| The objective of this research is to develop new soft output adaptive equalization techniques derived from trellis based soft output decoding algorithms.; We propose a receiver employing the soft output equalization concatenated with the soft decision Viterbi decoder and a receiver employing the iterative decoding process with the concatenation of the soft output equalization and the soft output decoder. Soft output equalization and soft output decoding techniques are derived from trellis based soft output decoding algorithms such as SOVA, SSA, the modified-MAP algorithm, and the MAP algorithm.; In this research, a TDMA cellular system using the GMSK modulation technique is proposed as a test bench for the evaluation of the proposed algorithms. The general framework of soft output equalization and turbo equalization is investigated. Modifications of suboptimum soft output decoding algorithms to be used for soft output equalization and turbo equalization are investigated. Performance of iterative decoding over continuous time varying channels with channel estimation errors is investigated. Generalization of the algorithms and complexity analysis are investigated. Proposed techniques are tested over three GSM channels by computer simulations. |
