Blind GSM channel estimation under channel coding conditions

Abstract

As a fundamental component of the global system for mobile communications (GSM), channel coding aims at improving speech and data transmission quality when the signal encounters disturbances. We put the main emphasis on investigating the application of blind channel estimation approaches based on second order statistics and/or higher order statistics to the identification of a full rate data traffic channel (TCH/F9.6). Giving the bit error rates after equalization and channel decoding in terms of the E~/sub b//N/sub 0/ ratio, we show that there are blind channel estimation algorithms which are almost as efficient as non-blind methods even if the disturbance is non-Gaussian.