PSO tuned ANFIS equalizer based on fuzzy C-means clustering algorithm


AEU-INTERNATIONAL JOURNAL OF ELECTRONICS AND COMMUNICATIONS, vol.70, no.6, pp.799-807, 2016 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 70 Issue: 6
  • Publication Date: 2016
  • Doi Number: 10.1016/j.aeue.2016.03.006
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Page Numbers: pp.799-807
  • Keywords: Channel equalization, Clustering, ANFIS, PSO, Wireless communications, PARTICLE SWARM OPTIMIZATION, CHANNEL EQUALIZATION
  • Karadeniz Technical University Affiliated: Yes


Wireless communication always suffers from intersymbol interference (ISI), multipath fading, co-channel interference (CCI). Equalization techniques try to combat these phenomena to recover distorted data at receivers. Most techniques need a separate process to estimate channel profile before equalization. In this paper, Particle Swarm Optimization tuned Adaptive Neuro Fuzzy Inference System (PSO-ANFIS) based channel equalizer, which is capable of system identification, estimation and equalization of wireless communication channels, is proposed. PSO-ANFIS equalizer uses the training data and employs fuzzy C-means (FCM) clustering to model a wireless communication channel without knowledge of channel dynamics. The PSO-ANFIS equalizer was simulated on a mobile communication model with ISI, CCI and the Additive White Gaussian Noise. The used training method and FCM provided the best regression of system modeling to fit to wireless channel. The performances of PSO-ANFIS equalizer were evaluated and compared to the equalizers of Maximum-Likelihood Sequence Estimation, Recursive Least Squares, ANFIS, Functional Link Artificial Neural Network, PSO-Variable Constriction Factor and Artificial Neural Network-PSO in terms of BER-SNR. The simulation results showed that the performance of the PSO-ANFIS equalizer with FCM clustering gives the best performance among the mentioned nonlinear equalization techniques. (C) 2016 Elsevier GmbH. All rights reserved.