Robust Underwater Acoustic Channel Estimation in Impulsive Noise Environment

Abundant impulsive noise exists in under-ice environment due to ice breaking, ice collision and ice melting phenomena. Conventional channel estimation algorithms are substantially degraded in this impulsive noise environment because of that the Gaussian noise assumption on longer holds. In this paper, a robust underwater acoustic channel estimation algorithm based on Sparse Bayesian learning (SBL) is proposed. The estimated channel is used to estimate and eliminate the impulsive noise iteratively. The proposed method is evaluated by simulations under Gaussian mixture model (GMM) noise and Arctic under-ice noise. The results show that the proposed iterative algorithm has higher estimation accuracy over several conventional methods.