Variable-Parameter Arctangent-Cost-Function Based NLMS Algorithm Robust Against Impulsive Interferences

In this paper, a variable-parameter arctangent-cost-function based NLMS (VP-Arc-NLMS) robust against impulsive interferences is proposed to solve the trade-off between tracking capability and low misadjustment. In particular, the energy of the filter update is controlled to be constrained by the maximum of the variable-parameter cost function. Finally, improved convergence and MSD performances of the proposed VP-Arc-NLMS are verified by considering system identification in an impulsive interference environment.