Adaptive estimation method for fluctuating background under active sonar multi-clutter strong interference

Abstract

The invention discloses a fluctuating background adaptive estimation method under active sonar multi-clutter strong interference, and belongs to the field of active sonar underwater acoustic signal processing. According to the method, backgrounds are respectively estimated through the left eccentric analysis window and the right eccentric analysis window, then the smaller value is taken as the interference background estimation value near the position of the target, and then the signal-to-interference ratio is calculated, so that the risk that a weak target is masked by neighbor strong interference can be avoided or reduced. The method provided by the invention can be used for estimating the uneven background in active sonar application, in particular to self-adaptive estimation of the uneven background in the presence of multi-clutter strong interference. The signal-to-noise ratio of the target can be directly obtained according to the estimated dynamic background, and a basis is established for target detection and positioning. Adaptive estimation of the fluctuating background under multi-clutter strong interference is a necessary preposed link of active sonar signal processing and has an important footstone position.

Description

Technical Field

[0001] This invention relates to the field of active sonar underwater acoustic signal processing technology, and in particular to an adaptive estimation method for undulating background under strong interference from multiple clutter waves in active sonar. Background Technology

[0002] The echo signals received by active sonar include not only target echoes but also severe reverberation interference, interference from numerous non-interest-of-interest scatterers such as seabed topography, as well as marine environmental noise, platform self-noise, and receiver circuitry noise. These non-target echo components collectively constitute the background interference for active sonar. This background interference is variable and unstable in spatial (azimuth + range), temporal (period interval), and Doppler (frequency) dimensions, and is often simply referred to as fluctuating background interference. Adaptive estimation of this fluctuating background under strong clutter interference is fundamental to active sonar target detection and tracking, and clutter suppression.

[0003] There is relatively little literature on active sonar. While active sonar and fire control radar share similarities, they also have significant differences. In the field of radar signal processing, numerous patents and documents exist on adaptive estimation of fluctuating backgrounds based on constant false alarm rate (CFAR) detection, but these are unsuitable for active sonar applications. The main reasons are: radar has a high carrier frequency, large signal bandwidth, high range resolution, fast waveguide velocity, short period interval, and is prone to multi-pulse accumulation, resulting in a sufficiently large number of echo signal samples within the same range. The echo signal can be considered a short-time stationary random signal, supporting various classical CFAR or knowledge-assisted and intelligent CFAR detection techniques. In contrast, sonar has a low carrier frequency, a much smaller signal bandwidth, significantly lower range resolution, slower waveguide velocity, longer period interval, and is less prone to multi-pulse accumulation, resulting in a smaller and less stable number of echo signal samples within the same range. Therefore, target detection against fluctuating backgrounds requires alternative approaches.

[0004] Active sonar underwater target detection has long been a global challenge. Research reports, patents, and published papers with engineering applications in this area are relatively rare. Most available methods for estimating active sonar background undulations involve homogenization and pre-whitening. While smoothing the background, these methods either destroy the fine structure of the target echo or are unsuitable for active sonar operating environments with strong clutter and interference. Summary of the Invention

[0005] The purpose of this invention is to provide an adaptive estimation method for undulating background under strong clutter interference of active sonar, so as to estimate a smooth undulating background and highlight weak targets under strong clutter interference.

[0006] To address the aforementioned technical problems, this invention provides an adaptive estimation method for undulating background under strong interference from multiple clutter waves in active sonar, comprising the following steps: Step 1: Denote the signal as The lengths of the signal protection window and the background window are related to the target scale. On the distance axis, the size of the protection window is related to the range resolution of the active signal. The target scale is set as follows: Distance resolution is Protect window length Background window length It is 2 to 3 times more protective than the window; Step 2: The length of the analysis window on the left. The right-side analysis window is the length, and the left-side off-center window is designed as follows: The right-side off-center window is designed as k is 5~10 The sampling intervals shall be no less than 10 and no greater than the target protection scale. The output result of the left window is denoted as The output on the right is denoted as ; Step 3: Final Output The mean is denoted as Threshold m can be 1 or 2; In the formula, express The samples in; Step 4: Signal With threshold value The amplitudes exceeding the threshold are used as the threshold value, while those not exceeding the threshold are used as the sample amplitude. The output is the dynamic reverberation / background noise estimation signal for that processed frame. .

[0007] The present invention provides an adaptive estimation method for undulating background under strong interference from multiple clutter waves in active sonar, which has the following beneficial effects: (1) The left and right eccentric analysis windows estimate the background respectively, and then the smaller value is taken as the estimated value of the interference background near the target location. Then the signal-to-interference ratio is calculated, which can avoid or reduce the risk of strong interference from the vicinity masking weak targets. (2) It can ensure that no matter which side of the target the strong interference is on, there is always an eccentric analysis window that is not affected or less affected by the strong interference on the background. The smaller value of the background value obtained by the left and right eccentric window analysis can be taken as the background value around the target. (3) is a target detection method based on measured data and Gaussian model, which is suitable for strong reverberation interference and provides robust detection performance even under complex environmental changes. Attached Figure Description

[0008] Figure 1 This is a schematic diagram of the left-right eccentricity analysis window design provided by the present invention.

[0009] Figure 2 This invention provides the original signal and estimated background image of the beam containing the large-scale target.

[0010] Figure 3 This invention provides the original signal and estimated background image of the beam containing the small-scale target. Detailed Implementation

[0011] The following detailed description, in conjunction with the accompanying drawings and specific embodiments, provides a more detailed explanation of the adaptive estimation method for undulating background under strong interference from multiple clutter waves in active sonar, as proposed in this invention. The advantages and features of this invention will become clearer from the following description. It should be noted that the accompanying drawings are all in a very simplified form and use non-precise scales, and are only used to facilitate and clarify the illustration of the embodiments of this invention.

[0012] The essential meaning of whether a sonar target is detectable is whether the target echo stands out from the surrounding interference. The signal-to-interference ratio (SIR) of the target echo is defined as the ratio of the target's peak power to the average power of its surrounding background interference. In fluctuating background conditions, the SIR of the target echo is a fundamental quantitative value for determining whether the target can be detected by the receiver system. While the peak power of the target echo is readily apparent, the core problem this invention aims to solve is how to estimate the average power of the background interference surrounding the target in fluctuating background conditions.

[0013] This invention provides an adaptive estimation method for undulating background under strong interference from multiple clutter waves in active sonar, comprising the following steps: (1) The signal is denoted as The lengths of the signal protection window and the background window are related to the target scale. On the distance axis, the size of the protection window is related to the range resolution of the active signal. The target scale is set as follows: Distance resolution is Then protect the window length Background window length It is 2 to 3 times more protective than the window.

[0014] (2) Since it is impossible to predict in advance whether the interference is on the left or right side of the target, the background estimation analysis window is designed with an eccentric structure, such as... Figure 1 As shown. When the interference is to the left of the target, the left-biased analysis window does not overlap with the strong interference on the left. The signal within the left-biased analysis window is used to estimate the background, which is unaffected or minimally affected by the strong interference. The same principle applies when the interference is to the right. The background is estimated separately for both the left and right sides, and the smaller value is taken as the estimated background interference value near the target's location. Then, the signal-to-interference ratio is calculated.

[0015] The length of the analysis window on the left. The right-side analysis window is the length, and the left-side off-center window is designed as follows: The right-side off-center window is designed as Typically, k is designed to be 5 to 10 times. The design is to have no less than 10 sampling intervals and no larger than the target protection scale. The output result of the left window is denoted as The output on the right is denoted as .

[0016] (3) Final output The mean is denoted as Design threshold m can be 1 or 2.

[0017] In the formula, express The samples in.

[0018] (4) Signal With threshold value The amplitudes exceeding the threshold are used as the threshold value, while those not exceeding the threshold are used as the sample amplitude. The output is the dynamic reverberation / background noise estimation signal for that processed frame. .

[0019] Figure 2 and Figure 3 The original signals and estimated background images of the beams containing large-scale and small-scale targets are shown respectively. It can be seen that the active sonar background estimation method proposed in this invention is feasible. By combining the target scale characteristics to set the length of the protection window and the detection window, the target signal is well protected while suppressing interference and highlighting the target, which has a certain degree of robustness.

[0020] The above description is merely a description of preferred embodiments of the present invention and is not intended to limit the scope of the present invention in any way. Any changes or modifications made by those skilled in the art based on the above disclosure shall fall within the protection scope of the claims.

Claims

1. An adaptive estimation method for undulating background under strong interference from multiple clutter waves in active sonar, characterized in that, Includes the following steps: Step 1: Denote the signal as The lengths of the signal protection window and the background window are related to the target scale. On the distance axis, the size of the protection window is related to the range resolution of the active signal. The target scale is set as follows: Distance resolution is Protect window length Background window length It is 2 to 3 times more protective than the window; Step 2: The length of the analysis window on the left. The right-side analysis window is the length, and the left-side off-center window is designed as follows: The right-side off-center window design is k is 5~10. The sampling intervals shall be no less than 10 and no greater than the target protection scale. The output result of the left window is denoted as The output on the right is denoted as Step 3: Final Output The mean is denoted as Threshold m can be 1 or 2; In the formula, express Samples in the sample; Step 4: Signal With threshold value The amplitudes exceeding the threshold are used as the threshold value, while those not exceeding the threshold are used as the sample amplitude. The output is the dynamic reverberation / background noise estimation signal for that processed frame. .

Images