Abstract

Nonlinear matched filtering involves the transformation of the signal spectrum and the filter transfer function through a nonlinearity before they are multiplied in the transform domain.¹ The resulting filter structures can be considered to be analogous to three-layer neural nets. They have better performance in terms of signal discrimination and lack of false correlation signals and artifacts than previously known filters. Because of nonlinearities, the analysis of nonlinear matched filters requires new approaches different from techniques valid in linear systems. We present recent results on analysis of nonlinear matched filters as well as their generalization to incorporate maximum clustering within class and maximum separation between classes in pattern recognition applications.

© 1989 Optical Society of America