Design of discrete coefficient FIR and IIR digital filters withprefilter-equalizer structure using linear programming

摘要

Optimal methods for designing multiplierless finite-impulsenresponse (FIR) and infinite-impulse response (IIR) filters with cascadednprefilter-equalizer structures are proposed. Assuming that an FIR filternconsists of a cyclotomic polynomial (CP) prefilter and an interpolatednsecond order polynomial (ISOP) equalizer, in the proposed method, thenprefilter and equalizer are simultaneously designed using mixed integernlinear programming (MILP). The resulting filter is a cascaded filternwith minimal complexity. For IIR filters, all-pole IIR equalizersnconsisting of inverse of interpolated first order polynomials (IIFOP's)nare introduced, and a CP-prefilter cascaded with this type of equalizernis designed. Design examples demonstrate that the proposed methodsnproduce a more efficient cascaded prefilter-equalizer than existingnmethods