Skip to Main Content
Estimation of time varying systems can be done by representing the whole system behaviour in terms of some fixed basis sequences. Available solutions use computationally expensive nonrecursive solutions. We derive an RLS solution for the wavelet based expansion of time varying systems. This method has been applied to speech signals and its performance is studied. A recursive lattice algorithm is derived for the above modeling. Also the performance of the method using both the Haar and Daubechies bases is studied.
Date of Conference: 6-10 April 2003