Sub-space based sparse channel estimation method with improved performance under noisy environment

Due to the time-varying nature of the channel and the non-Gaussian distribution of noise, underwater acoustic communication is extremely difficult. In recent years, a variety of techniques have been proposed that have had some success in establishing a communication link between two water submerged bodies, but it remains an open challenge for researchers. Underwater channel estimate is one of these issues. Since the nature of underwater channels is sparse, there are a variety of sparse channel estimation approaches available in the literature. Because these strategies do not account for noise in their mathematical models, they do not function well in underwater audio communication. We propose a technique that combines noise reduction and sparse channel estimation in this paper. We use Singular Value Decomposition (SVD) for noise reduction and Orthogonal Matching Pursuit (OMP) for sparse channel estimation in the proposed technique. In this paper, the suggested algorithm is compared against OMP in the presence of AWGN noise.