基于分簇特性的宽带信道估计算法  

Wideband channel estimation algorithm based on cluster characteristic

作　　者：王燕[1,2] 张立毅[3] 张锐[4] 

机构地区：[1]天津大学电子信息工程学院,天津300072 [2]天津城建大学计算机与信息工程学院,天津300384 [3]天津商业大学信息工程学院,天津300134 [4]天津科技大学电子信息与自动化学院,天津300222

出　　处：《计算机工程与设计》2015年第9期2329-2333,共5页Computer Engineering and Design

基　　金：国家自然科学基金项目(61071207);天津市应用基础研究计划基金项目(13JCYBJC15600);天津市高等学校科技发展基金计划基金项目(20110706);山西省自然科学基金项目(2011011015-3)

摘　　要：针对宽带稀疏信道具有分簇结构的特点,在原有基于压缩感知稀疏信道估计算法的基础上,加入稀疏信道分簇结构,提出一种基于簇稀疏特性的自适应正则匹配追踪压缩信道感知算法。在信道簇稀疏度未知的情况下,自适应调整候选簇的数目,应用正则化的思想对候选支撑簇进行二次筛选,达到信道的精确重构。仿真结果表明,该算法比传统的LS算法、BPDN算法、OMP算法、BOMP算法具有更低的误比特率(BER)、更小的均方误差(MSE)。According to cluster structure feature for wideband sparse channel, a cluster regularized adaptive matching pursuit （CRAMP） algorithm was presented based on the original compressed channel sensing estimation algorithm. Although the cluster sparsity of the channel was unknown, the channel was accurately reconstructed by the adaptive process in which the number of candidate clusters was chosen and the regularized process in which supports were provided for secondary screening of candidate clusters. Simulation results show that compared to the traditional LS algorithm, the original BP algorithm, OMP algorithm and BOMP algorithm, the proposed algorithm has lower bit error rate （BER） and smaller mean square error （MSE）.

关 键 词：簇稀疏性 压缩信道感知 S-V信道模型 信道估计 OFDM 

分 类 号：TN914.51[电子电信—通信与信息系统]