Abstract
Performance of turbo codes in fading channels can be improved when the channel fading amplitudes are available at the turbo decoders. Conventionally, channel estimation and turbo-decoding are treated as two independent tasks. In this paper, we propose an iterative structure for fading amplitude estimation where turbo-decoder decisions are directed to an adaptive channel estimator using the general optimum block adaptive (GOBA) algorithm. The performance of a turbo-coding system with the proposed channel estimation structure is also investigated through computer simulations for correlated Rayleigh flat-fading channels. This decision-directed approach for channel re-estimation combined with the excellent tracking capability of the GOBA algorithm significantly improves the turbo-coding system performance. Computer simulation results show that its bit error rate performance is close to that of a system with perfect knowledge of fading amplitudes especially for fast-fading channels.