Leaky Least Mean Mixed Norm Algorithm

Abstract

In this paper, the leakage-based variant of the Least Mean Mixed Norm (LMMN) algorithm, the leaky Least Mean Mixed Norm (LLMMN) algorithm, is derived.The proposed algorithm will help mitigate the weight drift problem expe-rienced in the conventional Least Mean Square (LMS) and Least Mean Fourth (LMF) algorithms.The aim of this paper is to derive the LLMMN adaptive algorithm and perform the transient analysis using the energy conservation relation framework. Finally, simulation results are carried out to support the theoretical findings, and show improved performance obtained through the use of LLMMN over the conventional LMMN algorithm in a weight drift environment.