Impulsive noise cancellation using CFAR and iterative techniques

In this paper we propose a new method to recover bandlimited signals corrupted by impulsive noise. The proposed method uses CFAR adaptive thresholding and soft decision making to detect the locations of the impulsive noise. Then, an iterative method cancels noise and reconstructs the signal. This estimate in turn will be used to make the approximation of the impulsive noise more accurate. Simulation results confirm the robustness of the proposed algorithm even when impulsive noise exceeds the theoretical reconstruction capacity.