Services on Demand
Journal
Article
Indicators
Cited by SciELO
Access statistics
Related links
Cited by Google
Similars in SciELO
Similars in Google
Share
Ingeniería y Universidad
Print version ISSN 0123-2126
Abstract
MELGAREJO-REY, Miguel; GAONA-BARRERA, Andrés and BARRETO-SUAREZ, Carlos. Adaptive Fuzzy Equalization Based on Neuron Grouping for Time-Varying Non-Linear Channels. Ing. Univ. [online]. 2011, vol.15, n.2, pp.423-443. ISSN 0123-2126.
This paper presents an approach for time varying non-linear channel equalization based on fuzzy systems and single-neuron training. The method consists of two stages: the first one uses supervised learning in order to determine channel states and to provide an initial tuning of the fuzzy equalizer parameters. The second one dynamically adjusts the equalizer to follow the varying behavior of the channel through unsupervised learning. This proposal is compared with a radial basis network over the equalization of a time-varying communication channel reported in previous works. Experiments are carried out through Monte Carlo simulations. Results show that the proposed approach presents a performance than that of a radial basis function in terms of the bit error rate of a communication system.
Keywords : Digital communications; equalizers (electronics); adaptive filters; neural networks (computer science); fuzzy system.