Performance of turbo codes in interleaved flat fading channels with estimated channel state information

This paper investigates the performance of a turbo coded system transmitting over correlated flat fading channels with channel interleaving and both perfect and estimated channel state information. We show that channel interleaving is necessary on a correlated fading channel, and that the performance degrades as the product of fading bandwidth and symbol duration (BT/sub s/) decreases. We consider both Rayleigh and Rician fading, and compare the performance of log-maximum a posteriori (log-MAP) and soft output Viterbi algorithm (SOVA) based decoding algorithms. It is shown that for fading channels, the performance of the log-MAP algorithm is considerably superior to that of the SOVA algorithm. A simple method for estimating the fading amplitudes and noise variance is proposed, and the impact of its use is investigated. It is shown that estimating the fades degrades the performance only slightly, and estimating the noise variance does not noticeably affect the performance.