay difference caused by the undulating seabed surface; Step 5: adjusting the FARROW structure filter parameters according to the sound path delay difference of each scattering unit, and inputting the preset broadband signal into the adjusted FARROW structure filter for filtering to generate a reverberation sequence; Step 6: Superimpose the reverberation sequences generated by all units in the scattering area to obtain the total reverberation simulation waveform; In step 3, the specific steps for finding the optimal weights of the FARROW structure filter are: Step 31: The objective function of fitting the frequency characteristics of the digital delay filter is minimax W(ω)|H(e jω )-H d (e jω )| Among them, H(e jω ) is the Fourier transform of the ideal infinite length h(n), H d (e jω ) is the target frequency characteristic, W(ω) is the frequency weighting function; Step 32: Assume that the range of the fractional delay d is [-0.5, 0.5], and the order of the polynomial fitting is p, then the delay filter is: Among them, C m (n) is the fitting coefficient of the fractional delay filter, N is the number of taps of the filter, and h d (n) is the filter coefficient of the time delay d, then the filter transfer function H d (e jω )for in, In step 4, the specific steps of setting the scattering intensity of each scattering unit and the sound path delay difference caused by the undulating seabed surface are: Step 41: setting the scattering coefficient A(i) of the i-th scattering unit according to the seabed scattering medium, where A(i) obeys a Gaussian distribution with a mean value of μ; Step 42: Set the delay difference Δτ of the i-th scattering unit according to the fluctuation of the seabed surface i , Δτ i It obeys the uniform distribution of (τ1, τ2), where τ1 and τ2 are pre-set values; In step 5, the formula for generating the reverberation sequence is: Step 51: The reverberation sequence of the i-th scattering unit is: Where A(i) is the scattering coefficient of the i-th scattering unit, t is and R is are the time and distance for the sound wave to propagate to the i-th scattering unit, t im and R im are the propagation time and distance of the signal received by the receiving hydrophone from the i-th scattering unit, dA is the area of ​​the scattering unit, the incident angle of the sound wave is θ, and the output angle is Step 52: Adjust the input delay d = Δτ of the FARROW structure filter i , the input signal of the ith scattering element is filtered by the FARROW structure filter to output the reverberation time series of the ith scattering element, and then the filtered time series is multiplied by the weight 2. The method for simulating the broadband signal seabed reverberation based on a variable fractional delay filter according to claim 1 is characterized in that: In step 1, the pulse width and bandwidth of the transmitting signal are set according to requirements, and the transmitting sound source depth, the receiving sound source depth and the sea depth are determined.

3. The method for simulating the broadband signal seabed reverberation based on a variable fractional delay filter according to claim 2 is characterized in that: In step 2, the distribution range of the seabed scattering units is first determined, and the reverberation scattering units are arranged within this range.

4. The method for simulating broadband signal seabed reverberation based on variable fractional delay filter according to claim 1, characterized in that: In step 6, the reverberation signal received by the hydrophone from the rough surface of the seabed is

Images