一种改进的稀疏多径信道均衡方法  

Improved sparse multipath channel equation method

作　　者：杨源[1] 李明阳[2] 王徐华[2] 

机构地区：[1]空军工程大学空管领航学院,陕西西安710051 [2]空军工程大学综合电子信息系统与电子对抗技术研究中心,陕西西安710051

出　　处：《西安电子科技大学学报》2014年第1期158-163,共6页Journal of Xidian University

基　　金：陕西省博士后科学基金资助项目(2012JQ8034)

摘　　要：针对传统的信道均衡算法在稀疏信道下效率低及实现复杂等问题,提出了一种改进的稀疏多径信道的均衡方法,利用少量训练序列进行信道估计作为先验知识,从而反演得到逆滤波均衡器.文中将求逆滤波过程建模为最优化问题,并提出一种获得近似最优解的贪婪算法.该算法相对于线性预测方法迭代次数极少,只要阶数足够高,就能获得几乎最佳的滤波器系数.设计了该算法模块化硬件结构,其复杂度低,易于工程实现.仿真结果表明,在稀疏信道下,文中的均衡方法相对于传统的最小均方线性预测方法,随着信噪比的增加,系统误码性能提升明显,在信噪比为15dB时,能够获得大约10dB的功率余度.Traditional channel equation methods are based on the multi-path richness hypothesis, which is complicated and inefficient in sparse channels. In this paper, a sparse multi-path channel equation method is proposed. The sparse channel estimation is carried out using a small number of pilot tones based on Compressed Sensing （CS）. The equation inverse filter is derived from the channel estimation. The procession of inverse filter solution is modeled as an optimization problem and a greedy algorithm is proposed which can bring about a near optimal solution. The new algorithm requires fewer iterations than linear prediction and gets almost optimal filter parameters when the rank is high enough. The modularized structure of the greedy algorithm is designed which is less complicated and can be easily realized. Simulation shows that the BER performance of the proposed equation method is improved significantly with the increase of SNRs. At 15 dB of the SNR it gains 10 dB in power efficiency relative to LMS.

关 键 词：稀疏信道 压缩感知 逆滤波 贪婪算法 

分 类 号：TN914.42[电子电信—通信与信息系统]