基于频率位置多项式的稀疏混叠频谱快速恢复算法  

Frequency Locator Polynomial Based Fast Algorithm for Sparse Aliased Spectrum Recovery

作　　者：曹凯 陆佩忠[2,3] 邹艳[2] 凌琳 CAO Kai;LU Peizhong;ZOU Yan;LING Lin(Institute of Information System Engineering,Information Engineering University,Zhengzhou 450001,China;Southwest Institute of Electronics and Telecommunication Technology,Shanghai 200434,China;Department of Computer Science and Engineering,Fudan University,Shanghai 200433,China)

机构地区：[1]信息工程大学信息系统工程学院,郑州450001 [2]西南电子电信技术研究所上海分所,上海200434 [3]复旦大学计算机科学与工程系,上海200433

出　　处：《电子与信息学报》2018年第9期2105-2111,共7页Journal of Electronics & Information Technology

基　　金：国家自然科学基金(60673082)~~

摘　　要：论文提出一种基于频率位置多项式的稀疏混叠频谱快速恢复算法。该算法使用不同延时的多通道欠采样得到的信号混叠频谱,通过建立频率位置多项式,快速定位非零频点,并有效地将非线性的频谱恢复问题转换成一系列线性方程组的求解问题。该算法的计算速度相对国外同类算法(Big Band)有显著提高,并且实验结果表明该算法具有更低的频谱恢复错误率。A fast algorithm based on Frequency Locator Polynomial （FLP） for sparse spectrum recovery is proposed. Using the shifted subsampled signals, the FLPs are constructed, thus to locate rapidly the nonzero frequencies. In particular, the nonlinear problem of sparse spectrum recovery is converted into solving a series of linear equations. Experimental results show that the proposed algorithm exhibits higher processing speed and lower error spectrum reconstruction rate than its predecessor BigBand.

关 键 词：宽带频谱感知 频谱恢复 欠采样 稀疏傅里叶变换 位置多项式 

分 类 号：TN911.12[电子电信—通信与信息系统]