On Reduced Complexity IIR Adaptive Filters

M.J. BRUNO; J.E. COUSSEAU; P.D. DOÑATE

Kobe

Congreso; International Symposium ob Circuits and Systems; 2005

IEEE Circuits and Systems Society

Reduced complexity IIR adaptive filters, like Laguerreadaptive filters with pole updating are being consideredas an alternative to full adaptive IIR schemes. On the other hand,low complexity realizations exist that allow a better compromisebetween filter-order and modeling performance. We discuss inthis work characteristics and properties of a reduced complexityIIR orthogonal realization that improve those of the Laguerreand sub-optimal IIR counterparts.After reviewing the Laguerre and sub-optimal adaptive filterswith pole updating, we introduce a reduced complexity IIRadaptive filter based on the Steiglitz-McBride method. The particularrealization proposed is based on a generalized orthonormalfamily of functions. The Steitglitz-McBride method is discussedmainly because it can avoid the local minima problem related todirect mean squared output error minimization.Computer simulations are given to illustrate the good performanceof the proposed adaptive IIR filter with respect tocomputational complexity and modeling capability if comparedwith their Laguerre counterpart.