非理想信噪比估计对Turbo均衡性能的影响  

Impact of non-ideal SNR estimation on Turbo equalizer performance

作　　者：聂远飞[1] 葛建华[1] 王勇[1] 

机构地区：[1]西安电子科技大学综合业务网理论及关键技术国家重点实验室,西安710071

出　　处：《吉林大学学报（工学版）》2007年第4期875-880,共6页Journal of Jilin University:Engineering and Technology Edition

基　　金：国家自然科学基金重大项目(60496316;60332030);'863'国家高技术研究发展计划项目(2006AA01Z270)

摘　　要：在信噪比估计非理想条件下,给出了一种简单的先验信息模型,使用外信息转移图分析了信噪比估计对基于线性最小均方误差的Turbo均衡系统性能的影响.所给示例表明：在-3～6 dB条件下,长帧的渐进性能对估计偏差并不敏感,而中短帧的不敏感区间则会大大缩小,系统仿真验证了分析的准确性,最后提出了一种新的结合先验信息的噪声方差估计算法,同时应用到Turbo均衡中,并通过仿真验证了算法的有效性.This paper proposes a simple model of a prior information with non-ideal signal-to-noise ratio （SNR） estimation. Extrinsic information transfer （EXIT） chart is applied to analyze the impact of SNR mismatch on Turbo equalizer based on linear minimum mean square error （LMMSE）. The example shows that asymptotic performance is indifferent when the estimated SNR is between -3 dB and 6 dB for long frame length, while the insensitive region becomes decreased for moderate and short length. The accuracy of our EXIT chart prediction is verified by simulation. Finally, a novel noise variance estimator is proposed with a prior information aided. It is employed to Turbo equalization system and simulation results validate its effect.

关 键 词：通信技术 信噪比 外信息转移 TURBO均衡 

分 类 号：TN913[电子电信—通信与信息系统]