Block-sparse enhancement and detection of envelope modulation on noise for ship radiated noise

Underwater target recognition is essential for ocean monitoring, with ship radiated noise analysis being a fundamental technique. This paper proposes a detection method based on block sparse enhancement. A harmonic dictionary is constructed and block sparse reconstruction is applied to enhance line spectrum components in the Detection of Envelope Modulation on Noise spectrum. A detection strategy that integrates harmonic structure and fundamental frequency continuity is designed. Simulations and SWellEx-96 data demonstrate that the proposed method achieves better clarity, higher signal-to-noise ratio (SNR) gain, and improved stability under low SNR conditions compared to traditional approaches.