A solution space principal component based adaptive filter

This paper introduces a new type of adaptive filter called PCP, principal component proportionate normalized least mean squares. It is an extension of PNLMS (proportionate normalized least mean squares), an adaptive filter that has been shown to provide exceptionally fast convergence and tracking when the underlying system parameters are sparse (as in network echo cancellation). PCP extends the application of PNLMS to certain nonsparse systems by applying it while using the principal components of the underlying system as basis vectors. Room acoustic echo systems are possible examples of such nonsparse systems. Simulations of acoustic echo paths and cancellers indicate that PCP converges and tracks much faster than the classical normalized least mean squares (NLMS) and fast recursive least squares (FRLS) adaptive filters.