Convergence analysis of Maximum Correntropy Criteria based adaptive filtering algorithm based on white input

Maximum Correntropy Criterion (MCC) based adaptive filters had received much attention due to its robustness against impulsive noise. In this paper the convergence analysis of MCC adaptive filter based on white input is performed. The condition for stability is analyzed and the steady state mean square error (MSE) and mean square deviation (MSD) error is derived in terms of step size, variance of noise source, length of the system and kernel width. Moreover a criteria for parameter selection to obtain improved performance of MCC adaptive filter is also proposed. Simulations in the context of unknown system identification scenario were performed to prove the validity of the theoretical analysis made.