Decision feedback approach to blind algorithms in impulsive noise environments

Abstract

In order to cope with impulsive noise and error propagation problems without any additional schemes, new decision feedback algorithms with inherent robustness against impulsive noise are introduced in this paper. The Gaussian kernel with a mathematically constructed information potential makes the algorithms insensitive to large differences between symbol points and outputs induced by impulsive noise. This property effectively prevents decision feedback strategies from incorrect decisions and error propagation. From the simulation results, the proposed strategy has yielded significant performance enhancement of 7 dB over the previously proposed scheme.