Turbo codes conventionally adopt modified BAHL et al.algorithm
a kind of probabilitic iterative decoding algorithm based on soft values.To fully make use of the information on the deterministic structure of turbo constituent encoders according to the algebraic decoding principle in conjunction with the probabilitic decoding algorithm
this paper proposed a joint probability-algebra decoding scheme for turbo codes.Combining the advantages of iterative soft decoding and algebraic decoding and efficiently avoiding the bit error propagation
the proposed scheme is capable of further improving the error-correcting performance of turbo codes.It is not only a new research method to lower the Bit Error Rate (BER) and the error floor of turbo codes but valuable for applications as
a result of better error correcting performance.Simulation results show that the turbo codes can obtain about 0.1dB asymptotic coding gain with the proposed scheme than those with the typical turbo-code decoding schemes when the BER is between 10