Computationally Efficient Fuzzy Adaptive Equalizer for Nonlinear Channels

Abstract

Fuzzy adaptive equalizers show a good performance in channels with highly nonlinear decision boundaries due to their universal function approximation capabilities. However, the computational load for the parameter adaptation is much higher than that of linear equalizers. This article considers a new approach to fuzzy adaptive equalizers to reduce the computational complexity. The approach in this article has the distinctive feature of conditional linear equalizations based on a perceptron-based filter. The premise part is automatically constructed by a competitive self-organizing scheme using decision feedback vectors as well as observed channel vectors. The corresponding resultant part is realized in the form of a perceptron-based filter for the observed channel vectors. As a result, the proposed equalizer shows a good performance with a computational load similar to that of linear equalizers at each symbol time. The effectiveness of our approach is shown through computer simulations of bit error rate performances and computational complexity.