摘要
针对未知稀疏度信号的重建,提出了一种改进的稀疏度自适应压缩采样匹配追踪(MACSMP)算法。该算法以压缩采样匹配追踪(CoSaMP)算法为基础,结合变步长自适应的思想,摆脱了对于信号稀疏度的依赖,并在迭代过程中引入正则化思想,从而提升了算法的重构精度。仿真结果表明,文中提出的MACSMP算法在重构性能与运行效率两方面都要优于SAMP、OMP、CoSaMP这几种算法,且其计算量较低,运行时间较短。
For the reconstruction of signals with unknown sparsity, a modified sparsity adaptive compressive sampling matching pursuit (MACSMP) algorithm is proposed. Based on the compressive sampling matching pursuit (CoSaMP) algorithm, the proposed algorithm adds the thought of regularizaton to iterative process,thus to improve the accuracy of the algorithm, and in combination with variable step adaptive idea, to solve the dependence on signal sparsity. Simulation results indicate that the proposed MACSMP algorithm is better than SAMP algorithm, OMP algorithm, CoSaMP algorithm in the reconstruction performance and operation efficiency, and in addition the calculation is lower and the running time shorter.
出处
《通信技术》
2016年第8期992-996,共5页
Communications Technology
关键词
压缩感知
重构算法
稀疏度自适应
正则化
compressed sensing
reconstruction algorithm
sparsity adaptation
regularization