An Anti-reverberation Approach for wideband Sonar: Based on Sparse Processing

Reverberation is the main factor that limits the detection of close-range targets by an active sonar system. In this paper, a sparse processing method of dynamic measurement vector is proposed to realize the parameter reconstruction of reverberation and target for broadband sonar scenario. In this method, the high-resolution parameters are reconstructed one by one along the coarse resolution range bin in the reverberation region, and a tolerance term is added in the iterative process of reconstruction to distinguish reverberation and target. The measurement vector is updated synchronously during the phase atom search of the dictionary matrix, which effectively overcomes the range migration of the target and reverberation. Simulation results show that the proposed method can achieve accurate reconstruction and detection of target parameters in a reverberation background.