一种自适应观测矩阵下的信号重构算法*

为提高压缩感知的性能，设计了一种自适应的稀疏观测矩阵，该观测矩阵由0和1组成。信号重构时，利用观测值的位置信息，避免了求解不定方程组，提高了重构速度。采用具有频域稀疏特性的深海隔水管受力参数作为仿真信号，仿真结果表明，观测值数目相同时，自适应观测矩阵下重构算法的平均误差比随机观测矩阵下基追踪算法的平均误差小。

Signal reconstruction algorithm under adaptive measurement matrix

To improve the performance of compressed sensing, designed an adaptive sparse measurement matrix including 0 and 1. In the signal reconstruction, used the location information of observations to avoid solving diophantine equation, which improved the reconstruction speed. The deep-sea riser force parameter which was sparse in frequency domain was adopted to simulate. The results of computer simulation show that the average error of reconstruction algorithm under the adaptive measurement matrix is smaller than the basis pursuit algorithm under the random measurement matrix when the number of observations is same.