Improvement on turbo code decoder by updating redundant likelihood information

The paper presents a novel decoding algorithm for turbo codes, in which the likelihood values for redundant parts are updated in order that those values become more reliable. A criterion for updating the redundant likelihood values is proposed; it is based on comparisons of the channel values with the values re-generated by the soft-input and soft-output encoders. It is shown that the proposed method can improve the error correcting capabilities, i.e., the improvement of BER/BLER performance and the achievable BER limit.