Adaptive SNR estimation based decoding algorithm for block turbo codes

The SNR values in conventional turbo decoding with the adaptive chase algorithm are pre-fixed, which is not practical in applications. To get rid of the deficiency, a novel decoding algorithm for block turbo codes was proposed, in which the channel levels of different SNR values were adaptively estimated. The proposed algorithm could tune the decoding complexity by using the relationship between the statistics of the received vectors and the channel SNR values in order to change the threshold function. Simulation results show that the proposed algorithm decreases the decoding complexity, speeds up the decoding process while achieving better BER performance. Furthermore, the proposed decoding algorithm can adapt to different SNR values without pre-fixing them.