Decision-feedback differential detection based on linear predictionfor MDPSK signals transmitted over Ricean fading channels

摘要

In this paper, linear prediction-based decision-feedbackndifferential detection (DF-DD) for M-ary differential phase-shift keyingn(MDPSK) signals transmitted over Ricean fading channels is proposed.nThis scheme can improve conventional DD significantly for a multitude ofnfrequency-nonselective channels, as shown analytically and by computernsimulations. Prediction-based DF-DD is particularly well suited fornapplication in mobile communications since the predictor coefficientsnmay be updated regularly using the recursive least squares (RLS)nalgorithm. Here, adaptation can start blind, i.e., no training sequencenand no a prior knowledge about the channel statistics are required. Anfurther important characteristic of the proposed detection scheme isnthat no degradation occurs under frequency offset. The bit error raten(BER) performance of QDPSK with genie-aided prediction-based DF-DD isnanalyzed, and it is shown under which conditions the irreducible errornfloor of conventional DD can be removed entirely. In addition, theninfluence of Doppler shift is discussed. Last, the proposed scheme isncompared with a second DF-DD scheme, which is based on multiple-symbolndetection