Reduced-complexity iterative decoding using a variable-size trellis structure

We propose a new reduced-complexity iterative decoding algorithm. The main concept is to approximate the aposteriori probabilities (APP) by a set of few sequences, and to reduce the size of this set with iterations. According to the numerical results on the bit error rate (BER) performance and the convergence behavior by extrinsic information transfer (EXIT) chart analysis, the algorithm could obtain a good balance between performance and complexity.