Title: Stochastic MIMO channel modelling using FMLP-based inference engine

Authors: Kandarpa Kumar Sarma; Abhijit Mitra

Addresses: Department of Electronics and Communication Technology, Gauhati University, Guwahati-781014, Assam, India ' Department of Electronics and Electrical Engineering (EEE), Indian Institute of Technology Guwahati (IITG), Guwahati-7810139, Assam, India

Abstract: A fuzzy-neural (FN) estimator of stochastic multi-input-multi-output (MIMO) wireless channels shows dependence on the membership and inference rule generation norms adopted. These two factors decide the ability of the fuzzy-estimator to capture the subtle variations in the input signal patterns and provide corresponding responses with optimum performance. They also determine the precision and processing speed of the FN-estimator while tackling stochastic behaviour of the MIMO channels. The membership and inference rule generation norms must not only capture the subtle variations but also should contribute towards lower bit error rate (BER), reduced design and time complexity. Here, we propose the design of a fuzzy multilayer perceptron (FMLP)-based inference engine design for a FN-based MIMO modelling using multiple membership and inference rule generation norms. Experimental results derived show that a set formulated with seven linguistic hedges and six inference states helps in designing a FN MIMO estimator which can be an important element in the design of adaptive receivers for high data rate wireless communication.

Keywords: multi-input-multi-output; MIMO channel modelling; stochastic modelling; fuzzy multilayer perceptron; FMLP; inference engine; artificial neural networks; ANNs; recurrent neural networks; self-organising maps; optimisation; fuzzy logic; fuzzy-neural estimation; wireless communications.

DOI: 10.1504/IJICT.2013.053112

International Journal of Information and Communication Technology, 2013 Vol.5 No.2, pp.122 - 136

Received: 07 Mar 2012
Accepted: 05 Sep 2012
Published online: 19 Dec 2013 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article