Abstract
Symmetric nonlinear matched filters (SNMF’s) involve the transformation of the signal spectrum and the filter transfer function through pointwise nonlinearities before they are multiplied in the transform domain. The resulting system is analogous to a 3-layer neural net The experimental and theoretical results discussed indicate that SNMF’s hold considerable potential to achieve high-power of discrimination, resolution and large SNR. The statistical analysis of a particular SNMF in the 2-class problem indicates that the performance coefficient of the SNMF is about four times larger than the performance coefficient of the classical matched filter. In terms of resolving closeby signals, there seems to be no limit to die achievable resolution. However, intermodulation noise has to be carefully monitored.
Date of this Version
3-1-1990