A NEW PREFILTER DESIGN FOR DISCRETE MULTIWAVELET TRANSFORMS

In conventional wavelet transforms, prefiltering is not necessary due to the lowpass property of a scaling function. This is no longer true for multiwavelet transforms. A few research papers on the design of pr efilters have appeared recently, but the existing prefilters are usual ly not orthogonal, which often causes problems in coding. Moreover, th e condition on the prefilters was imposed based on the first-step disc rete multiwavelet decomposition. In this paper, we propose a new prefi lter design that combines the ideas of the conventional wavelet transf orms and multiwavelet transforms. The prefilters are orthogonal but no nmaximally decimated. They are derived from a very natural calculation of multiwavelet transform coefficients. in this new prefilter design, multiple step discrete multiwavelet decomposition is taken into accou nt, Our numerical examples (by taking care of the redundant prefilteri ng) indicate that the energy compaction ratio with the Geronimo-Hardin -Massopust 2 wavelet transform and our new prefiltering is better than the one with Daubechies D-4 wavelet transform.