基于压缩感知的测量矩阵研究  

Research on measurement matrix based on compressed sensing

作　　者：马庆涛[1] 唐加山[2] 

机构地区：[1]南京邮电大学通信与信息工程学院,江苏南京210003 [2]南京邮电大学数理学院,江苏南京210003

出　　处：《微型机与应用》2013年第8期64-67,共4页Microcomputer & Its Applications

摘　　要：压缩感知打破了传统采样定理的限制,提供了一种从少量的非自适应线性测量值中就能恢复原始信号的方法。测量矩阵正是获取这些测量值的关键所在,寻求结构简单、性能稳定的测量矩阵一直是研究人员的目标。在介绍压缩感知测量矩阵的基础上,提出了广义轮换矩阵的改进方法,结合正交基线性表示的思想,利用广义轮换构造的正交矩阵来生成新的测量矩阵。通过仿真实验,证明了新的测量矩阵具有较好的性能。Compressed sensing broke the limits of traditional sampling theorem, showed a new method to recover a signal from a small set of nonadaptive linear measurements. The measurement matrix is the key part of the measurements acquisition, seeking the measurement matrix witn simple structure and stable performance is the goal of researchers. This article introduces the measure- ment matrices for compressed sensing and states a new improved method for generalized rotation matrix. And introduces a new measurement matrix which is generated by the orthogonal basises from the improved generalized rotation matrix. Finally, the simula- tion experiment proves that the new measurement matrix has better perfonnance.

关 键 词：压缩感知 稀疏信号 测量矩阵 广义轮换矩阵 正交基 

分 类 号：TN911.7[电子电信—通信与信息系统]