A Low-Complexity Underwater Acoustic Signal Denoising Technique based on Multi-Stage Adaptive Filter Configuration

Underwater communication is highly challenging due to natural and anthropogenic noise in the ocean. The noise-corrupted signal should be denoised before being employed for further processing. This paper introduces an optimal cascaded adaptive filter structure using a Sign Data Least Mean Square (SDLMS) algorithm to denoise a noisy signal with maximum accuracy. Multiple stages are automatically added in the proposed filter structure, and the step size at every stage is varied. The proposed Multistage Sign Data LMS adaptive filter model has been evaluated for denoising an underwater fin whale sound signal corrupted by underwater vessel noise taken from the ShipsEar database. The simulation results prove that the proposed filter model performs remarkably well and offers a cost-effective hardware implementation of Adaptive Noise Canceller (ANC).

Automated summary

Underwater communication is challenging due to ocean noise. This paper proposes a cascaded adaptive filter structure using the SDLMS algorithm for denoising signals. Simulation experiments on underwater fin whale sound corrupted by vessel noise demonstrate the remarkable performance and cost-effectiveness of the proposed filter model.