An RLS type algorithm for generalized eigendecomposition

Abstract

Eigendecompositions play a very important role in a variety of signal processing applications. The authors derive and study an algorithm for Generalized Eigendecomposition which is both online and fast converging. A rule to extract the maximum eigencomponent is first presented, and then online deflation is applied to estimate the minor components. Proof of convergence has been established using stochastic approximation theory.