A method for joint multi-parameter estimation of wideband moving targets based on vector array

Abstract

The present invention belongs to the technical field of underwater acoustic vector signal processing, and specifically relates to a method for joint estimation of multiple parameters of broadband moving targets based on vector arrays. The present invention performs sub-band division on the broadband sound pressure and velocity signals of noisy targets received by the vector vertical array to construct an acoustic field interference structure containing target parameter information; reflects the interference structure to the time and space domain through vector beamforming, and extracts the interference structure and vertical arrival angle; combines the interference structure and vertical arrival angle to estimate the target depth based on the zero-point periodicity characteristics of the interference structure; cross-correlates the sound pressure and velocity signals received by the vector hydrophone to construct sound pressure and velocity cross-correlation velocity interference fringes, and solves the velocity parameters based on the characteristics of the fringes containing physical quantities. The present invention uses a vector vertical array to obtain the target excitation sound field interference structure and the target vertical arrival angle, and is suitable for fields such as long-range early warning and target detection of underwater targets.

Description

Technical Field

[0001] The present invention belongs to the technical field of underwater acoustic vector signal processing, and in particular relates to a method for joint estimation of multiple parameters of broadband moving targets based on a vector array. Background Art

[0002] The problem of target parameter estimation is an important research direction for underwater target detection. Compared with active detection, passive detection has better concealment. In the deep-sea environment, there are multiple sound propagation channels, among which reliable sound paths have the advantages of long propagation distance, channel stability and low environmental noise level. Compared with traditional acoustic pressure hydrophones, single vector hydrophones can obtain sound pressure and vibration velocity information simultaneously, and have a certain inhibitory effect on isotropic noise; the acoustic vector array based on vector hydrophones further organically combines the spatial azimuth resolution and noise suppression capabilities of vector sensors with the spatial resolution capabilities of the array, greatly expanding the signal processing space, and has better azimuth estimation and noise suppression capabilities than single acoustic pressure arrays.

[0003] Reference 1 (“Performance metrics for depth-based signal separation using deep vertical line arrays”, The Journal of the Acoustical Society of America 139, 418-425 (2016)) proposes using a vertical array deployed near a critical depth to receive acoustic field information from a target near the sea surface via a reliable acoustic path. Target depth is estimated by analyzing the periodic interference structure between the energy and vertical angle of arrival obtained from the received narrowband sound pressure signal beam output. This method ignores variations in sound velocity in water and does not consider the impact of noise on the method.

[0004] Reference 2 (“Source localization by matching sound intensity with an avertical array in the deep ocean”, The Journal of the Acoustical Society of America 146, EL477-EL481 (2019)) proposes a broadband sound source localization method using an asynchronous vertical array in a reliable acoustic path. This method locates the target by copying the sound intensity matrix and constructing a cost function. This method requires relatively precise ocean environmental parameters to achieve a more accurate model, and therefore places high demands on prior information.

[0005] Vector hydrophones can simultaneously receive both the pressure and velocity components of the sound field, enabling them to capture multidimensional sound field information compared to scalar hydrophones. In deep-sea environments, sound waves generated by targets near the sea surface can propagate to the seafloor via direct wave zones or reliable acoustic paths, with minimal propagation loss and stable propagation characteristics.

[0006] Reference 3 (“Passive broadband source depth estimation in the deep ocean using a single vector sensor”, The Journal of the Acoustical Society of America 148, EL88-EL92 (2020)) proposes a method for estimating the depth of targets in the deep ocean's direct sound zone by leveraging the interference structure between the frequency and grazing angle of a broadband sound field. This method can estimate target depth with limited prior information, but it only considers the propagation of sound rays under constant sound speed conditions and fails to account for the bending of sound rays due to the influence of sound speed.

[0007] Reference 4 (“Analysis on the Characteristic of Cross-Correlated Field and Its Potential Application on Source Localization in Deep Water”. J. Comput. Acoust. 2017, 25(2).) proposes a broadband target velocity estimation method. This method uses the sound pressure signal received by a single hydrophone to construct a radial velocity interference structure based on the spatial cross-correlation of the sound pressure. Based on this structure, the target velocity is estimated through Fourier transform. This method does not consider the impact of noise on the radial velocity interference fringe structure and the velocity estimation results. Summary of the Invention

[0008] The purpose of the present invention is to provide a method for joint estimation of multiple parameters of broadband moving targets based on vector array.

[0009] A method for joint estimation of multiple parameters of a wideband moving target based on a vector array comprises the following steps:

[0010] Step 1: Use a vector vertical array deployed near the critical depth to receive the noise-containing pressure signal, horizontal velocity signal, and vertical velocity signal of a near-sea surface broadband target propagating through a reliable acoustic path;

[0011] Step 2: Perform sub-band decomposition on the received broadband sound pressure signal and broadband vibration velocity signal to divide them into a certain number of narrowband signals;

[0012] Step 3: Perform joint sound pressure and velocity beamforming on the divided narrowband sound pressure signal and vibration velocity signal respectively to obtain the sound pressure and velocity joint output beam interference structure cloud map corresponding to each narrowband signal;

[0013] Step 4: Perform structure extraction based on the sound pressure-velocity joint output beam interference structure cloud corresponding to each narrowband signal to obtain the sound pressure-velocity joint output beam interference structure;

[0014] Step 5: Extract the structure based on the sound pressure and velocity corresponding to each narrowband signal and the output beam interference structure cloud map to obtain the target vertical arrival angle curve;

[0015] Step 6: Combine the sound pressure and vibration velocity to output the beam interference structure and the target vertical arrival angle curve, and estimate the target depth parameters based on the characteristics of the interference structure itself;

[0016] Step 7: Cross-correlate the sound pressure signal and the vibration velocity signal received by any vector hydrophone to obtain the sound pressure-vibration velocity cross-correlation velocity interference fringes;

[0017] Step 8: Perform a Fourier-like transform on the sound pressure and velocity combined interference fringes along the frequency direction, and transform the interference fringes from the time-frequency domain to the time-velocity domain to estimate the target velocity.

[0018] Furthermore, step 3 is specifically as follows:

[0019] The sound pressure and velocity signals received by a vector vertical array can be expressed as:

[0020] x(t)=a(θ s )s(t)+noise x (t)

[0021] x vx (t) = x(t) cosθ s =a(θ s )s(t)cosθ s +noise vr (t)

[0022] x vy (t) = x(t) sinθ s =a(θ s )S(T)sinθ s +noise vz (t)

[0023]

[0024] Among them, θ srepresents the vertical arrival angle of the target; s(t) represents the target signal; noise(t) represents the noise received by each channel;

[0025] Combine the horizontal and vertical velocity channels received by the vector vertical array to obtain a new combined velocity signal:

[0026] v c (t) = x vx (t)cosθ+x vy (t)sinθ=a(θ s )s(t)cos(θ-θ s )

[0027] Add the sound pressure signal and the combined vibration velocity signal to obtain the following combined quantity:

[0028] x(t)+v c (t) = a(θ s )s(t)(1+cos(θ-θ s ))

[0029] The combined processing of sound pressure and vibration velocity (x+v c )v c The vector beamforming beam output power is:

[0030]

[0031] Among them, P CBF (θ) is the conventional beamforming beam output power of the scalar acoustic pressure array;

[0032]

[0033] According to the virtual source theory, the sound pressure velocity signal received at a certain point in space can be expressed as:

[0034]

[0035] v r (t,z,ω)=p(t,z,ω)·cosθ s (t)

[0036] v z (t,z,ω)=p(t,z,ω)·sinθ s (t)

[0037]

[0038]

[0039]

[0040] For the given (p+vc )v c The interference structure of the combined vector beam output can be expressed as:

[0041]

[0042] B(ω,sinθ s (t))=2|S(ω)| 2 A 2 (1-cos(2kz s sinθ s (t))).

[0043] Furthermore, step 6 is specifically as follows:

[0044] There is a clear structural connection between the sound pressure and velocity combined output beam interference structure and the vertical arrival angle. The periodicity of the interference structure zero point is expressed as:

[0045] 2kz s Δsinθ s-zero (t)=2π

[0046] The relationship between the target depth parameter and the zero point of the interference period is expressed as:

[0047]

[0048] Where c is the reference sound velocity; f is the target sound source frequency, and its interference period zero point difference Δsinθ s-zero (t) is given by combining the spatial periodic interference modulation structure of the sound pressure velocity and the target vertical arrival angle curve.

[0049] Furthermore, step 7 is specifically as follows:

[0050] The sound pressure signal and the vibration velocity signal are cross-correlated to obtain

[0051]

[0052] Taking the real part, we get:

[0053]

[0054] ΔR(t)=R0(t+Δt)-R0(t)

[0055] The relationship between radial velocity and radial distance is expressed as:

[0056] ΔR(t)=v(t)Δt

[0057] If the periodic oscillation term of the cross-correlation velocity interference fringe of the acoustic pressure velocity satisfies the relationship kΔR(t) = 2π, the estimated value of the radial velocity is expressed as:

[0058]

[0059] Furthermore, step 8 is specifically as follows:

[0060] The Fourier transform function of the cross-correlation velocity interference fringe along the frequency is specifically expressed as:

[0061]

[0062] Where k is the wave number, P represents the number of integration points, and Δf(p)=f(p)-f(p-1).

[0063] The beneficial effects of the present invention are:

[0064] This invention utilizes a vector vertical array to determine the interference structure of the target's excited acoustic field and the target's vertical angle of arrival, eliminating the need for precise measurements of ocean environmental parameters. By combining vector hydrophone technology with array signal processing techniques, this expands the signal processing space compared to traditional acoustic pressure arrays and offers improved isotropic noise suppression. Compared to single vector hydrophones, vector arrays more fully utilize acoustic field information, offering significant advantages in spatial gain and detection range. This invention is suitable for applications such as long-range early warning and detection of underwater targets. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 Flowchart of the present invention.

[0066] Figure 2 Schematic diagram of the vector vertical array and target spatial position.

[0067] Figure 3 Schematic diagram of the target motion state

[0068] Figure 4 This is the sound pressure and velocity joint output beam interference structure and vertical arrival angle curve.

[0069] Figure 5 This is the joint diagram of the sound pressure and velocity joint output beam interference structure and vertical arrival angle.

[0070] Figure 6 It is the velocity interference fringe diagram of the cross-correlation of sound pressure and vibration velocity.

[0071] Figure 7 Comparison chart of radial velocity estimates.

[0072] Figure 8 Output interference structure comparison diagram for sound pressure array and vector vertical array.

[0073] Figure 9 It is the sound pressure and vibration velocity joint output beam interferometry structure depth estimation table. DETAILED DESCRIPTION

[0074] The present invention will be further described below with reference to the accompanying drawings.

[0075] The purpose of the present invention is to use an array composed of vector hydrophones to receive target excitation sound field information, reflect the interference structure characteristics between the direct wave and the sea surface reflection wave in the received sound field into the time and space domain through the beamforming method, and estimate the target depth parameters based on the relationship between the interference structure characteristics and the target depth. Considering that the target signal is a broadband signal, the method performs corresponding processing methods on the received broadband signal to match the above-mentioned depth estimation method. Since the broadband received signal has more and richer target parameters, the sound pressure vibration velocity signal is used to construct the sound pressure vibration velocity cross-correlation velocity interference stripes, and the target velocity parameters are estimated based on the physical structural characteristics of the stripes. The above methods all take into account the influence of certain noise and have good practical application prospects in engineering.

[0076] Step 1: The vector vertical array is placed near the critical depth in the deep-sea waveguide environment to receive the near-sea surface broadband target noise pressure signal p(r,ω,t) and horizontal velocity signal v transmitted through a reliable acoustic path. r (r,ω,t) and vertical vibration velocity signal v z (r,ω,t).

[0077] Step 2: Perform sub-band decomposition on the received broadband sound pressure signal and broadband vibration velocity signal to divide them into a certain number of narrowband signals;

[0078] s(t) is a broadband signal, which is converted from the time domain to the frequency domain, namely S(f). S(f) is divided into L narrowband signal sub-bands, and the center frequency of the lth narrowband signal sub-band is f l .

[0079] Step 3: Perform joint sound pressure and velocity beamforming on the divided narrowband sound pressure signal and velocity signal respectively to obtain the sound pressure and velocity joint output beam interference structure cloud map corresponding to each narrowband signal;

[0080] The sound pressure and velocity signals received by a vector vertical array can be expressed as:

[0081] x(t)=a(θ s )s(t)+noise x (t)

[0082] x vx (t) = x(t) cosθ s =a(θ s )s(t)cosθ s +noise vr (t)

[0083] xvy (t) = x(t) sinθ s =a(θ s )s(t)sinθ s +noise vz (t)

[0084]

[0085] Among them, θ s represents the vertical arrival angle of the target, s(t) represents the target signal, and noise(t) represents the noise received by each channel.

[0086] For ease of analysis, the noise term is ignored in the following analysis. The horizontal velocity and vertical velocity channels received by the vector vertical array are combined to obtain a new combined velocity signal:

[0087] v c (t) = x vx (t)cosθ+x vy (t)sinθ=a(θ s )s(t)cos(θ-θ s )

[0088] By adding the sound pressure signal and the combined vibration velocity signal, the following combined quantity can be obtained:

[0089] x(t)+v c (t) = a(θ s )s(t)(1+cos(θ-θ s ))

[0090] The combined processing of sound pressure and vibration velocity (x+v c )v c The vector beamforming beam output power is:

[0091]

[0092] Among them, P CBF (θ) is the conventional beamforming beam output power of the scalar acoustic pressure array:

[0093]

[0094] According to the virtual source theory, the sound pressure velocity signal received at a certain point in space can be expressed as:

[0095]

[0096] v r (t,z,ω)=p(t,z,ω)·cosθ s (t)

[0097] vz (t,z,ω)=p(t,z,ω)·sinθ s (t)

[0098]

[0099]

[0100]

[0101] For the given (p+v c )v c The interference structure of the combined vector beam output can be expressed as:

[0102]

[0103] B(ω,sinθ s (t))=2|S(ω)| 2 A 2 (1-cos(2kz s sinθ s (t)))

[0104] Step 4: Extract the structure of the sound pressure-velocity joint output beam interference structure corresponding to each narrowband signal based on the sound pressure-velocity joint output beam interference structure.

[0105] Step 5: Extract the structure based on the sound pressure and velocity corresponding to each narrowband signal and the combined output beam interference structure cloud map to obtain the target vertical arrival angle curve;

[0106] Step 6: Combine the sound pressure and vibration velocity to output the beam interference structure and the target vertical arrival angle curve, and estimate the target depth parameters based on the characteristics of the interference structure itself;

[0107] There is a clear structural connection between the sound pressure and velocity combined output beam interference structure and the vertical arrival angle. The periodicity of the interference structure zero point can be expressed as

[0108] 2kz s Δsinθ s-z (t)=2π

[0109] The relationship between the target depth parameter and the zero point of the interference period can be expressed as

[0110]

[0111] Where c is the reference sound velocity, f is the target sound source frequency, and the interference period zero point difference Δsinθ s-zero (t) can be given by combining the spatial periodic interference modulation structure of the sound pressure velocity and the target vertical arrival angle curve.

[0112] Step 7: Cross-correlate the sound pressure signal and the vibration velocity signal received by any vector hydrophone to obtain the sound pressure-vibration velocity cross-correlation velocity interference fringes;

[0113] From the sound pressure signal and vibration velocity signal expressions in step 3, ignoring the time factor, we can get the sound pressure corresponding to time t and the horizontal vibration velocity corresponding to time t+Δt:

[0114]

[0115]

[0116] If Δt is small, then

[0117] sin(kz s sinθ s (t))≈sin(kz s sinθ s (t+Δt))

[0118] The cross-correlation of sound pressure and vibration velocity is:

[0119]

[0120] Taking the real part, we can get

[0121]

[0122] ΔR(t)=R0(t+Δt)-R0(t)

[0123] The relationship between radial velocity and radial distance can be expressed as

[0124] ΔR(t)=v(t)Δt

[0125] If the periodic oscillation term of the cross-correlation velocity interference fringe satisfies the relationship kΔR(t) = 2π, the estimated value of the radial velocity can be expressed as

[0126]

[0127] Step 8: Perform a Fourier-like transform on the sound pressure and velocity combined interference fringes along the frequency direction, and transform the interference fringes from the time-frequency domain to the time-velocity domain to estimate the target velocity.

[0128] The Fourier transform function of the cross-correlation velocity interference fringe along the frequency is specifically expressed as:

[0129]

[0130] Where k is the wave number, P represents the number of integration points, and for Δf, the following relationship holds:

[0131] Δf(p)=f(p)-f(p-1)

[0132] The present invention realizes the joint estimation of depth parameters and motion velocity parameters of a moving target. The broadband sound pressure and velocity signals of a noisy target received by a vector vertical array are divided into sub-bands to construct an acoustic field interference structure containing target parameter information; the interference structure is reflected in the time and space domain through vector beamforming, and the interference structure and vertical arrival angle are extracted; the interference structure and vertical arrival angle are combined to estimate the target depth based on the zero-point periodicity characteristics of the interference structure; the sound pressure and velocity signals received by the vector hydrophone are cross-correlated to construct sound pressure and velocity cross-correlation velocity interference fringes, and the velocity parameters are calculated based on the characteristics of the physical quantities contained in the fringes. The present invention is further described below through simulation experiments.

[0133] The parameters are as follows: the first element of the vector vertical array is set at a depth of 4600m, with a frequency band distribution of 50 to 100Hz. The target signal propagates from the far field to the vector vertical array via a reliable acoustic path, with a target depth of 100m, a signal sampling frequency of 1kHz, and a total sampling time of 600s. Gaussian white noise with a signal-to-noise ratio of 0dB is added within the frequency band of 50Hz to 150Hz. Furthermore, a comparison of the output beam interferometry structures of the vector array and the acoustic pressure array is presented when Gaussian white noise with a signal-to-noise ratio of -10dB is added.

[0134] It can be seen from the above simulation examples that the present invention can achieve the acquisition of sound field interference structure and vertical arrival angle at a lower signal-to-noise ratio. The sound pressure and vibration velocity jointly output beam interference structure and vertical arrival angle curves are shown in Figure 2. Figure 4 As shown, the sound pressure and velocity combined output beam interference structure - vertical arrival angle combined as shown Figure 5 As shown, the target depth estimation results are as follows Figure 9 As shown, the sound pressure velocity cross-correlation velocity interference fringes are as follows Figure 6 The radial velocity estimation results are shown in Figure 7 As shown, the output interference structure comparison between the sound pressure array and the vector vertical array is as follows: Figure 8 The simulation experiments show that the method of the present invention can better realize the joint estimation of target depth parameters and velocity parameters, and compared with the sound pressure array, the vector vertical array has better noise suppression effect.

[0135] This invention utilizes a vector vertical array to determine the interference structure of the target's excited acoustic field and the target's vertical angle of arrival, eliminating the need for precise measurements of ocean environmental parameters. By combining vector hydrophone technology with array signal processing techniques, this expands the signal processing space compared to traditional acoustic pressure arrays and offers improved isotropic noise suppression. Compared to single vector hydrophones, vector arrays more fully utilize acoustic field information, offering significant advantages in spatial gain and detection range. This invention is suitable for applications such as long-range early warning and detection of underwater targets.

[0136] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be within the scope of protection of the present invention.

Claims

1. A method for joint estimation of multiple parameters of broadband moving targets based on vector array, characterized in that: The following steps are involved: Step 1: Use a vector vertical array deployed near the critical depth to receive the noise-containing pressure signal, horizontal velocity signal, and vertical velocity signal of a near-sea surface broadband target propagating through a reliable acoustic path; Step 2: Perform sub-band decomposition on the received broadband sound pressure signal and broadband vibration velocity signal to divide them into a certain number of narrowband signals; Step 3: Perform joint sound pressure and velocity beamforming on the divided narrowband sound pressure signal and vibration velocity signal respectively to obtain the sound pressure and velocity joint output beam interference structure cloud map corresponding to each narrowband signal; Step 4: Perform structure extraction based on the sound pressure-velocity joint output beam interference structure cloud corresponding to each narrowband signal to obtain the sound pressure-velocity joint output beam interference structure; Step 5: Extract the structure based on the sound pressure and velocity corresponding to each narrowband signal and the output beam interference structure cloud map to obtain the target vertical arrival angle curve; Step 6: Combine the sound pressure and vibration velocity to output the beam interference structure and the target vertical arrival angle curve, and estimate the target depth parameters based on the characteristics of the interference structure itself; Step 7: Cross-correlate the sound pressure signal and the vibration velocity signal received by any vector hydrophone to obtain the sound pressure-vibration velocity cross-correlation velocity interference fringes; Step 8: Perform a Fourier-like transform on the sound pressure and velocity combined interference fringes along the frequency direction, and transform the interference fringes from the time-frequency domain to the time-velocity domain to estimate the target velocity.

2. The method for joint estimation of multiple parameters of a wideband moving target based on a vector array according to claim 1, characterized in that: Step 3 is as follows: The sound pressure and velocity signals received by a vector vertical array can be expressed as: x(t)=a(θ s )s(t)+noise x (t) x vx (t)=x(t)cosθ s =a(θ s )s(t)cosθ s +noise vr (t) x vy (t)=x(t)sinθ s =a(θ s )s(t)sinθ s +noise vz (t) Among them, θ s represents the vertical arrival angle of the target; s(t) represents the target signal; noise(t) represents the noise received by each channel; Combine the horizontal and vertical velocity channels received by the vector vertical array to obtain a new combined velocity signal: v c (t)=x vx (t)cosθ+x vy (t)sinθ=a(θ s )s(t)cos(θ-θ s ) Add the sound pressure signal and the combined vibration velocity signal to obtain the following combined quantity: x(t)+v c (t)=a(θ s )s(t)(1+cos(θ-θ s )) The combined processing of sound pressure and vibration velocity (x+v c )v c The vector beamforming beam output power is: Among them, P CBF (θ) is the conventional beamforming beam output power of the scalar acoustic pressure array; According to the virtual source theory, the sound pressure velocity signal received at a certain point in space can be expressed as: v r (t,z,ω)=p(t,z,ω)·cosθ s (t) v z (t,z,ω)=p(t,z,ω)·sinθ s (t) For the given (p+v c )v c The interference structure of the combined vector beam output can be expressed as: B(ω,sinθ s (t))=2|S(ω)| 2 A 2 (1-cos(2kz s sinth s (t)))。 3. The method for joint estimation of multiple parameters of a wideband moving target based on a vector array according to claim 1, characterized in that: Step 6 is as follows: There is a clear structural connection between the sound pressure and velocity combined output beam interference structure and the vertical arrival angle. The periodicity of the interference structure zero point is expressed as: 2kz s Δsinθ s-zer (t)=2π The relationship between the target depth parameter and the zero point of the interference period is expressed as: Where c is the reference sound velocity; f is the target sound source frequency, and its interference period zero point difference Δsinθ s-zer (t) is given by combining the spatial periodic interference modulation structure of the sound pressure velocity and the target vertical arrival angle curve.

4. The method for joint estimation of multiple parameters of a wideband moving target based on a vector array according to claim 2, characterized in that: Step 7 is as follows: The sound pressure signal and the vibration velocity signal are cross-correlated to obtain Taking the real part, we get: ΔR(t)=R0(t+Δt)-R0(t) The relationship between radial velocity and radial distance is expressed as: ΔR(t)=v(t)Δt If the periodic oscillation term of the cross-correlation velocity interference fringe of the acoustic pressure velocity satisfies the relationship kΔR(t) = 2π, the estimated value of the radial velocity is expressed as:

5. The method for joint estimation of multiple parameters of a wideband moving target based on a vector array according to claim 4, characterized in that: Step 8 is as follows: The Fourier transform function of the cross-correlation velocity interference fringe along the frequency is specifically expressed as: Where k is the wave number, P represents the number of integration points, and Δf(p)=f(p)-f(p-1).

Images