对数域内T-TCM译码算法的分析与简化  

Analysis and simplification of T-TCM decoding algorithm in log domain

作　　者：闫涛[1] 杜兴民[1] 

机构地区：[1]空军工程大学工程学院,西安710038

出　　处：《计算机工程与应用》2008年第8期145-147,186,共4页Computer Engineering and Applications

基　　金：国家部委重点实验室基金

摘　　要：基于AWGN信道研究了对数域内T-TCM译码算法的表述与简化,重点分析了外信息在迭代译码过程中的产生与使用,结合雅可比对数式的特点详细给出分支度量计算的简化策略。仿真结果表明,忽略雅可比对数式的修正项并不会造成T-TCM系统性能的下降,通过理论分析给出了解释,并进一步提出无需SNR信息的T-TCM译码方案,在系统性能无损的情况下,降低了T-TCM在工程应用中的要求与实现复杂度。In this paper,expression and simplification of T-TCM decoding algorithm in Log domain is investigated for AWGN.The generation and utilization of extrinsic information is deeply analyzed during the iterative decoding process,and some simplification for branch metric calculation is presented in detail with the trait of Jacobian logarithm formula.According to the simulation results,discarding the correction function of Jacobian will not degrade the T-TCM performance,and a theoretical explanation is given in this paper,further a new T-TCM decoding scheme with no SNR is proposed,which can lower requirements and simplify implementation in the application with no performance degradation.

关 键 词：T-TCM 雅可比对数式 修正项 S-Max-Log-MAP算法 

分 类 号：TN919[电子电信—通信与信息系统]