Abstract
Tracking properties are considered for adaptive filters which are adjusted within a vector space of filtering operations. For a given vector space of systems, the adaptive filter structure is determined by a choice of basis for the vector space. Basis dependent expressions are developed for the asymptotic mean square error (MSE) under least-mean-square (LMS) and recursive-least-square (RLS) adaptation, when the optimal filter specification is subject to constant drift. For the idealized setting considered, it is shown that the minimum achievable MSE using LMS adaptation is less than or equal to that achievable under RLS adaptation.

This publication has 9 references indexed in Scilit: