A Robust Generalized Maximum Correntropy Criterion Algorithm for Active Noise Control

Abstract The main content of noise control for vehicle is to restrain the noise in the cab. As a new method for noise reduction, active noise control (ANC) gradually fills in the blank part that high frequency noise could not be eliminated by the traditional passive noise cancellation (PNC) method. In this brief, a robust generalized maximum correntropy criterion (FxGMCC) algorithm for ANC controller is proposed against impulsive input. The generalized maximum correntropy criterion (FxGMCC) algorithm adopts a more flexible generalized Gaussian density (GGD) function as kernel, it performs better than ordinary maximum correntropy criterion (MCC) algorithm and with strong robust property. To demonstrate the great performance of the proposed algorithm, computational complexity is analyzed and comparison of simulations between existing algorithms in ANC model is carried out under three different intensity cases of impulsive noise.