MULTICHANNEL L-FILTER BASED ON REDUCED ORDERING

Nonlinear multichannel signal processing is an emerging research topic with numerous applications, In this paper we use the so-called reduce d ordering (R-ordering) principle to introduce a new family of L filte rs for vector-valued observations, The coefficients of the proposed fi lters can be deduced so that the filters are optimal with respect to t he output mean squared error, Expressions for the unconstrained, unbia sed and location invariant optimal filter coefficients are derived, Th e calculation of moments of the R-ordered vectors that are involved in these expressions is also discussed, Experiments with noisy two-chann el vector fields and noisy color images are presented in order to demo nstrate the superiority of the proposed filters over other multichanne l filters.