The weak modulation line spectrum enhancement and detection method for ship radiated noise in the absence of prior information

With the development of vibration and noise reduction technology, the radiated noise of underwater ship targets is decreasing. To achieve underwater target detection under low signal-to-noise ratio conditions, this paper proposes a line spectrum detection algorithm based on cascaded bistable stochastic resonance and time-domain synchronous averaging (CBSRTSA). At the first step, integrated processing of the first-stage bistable stochastic resonance and time-domain synchronous averaging is employed to determine a possible set of line spectrum frequencies in the absence of prior frequency knowledge. This approach effectively enhances harmonic line spectra and reduces the impact of stochastic resonance on low-frequency noise. Subsequently, a comprehensive index based on spectrum kurtosis and information entropy is proposed for effective estimation of line spectrum frequencies. Additionally, the parameters of the second-stage bistable stochastic resonance system are optimized based on the estimated line spectrum frequencies. Cascade stochastic resonance systems further achieve line spectrum enhancement and waveform smoothing. Finally, a CSRTSA-based detection algorithm is proposed for line spectrum detection. The effectiveness of the proposed method for weak line spectrum enhancement is verified by numerical simulation and measured data, respectively, and the performance of the proposed method is compared with other line spectrum detection methods.