Adaptive Arma Lattice Based On Generalized Mullis-roberts Criterion
Donald W. Mennecke, M. Tummala · 2005
In this paper, an adaptive algorithm is presented for an ARMA digital lattice filter, whose parameters are estimated based on a formulation using generalized Mullis-Roberts criterion. The ARMA digital lattice filter is obtained directly from independent AR and MA prediction error recursion formulas which evolve from a parameter estimation algorithm based on the generalized criterion. The adaptive lattice algorithm is based on the well known LMS method of error minimization. Experimental results obtained from applying this adaptive algorithm to several system identification type modeling problems are shown.