The Characterization of the Wideband and Ultra-Wideband Mobile Channels and Their Equalization using Neuro-Fuzzy Techniques

Abstract

The 3G mobile and 4G mobile systems have generated ever-increasing interest over the last few years, fuelled by a demand for high frequency utilization and a large number of users requiring simultaneous multidimensional high data rate access for applications of wireless Internet and e-commerce. The characterization of mobile cellular channels and their equalization had been a topic of intense research the world over. We have linear time variant (LTV) models of channels appropriate for different classes/generations of systems like GSM, CDMA, and WCDMA. The need for more bandwidth resulting from the proliferation of users demands better ways to model the mobile channel and mitigate the aggravated co-channel interference (CCI). In this paper we present a novel approach for the modeling of mobile cellular channels in the wideband and ultra wideband frequencies. It should be noted that at the above frequency bands, the channel impulse response is both frequency dependent and time varying. The classical methods support narrow band characterization of these channels especially when parameter spread of channel matrix is less significant. Neuro-fuzzy modeling provides for a higher spread of these parameters making the wideband analysis easier. Therefore the use of Neuro-Fuzzy model for the channel equalizer design is more elegant and simple. Simulations do indicate that this paradigm is indeed valid.