Decision Feedback Equalizers Using Self-Constructing Fuzzy Neural Networks

A self-constructing fuzzy neural network decision feedback equalizer (SCFNN DFE), which does not have to estimate the channel order first, is proposed in this paper. An online learning, where the structure and the parameter learning phases are performed concurrently, is used in SCFNN. Specifically, structure and parameter learning phases respectively based on the partition of input space and the gradient method are also described. The performance of SCFNN DFE is compared with the traditional nonlinear equalizers. The reduced complexity and high performance of the SCFNN DFE makes it suitable for high-speed channel equalization.