Transmission rate prediction for Cognitive Radio using Adaptive Neural Fuzzy Inference System

Advances in applications demanding high data rate wireless applications and existing wireless system upgrading has lead to scarcity in spectrum. Unlicensed new technologies like Digital video broadcast (DVB), Digital audio broadcast (DAB), internet, WiMAX etc. launched recently are reaching thousands of customers at rapid speed. Most of the primary spectrum is assigned, so it is becoming very difficult to find spectrum for either new services or expanding existing infrastructure. Present government policies do not allow unlicensed access of licensed spectrum, constraining them instead to heavily populated, interference-prone frequency bands. Cognitive Radio systems promise to handle this situation by utilizing intelligent software packages that enrich their transceiver with radio-awareness, adaptability and capability to learn. In this paper, we present the working of the fifth generation intelligent radio that is Cognitive Radio (CR) system which works on predictive data rate and propose ANFIS based learning scheme to introduce intelligence in it. The performance of this is seen to be comparable to neural network based scheme with reduced complexity.