It is known that Block Turbo Codes （BTC） can be nearly optimally decoded by Chase-II algorithm, in which the Least Reliable Bits （LRBs） are chosen empirically to keep the size of the test patterns （sequences） re...It is known that Block Turbo Codes （BTC） can be nearly optimally decoded by Chase-II algorithm, in which the Least Reliable Bits （LRBs） are chosen empirically to keep the size of the test patterns （sequences） relatively small and to reduce the decoding complexity. While there are also other adaptive techniques, where the decoder＇s LRBs adapt to the external parameter of the decoder like SNR （Signal Noise Ratio） level, a novel adaptive algorithm for BTC based on the statistics of an internal variable of the decoder itself is proposed in this paper. Different from the previous reported results, it collects the statistics of the multiplicity of the candidate sequences, i.e., the number of the same candidate sequences with the same minimum squared Euclidean distance resulted from the decoding of test sequences. It is shown by Monte Carlo simulations that the proposed adaptive algorithm has only about 0.02dB coding loss but the average complexity of the proposed algorithm is about 42% less compared with Pyndiah＇s iterative decoding algorithm using the fixed LRBs parameter.展开更多
基金the National Natural Science Foundation of China under grants，NUAA research funding
文摘It is known that Block Turbo Codes （BTC） can be nearly optimally decoded by Chase-II algorithm, in which the Least Reliable Bits （LRBs） are chosen empirically to keep the size of the test patterns （sequences） relatively small and to reduce the decoding complexity. While there are also other adaptive techniques, where the decoder＇s LRBs adapt to the external parameter of the decoder like SNR （Signal Noise Ratio） level, a novel adaptive algorithm for BTC based on the statistics of an internal variable of the decoder itself is proposed in this paper. Different from the previous reported results, it collects the statistics of the multiplicity of the candidate sequences, i.e., the number of the same candidate sequences with the same minimum squared Euclidean distance resulted from the decoding of test sequences. It is shown by Monte Carlo simulations that the proposed adaptive algorithm has only about 0.02dB coding loss but the average complexity of the proposed algorithm is about 42% less compared with Pyndiah＇s iterative decoding algorithm using the fixed LRBs parameter.
