Generalization of channel estimation weights for multiple differential detection based on per-survivor processing

Abstract

This paper proposes novel channel estimation weights for multiple differential detection based on per-survivor processing (PSP-MDD). Channel estimation weights for PSP-MDD can control a trade-off between performance in the required signal-to-noise power ratio (SNR) and tracking capability. There have been proposed the two channel estimation weights: weights of simple first-order channel prediction and weights of higher-order channel prediction based on the binomial theorem. In order to control this trade-off widely and precisely, this paper proposes channel estimation weights generalizing the two conventional weights. Finally, computer simulation results confirm that the proposed weights can control the trade-off widely and precisely.