A causal optimal filter of the second order

A. Torokhti1,2, P. Howlett1, C. Pearce2
1Centre for lndustrial and Applicable Mathematics, University of South Australia, SA, Australia
2Applied Mathematics Department, University of Adelaide, SA, Australia

Tóm tắt

We provide a non-linear optimal physically realizable filter which guarantees a smaller associated error than those of the known linear optimal filters proposed by H.W. Bode and C.E. Shannon (see Proc. IRE, vol.38, p.417-25, 1950) and M.V. Ruzhansky and V.N. Fomin (see Bulletin of St. Petersburg University, Mathematics, vol.28, p.50-5, 1995). The technique presented has potential applications to numerous areas in signal processing including, for example, filtering, blind channel equalization, feature selection and classification in pattern recognition, target detection, etc. The technique is based on the best approximation of a stochastic signal by a specific non-linear operator acting on the noisy observed data.

Từ khóa

#Nonlinear filters #Electronic mail #Mathematics #Signal processing #Blind equalizers #Pattern recognition #Stochastic processes #Extraterrestrial measurements #Equations #Digital signal processing

Tài liệu tham khảo

10.1109/81.922461 10.1109/78.917807 10.1080/01630569708816764 10.1109/82.924074 golub, 1990, Matrix Computations howlett, 1999, A best linear estimator for random vectors with values in Hilbert space, Maths Res (St Petersburg), 4, 99 10.1016/S0362-546X(01)00659-9 ruzhaneky, 1995, Optimal Filter Construction for a General Quadratic Cost Functional, Bull St Petersburg Univ Math, 28, 50 torokhti, 0, Method of Best Successive Approximations for Nonlinear Operators, Journal of Computational Analysis and Applications 10.1109/JRPROC.1950.231821