ADAPTIVE IIR FILTERING USING QR MATRIX DECOMPOSITION

In this correspondence, an approach to adaptive IIR filtering based on a pseudo-linear regression and applying an iterative QR matrix decomp osition is developed. In simulations, the algorithm has shown a high s tability and excellent convergence properties. The derivation of the I IR-QR adaptive filter is straightforward, and the computational comple xity of the algorithm is comparable with the FIR-QR adaptive algorithm of the same order and less than gradient-based adaptive IIR filters s uch as the simplified-gradient recursive prediction error (RPE) algori thm. Fast versions of O(N) computational complexity are readily availa ble.