Title :
The gradient adaptive split lattice algorithm
Author :
Wu, Chi-Hsin ; Yagle, Andrew E.
Author_Institution :
Dept. of Electr. & Comput. Sci., Michigan Univ., Ann Arbor, MI, USA
Abstract :
A new algorithm for adaptive filtering of autoregressive processes is presented. The new algorithm is based on a three-term recurrence that replaces, but is mathematically equivalent to, the lattice equations of the gradient adaptive lattice (GAL) algorithm of Griffiths (1977). This three-term recurrence requires only half as many multiplications per recursion as do the lattice equations. A single-stage convergence analysis of the new algorithm is presented, and some numerical examples comparing it to the GAL are given. The new algorithm performs similarly to the GAL, but requires only 2/3 as many multiplications
Keywords :
adaptive filters; convergence of numerical methods; digital filters; filtering and prediction theory; recursive functions; statistical analysis; adaptive filtering; autoregressive processes; gradient adaptive split lattice algorithm; single-stage convergence analysis; three-term recurrence; Adaptive algorithm; Adaptive filters; Algorithm design and analysis; Convergence of numerical methods; Difference equations; Filtering algorithms; Lattices; Reflection; Signal analysis; Signal processing algorithms;
Conference_Titel :
Acoustics, Speech, and Signal Processing, 1992. ICASSP-92., 1992 IEEE International Conference on
Conference_Location :
San Francisco, CA
Print_ISBN :
0-7803-0532-9
DOI :
10.1109/ICASSP.1992.226447
