Sonar echo denoising method
By combining the DWT and RLS algorithms to create a sonar echo denoising method, the imaging ambiguity problem of traditional filtering algorithms in complex water environments is solved, effective denoising and high-frequency feature protection of sonar echo signals are achieved, and the clarity of sonar imaging is improved.
Patent Information
- Application Number
- CN202510476086.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-16
- Publication Date
- 2025-10-10
AI Technical Summary
Traditional filtering algorithms cannot effectively suppress sidelobe beams in complex water environments, and the fixed cutoff frequency cannot adapt to the time-varying noise characteristics, resulting in sonar echo distortion and blurred imaging.
The discrete wavelet transform (DWT) is combined with the recursive least squares (RLS) algorithm to first reduce the noise of the sonar echo signal, and then perform RLS filtering, adaptive beamforming and pulse compression. The noise is suppressed and the azimuth resolution is maintained by quickly adjusting the weight update.
The filtering effect of multi-beam sonar echo signals is improved, the imaging ambiguity problem of traditional filtering algorithms in complex water environments is solved, and the effective suppression of time-varying noise and protection of the target's high-frequency characteristics are achieved.
Smart Images
Figure CN120763459A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of sonar echo denoising, and in particular to a sonar echo denoising method. Background Art
[0002] The basic principle of the forward-looking multi-beam sonar imaging and positioning system is: first, transmit an acoustic signal to the target, and calculate the target azimuth angle based on the time delay of the echo signal reaching each element of the receiving array; second, calculate the distance between the target and the sonar by the time difference between the transmitted signal and the echo signal.
[0003] However, complex water environments, especially those in narrow waters, are susceptible to interference from water noise, namely the composite noise field formed by water boundary reflections, biological activity, and instrument self-noise, as well as the multipath effect, which is the coherent superposition of sound waves reflected multiple times by the water surface and bottom, resulting in target echo distortion. Underwater target imaging detection has a certain degree of ambiguity. Therefore, filtering algorithms play a very important role in forward-looking multi-beam sonar imaging.
[0004] Traditional filtering algorithms mainly use square-law detection and low-pass filtering to reduce the impact of environmental noise on imaging quality. Their advantages are simple implementation, low algorithm complexity, hardware-friendly implementation, and small algorithm calculation amount. However, since the filtering process is after beamforming and the processing timing is limited, traditional filtering algorithms cannot effectively suppress sidelobe beams. The frequency band truncation characteristics of the low-pass filter can easily damage the high-frequency characteristic components of the target, and the fixed cutoff frequency cannot adapt to the time-varying noise characteristics, and there is a risk of filtering and under-filtering. Summary of the Invention
[0005] The purpose of the present invention is to solve the problems existing in the prior art and to propose a sonar echo denoising method.
[0006] In order to achieve the above object, the present invention adopts the following technical solutions: A sonar echo denoising method comprises the following steps: S1, use DWT algorithm to reduce noise of the original signal; S2, performs RLS filtering on the denoised signal; S3, performing beamforming processing on the filtered signal; S4, performing pulse compression processing on the formed beam to obtain the final sonar image.
[0007] Preferably, the specific process of step S2 includes: performing noise filtering on the denoised signal using a spatial filter, and the spatial filter quickly adjusts the speed of weight update according to the error size to ensure that the azimuth resolution is not affected.
[0008] Preferably, in step S2, the weight adjustment uses a recursive least squares RLS algorithm to perform value search.
[0009] Preferably, the specific steps of step S3 include: using adaptive beamforming calculation to obtain pre-formed beams in various directions.
[0010] Preferably, in step S4, the specific steps of performing pulse compression processing on the formed beam include: The frequency domain signal after beamforming is obtained through fast Fourier transform; The frequency domain of the transmitted signal is complex conjugate multiplied by the frequency domain signal after beamforming, and then an inverse fast Fourier transform is performed to obtain the time domain result after pulse compression.
[0011] Preferably, the specific steps of step S1 include: Acquire a noise signal; Select wavelet basis function to perform multi-layer decomposition on the signal, decompose the signal into multiple components, and obtain low-frequency components and high-frequency components; The high-frequency coefficients of each layer in the high-frequency component are selected by threshold quantization; According to the low-frequency coefficients of the Nth layer of wavelet decomposition and the high-frequency coefficients of the 1st to Nth layers after quantization, wavelet reconstruction is performed; Output the reconstructed noise-reduced signal.
[0012] Compared with the prior art, the present invention has the following beneficial effects: The present invention combines the discrete wavelet transform (DWT) with the recursive least squares (RLS) algorithm to perform sonar echo denoising, solving the problem of distortion generated when using the RLS algorithm alone to process echo signals with a small signal-to-noise ratio, and improving the filtering effect of multi-beam sonar echo signals. BRIEF DESCRIPTION OF THE DRAWINGS
[0013] Figure 1 This is a flow chart of a sonar echo denoising method proposed in the present invention. DETAILED DESCRIPTION
[0014] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.
[0015] Reference Figure 1 , a sonar echo denoising method, comprising the following steps: S1, use DWT algorithm to reduce noise of the original signal; S2, performs RLS filtering on the denoised signal; S3, performing beamforming processing on the filtered signal; S4, performing pulse compression processing on the formed beam to obtain the final sonar image.
[0016] In this embodiment, the specific process of step S2 includes: using a spatial filter to perform noise filtering on the de-noised signal, and the spatial filter quickly adjusts the speed of weight update according to the error size to ensure that the azimuth resolution is not affected; Among them, the weight adjustment uses the recursive least squares RLS algorithm to search for values; The filter weights are updated by recursively minimizing the weighted sum of squares of the error signal. Its convergence speed is an order of magnitude faster than the traditional LMS algorithm, which is particularly suitable for scenarios that require rapid tracking of changes. This method uses the RLS algorithm to quickly adjust the spatial filter weights, effectively suppressing noise while maintaining azimuth resolution.
[0017] In this embodiment, the specific steps of step S3 include: using adaptive beamforming calculation to obtain pre-formed beams in various directions.
[0018] In this embodiment, in step S4, the specific steps of performing pulse compression processing on the formed beam include: Pulse compression is performed through fast Fourier transform to obtain the frequency domain signal after beamforming: ; The frequency domain expression of the transmitted signal is: ; The signal With signal After multiplying the complex conjugate of , the inverse fast Fourier transform is performed to obtain the time domain result after pulse compression: .
[0019] In this embodiment, the specific steps of step S1 include: Acquire a noise signal; Select wavelet basis function to perform multi-layer decomposition on the signal, decompose the signal into multiple components, and obtain low-frequency components and high-frequency components; The high-frequency coefficients of each layer in the high-frequency component are selected by threshold quantization; According to the low-frequency coefficients of the Nth layer of wavelet decomposition and the high-frequency coefficients of the 1st to Nth layers after quantization, wavelet reconstruction is performed; Output the reconstructed noise-reduced signal.
[0020] Compared with the existing technology, the present invention first performs noise reduction processing on the noisy signal, then performs filtering, and finally performs adaptive beamforming, pulse compression and imaging processing; Sonar echo denoising is performed based on the combination of discrete wavelet transform (DWT) and recursive least squares (RLS) algorithm. This solves the problem of distortion when using the RLS algorithm alone to process echo signals with low signal-to-noise ratio, and improves the filtering effect of multi-beam sonar echo signals. The present invention solves the problems that the filtering process in the existing technology is carried out after beamforming, the processing timing is limited, the sidelobe beam cannot be effectively suppressed, the frequency band truncation characteristics of the low-pass filter easily damage the high-frequency characteristic components of the target, and the fixed cutoff frequency cannot adapt to the time-varying noise characteristics, resulting in the risk of filtering and under-filtering.
[0021] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with the technical field, within the technical scope disclosed by the present invention, who makes equivalent replacements or changes based on the technical solution and inventive concept of the present invention, should be covered by the scope of protection of the present invention.
Claims
1. A sonar echo denoising method, characterized by: The following steps are involved: S1, use DWT algorithm to reduce noise of the original signal; S2, performs RLS filtering on the denoised signal; S3, performing beamforming processing on the filtered signal; S4, performing pulse compression processing on the formed beam to obtain the final sonar image.
2. The sonar echo denoising method according to claim 1, wherein: The specific process of step S2 includes: using a spatial filter to perform noise filtering on the denoised signal, and the spatial filter quickly adjusts the speed of weight update according to the error size to ensure that the azimuth resolution is not affected.
3. The sonar echo denoising method according to claim 2, wherein: In step S2, the weight adjustment is performed by searching for values using a recursive least squares (RLS) algorithm.
4. The sonar echo denoising method according to claim 1, wherein: The specific steps of step S3 include: using adaptive beamforming to calculate and obtain pre-formed beams in various directions.
5. The sonar echo denoising method according to claim 1, wherein: In step S4, the specific steps of performing pulse compression processing on the formed beam include: The frequency domain signal after beamforming is obtained through fast Fourier transform; The frequency domain of the transmitted signal is complex conjugate multiplied by the frequency domain signal after beamforming, and then an inverse fast Fourier transform is performed to obtain the time domain result after pulse compression.
6. The sonar echo denoising method according to claim 1, wherein: The specific steps of step S1 include: Acquire a noise signal; Select wavelet basis function to perform multi-layer decomposition on the signal, decompose the signal into multiple components, and obtain low-frequency components and high-frequency components; The high-frequency coefficients of each layer in the high-frequency component are selected by threshold quantization; According to the low-frequency coefficients of the Nth layer of wavelet decomposition and the high-frequency coefficients of the 1st to Nth layers after quantization, wavelet reconstruction is performed; Output the reconstructed noise-reduced signal.