Proportionate Adaptive Filtering under Correntropy Criterion in Impulsive Noise Environments

TL;DR

This paper introduces an improved proportionate adaptive filter based on the Maximum Correntropy Criterion (IP-MCC) for system identification in impulsive noise environments, combining robustness and efficiency.

Contribution

It proposes a novel IP-MCC algorithm that enhances convergence speed and reduces computational complexity while effectively handling impulsive noise.

Findings

IP-MCC achieves similar steady-state EMSE as MCC.

It outperforms existing algorithms in convergence and robustness.

Requires less computational effort than comparable methods.

Abstract

An improved proportionate adaptive filter based on the Maximum Correntropy Criterion (IP-MCC) is proposed for identifying the system with variable sparsity in an impulsive noise environment. Utilization of MCC mitigates the effect of impulse noise while the improved proportionate concepts exploit the underlying system sparsity to improve the convergence rate. Performance analysis of the proposed IP-MCC is carried out in the steady state and our analysis reveals that the steady state Excess Mean Square Error (EMSE) of the proposed IP-MCC filter is similar to the MCC filter. The proposed IP-MCC algorithm outperforms the state of the art algorithms and requires much less computational effort. The claims made are validated through exhaustive simulation studies using the correlated input.