TURBO编码系统中信噪比盲估计算法  

Improving Algorithm for Blind Estimation of SNR in Turbo Coded System

作　　者：姚如贵[1] 王永生[1] 苗睿锋[1] 徐娟[1] 

机构地区：[1]西北工业大学电子信息学院,陕西西安710072

出　　处：《西北工业大学学报》2006年第1期54-58,共5页Journal of Northwestern Polytechnical University

摘　　要：针对盲估计算法复杂度高、运算量大,研究了TURBO译码器中两个成员译码器输出的外部信息与信道信噪比的关系,提出一种基于译码器外部信息量的信噪比盲估计算法。仿真结果说明,采用此信噪比盲估计算法,在较小的系统性能恶化的前提下,大大简化了信道估计器的复杂度。分析了信噪比盲估计算法的误差性能,理论分析表明,发生误判的几率很小,均在1-0 3量级以下,可以认为信噪比估计值十分可靠。Existing algorithms for blind estimation of SNR （signal-to-noise ratio） in Turbo coded system are too complex and, in our opinion, can be improved. We propose a better algorithm that utilizes the extrinsic information of the two decoders within the Turbo code system. In the full paper, we describe our better algorithm in much detail and we give numerical results in Fig. 4 and Table 2; here we just give a briefing. The proposed blind estimation algorithm can greatly decrease the complexity of the channel estimator but at the cost of some sacrifice in performance. However, as we are mostly interested in the 7- 10 dB range of SNR, the sacrifice in performance is quite small. As to the error performance of our better algorithm, the estimated SNR is quite reliable as its fault probability is below 10^-3.

关 键 词：TURBO码 信噪比 盲估计 外部信息 

分 类 号：TN919.7[电子电信—通信与信息系统]