AN ADAPTIVE RECURSIVE FILTER FOR AUTOREGRESSIVE INPUTS
Abstract:
A novel o-line procedure for IIR system identi cation problems has been recently proposed and analysed under the assumption of a white input signal. The method happens to have a single, unbiased stationary point under this condition. This approach is based in interpolation theory and tries to overcome the drawbacks of equation- and output-error strategies. Here it is shown that the unicity of the stationary point is still preserved for a wider class of input signals, namely autoregressive processes of a certain order. With coloured input signals, an SPR condition arises for the convergence of the original on-line algorithm. A modi cation is proposed that eliminates this SPR condition, while leaving the stationary point untouched. Two di erent structures that implement the on-line algorithm are presented; the use of a lattice structure for the implementation of the recursive part of the lter is considered as well.? 1998 John Wiley & Sons, Ltd. Key words: adaptive IIR lters; system identi cation; interpolation theory; lattice lters
Citations
| 196 | The Theory of Matrices – Lancaster, Tismenetsky - 1985 |
| 30 | Adaptive IIR Filtering – Regalia - 1995 |
| 7 | Adaptive Filter Theory. Prentice-Hall, Englewood Cliffs – Haykin - 1985 |
| 4 | An unbiased equation error identi er and reduced-order approximations – Regalia - 1994 |
| 3 | Adaptive IIR ltering – Shynk - 1989 |
| 3 | Theory and Design of Adaptive Filters – Larimore - 1987 |
| 3 | On the stability of dynamic models obtained by least squares identification – Söderström, Stoica - 1981 |
| 2 | A new adaptive IIR lter – Fan, Jenkins - 1986 |
| 2 | New results and theoretical analysis of the master-slave family of recursive identi cation algorithms – Lopez-Valcarce, Perez-Gonzalez - 1997 |
| 1 | Undermodeled adaptive ltering: an a priori error bound for the Steiglitz-McBride method – Regalia, Mboup - 1996 |
| 1 | Lattice lters for adaptive processing’, Proc – Friedlander - 1982 |
