Turbo joint multiuser detection and channel estimation for space-time turbo coded TDMA systems

We consider multiple-input multiple-output (MIMO) downlink TDMA systems employing space-time turbo codes (STTuC) over the multipath fading channel. The mixture of intercell interference (ICI), inter-antenna interference (IAI) and intersymbol interference (ISI) can prevent the receiver from achieving a satisfactory detection/estimation performance. Here we develop an iterative receiver structure that can effectively detect the wanted signal by simultaneously suppressing all types of interference and estimating the channels in this complex interference environment. The proposed receiver follows the ´turbo principle´ and is based on parallel interference cancellation (PIC) and MMSE joint channel estimation (MJCE). In each iteration, the soft output of the space-time turbo decoder is fed back to improve the performance of the PIC and MJCE. Since symbol-level synchronization between base stations is unachievable, the ´optimal´ MIMO training sequences cannot achieve optimal performance. The simulation shows that, in this case, the iterative detection/estimation yields a substantial performance gain over both the iterative detection/one-shot estimation and the RAKE reception/iterative estimation.