Analysis of Adaptive Threshold Nonlinear Correlation Algorithm

This paper proposes an adaptation algorithm named Adaptive Threshold Nonlinear Correlation Algorithm (ATNCA) which makes adaptive filters robust against two types of impulse noise: impulsive observation noise at the filter output and impulse noise at the filter input. Analysis of the ATNCA is developed to theoretically calculate filter convergence behavior. Through experiments, we demonstrate the effectiveness of the proposed algorithm in realizing fast convergent and robust adaptive filters in impulsive noise environments. Good agreement between simulated and theoretical filter convergence curves shows the validity of the analysis.