Tracking performance of incremental augmented complex least mean square adaptive network in the presence of model non‐stationarity

Abstract

This study addresses the tracking performance of the incremental augmented complex least mean square (IAC-LMS) algorithm, operating in the presence of model non-stationarities. The authors consider the mean-square deviation and excess mean square error as performance metrics and use energy conservation argument to derive closed-form expressions for the mentioned metrics. The expression describes how the IAC-LMS algorithm performs under such non-stationary conditions. The authors further find the step size range where the mean-square stability of the IAC-LMS algorithm is guaranteed. The authors provide some simulation results to support the theoretical derivations.