Performance Analysis of Long Range Prediction for Fast Fading Channels

– Long range prediction capability for fading channels would provide enabling technology for accurate power control, reliable transmitter and/or receiver diversity, more effective adaptive modulation and coding and improvements in many other components of wireless systems. In order to achieve accurate long range prediction, the authors recently introduced a novel algorithm that finds the linear Minimum Mean Squared Error (MMSE) estimate of the future fading coefficient given a fixed number of previous observations. In this paper, we show that the superior performance of this algorithm is due to its lower sampling rate relative to the conventional (data rate) methods of fading estimation. We present the theoretical analysis of the MMSE of long range prediction as a function of several parameters: the number of scatterers, model order, sampling rate and the Signal-to-Noise ratio (SNR). Moreover, we show that the long range prediction can be further improved by employing adaptive tracking combined with the truncated channel inversion algorithm.