A new class of sinusoidal-preserving FMH filters
J. Neejärvi, S. Fotopoulos, Yrjo A. Neuvo · 1991
A new class of FIR (finite impulse response) median hybrid filters for nonlinear signal processing is proposed. The proposed filters, five-point weighted FIR median hybrid (WFMH) filters, consist of combinations of averaging and ramp predicting FIR substructures, the input signal, and the median operation. The central weighted FMH (CWFMH) filter structure is analyzed in detail. Due to the selection of subfilters the CWFMH filter has low-pass characteristics and preserves sinusoidal signals. It is shown that sinusoidal signals are root signals of the filter. An upper bound for the preservation of sinusoids is derived. In addition, the root length for a rectangular pulse is derived. The behavior of the CWFMH filter around noisy edges is also studied. An efficient implementation structure for the WFMH filter with complexity independent of the filter length is shown.>