Nonlinear filtering effects of intrawell matched stochastic resonance with barrier constrainted duffing system for ship radiated line signature extraction

Abstract

Abstract Harmonic signatures extraction is of key interests in remote passive sonar detection and classification for its relevant to the inherent characteristics of engine and shaft/propeller rotations, of which the long-term challenging problem is to address the weak signatures from heavy noise background in the far field scenario. Signal filtering is often utilized as an essential preprocessing tool to purify the acquired noisy signals for applications, whereas stochastic resonance (SR) based nonlinear filter, has been proven effective for weak signature extraction especially under low signal-to-noise ratio (SNR) conditions. In this paper, a novel intrawell matched stochastic resonance (IMSR) is parameterized and implemented to further improve the output signal-to-noise ratio (SNR) with a barrier constrained quartic double-well potential. It can release the limitation on nonlinear system response, and ease the insufficient time-scale matching constraint and the inaccurate noise intensity estimation on potential parameters tuning. The nonlinear filtering effects are comparatively analyzed, which reflect superior performance and can break through the large sampling rate limitation as well. The distinct merits of IMSR are summarized as follows: 1) a superior low-noise output performance to the interwell response in the sense of nonlinear filtering effect; 2) a wider range of frequency response so that can ease the large sampling frequency limitation; 3) better anti-noise capability that expected to perform well in extracting weak line signatures from heavy background noise. Numerical analysis and application verification are performed to confirm the effectiveness and efficiency of the proposed method in comparison with a state-of-the-art second order matched stochastic resonance (SMSR) method, which reflect superior in different contexts.