基于卡尔曼滤波的稀疏流信号动态压缩感知  

Dynamic compressed sensing of sparsing streaming signal based on Kalman filter

作　　者：田金鹏[1] 薛莹[1] 闵天 Tian Jinpeng;Xue Ying;Min Tian(School of Communication and Information Engineering,Shanghai University,Shanghai 200444,China)

机构地区：[1]上海大学通信与信息工程学院

出　　处：《电子测量技术》2018年第19期73-78,共6页Electronic Measurement Technology

摘　　要：提出一种针对稀疏流信号的卡尔曼滤波压缩感知恢复方法。该算法采用交叉随机模型对信号进行压缩采样,基于前后窗口内信号之间的相关性,建立信号的状态转移方程,利用卡尔曼滤波算法与贪婪算法相结合得到信号最小均方误差估计。信号重构阶段,基于压缩感知贪婪算法进行支撑集更新,采用自适应误差阈值判别方法,不断迭代减小残差以找到精确程度较高的支撑集,将其用于卡尔曼滤波得到信号最优估计。仿真结果表明,与同类算法相比,本算法提高了重构精度和抗干扰能力,并且算法的复杂度较低。Kalman filter compression perceptual recovery method for sparse flow signals is proposed. The algorithm uses the cross-random model to compress the signal. Based on the correlation between the signals in the adjacent windows, the state transition equation of the signal is established. The least mean square error estimation of the signal is obtained by using the Kalman filter algorithm and the greedy algorithm. In the signal reconstruction stage, the support set is updated based on the compression-aware greedy algorithm, and the adaptive error threshold method is used to iterate to reduce the residual set update. The adaptive error threshold method is used to iterate the residuals to find the accuracy. The higher support set is used for Kalman filtering to do signal estimation. The simulation results show that compared with the same kind of algorithms, this algorithm improves the reconstruction precision and anti-jamming ability, and the complexity of the algorithm is low.

关 键 词：动态压缩感知 卡尔曼滤波 流信号重构 自适应判别 

分 类 号：TN911.7[电子电信—通信与信息系统]