Journal of Systems Engineering and Electronics ›› 2011, Vol. 33 ›› Issue (9): 1948-1953.doi: 10.3969/j.issn.1001-506X.2011.09.07

• 电子技术 • 上一篇    下一篇


甘伟, 许录平, 罗楠, 谢强   

  1. 西安电子科技大学电子工程学院, 陕西 西安 710071
  • 出版日期:2011-09-17 发布日期:2010-01-03

Adaptive recovery algorithm for compressive sensing

GAN Wei, XU Lu-ping, LUO Nan, XIE Qiang   

  1. School of Electronic Engineering, Xidian University, Xi’an 710071, China
  • Online:2011-09-17 Published:2010-01-03


为优化稀疏自适应匹配追踪(sparsity adaptive matching pursuit, SAMP)算法的性能,给出了一种修正自适应匹配追踪(modified adaptive matching pursuit, MAMP)算法。该算法采用模糊阈值预选方案,改进了步长选择方法,设置了初次裁剪门限。仿真结果表明,在同等稀疏的条件下实现精确重构,该算法的运算速度较原算法提高了1倍,所需的观测值个数减少了1%,并提高了重构精度。


In order to optimize the performance of sparsity adaptive matching pursuit (SAMP) algorithm, a modified adaptive matching pursuit (MAMP) algorithm is proposed. The proposed algorithm adoptes a fuzzy threshold preliminary rule, improves the method of choosing step size and sets the initial threshold to cut out the candidate set. Simulation results show that, compared with SAMP, the operation speed of MAMP is increased by a factor of 2 for the same sparsity level, and the required measurement number decreases about 1%. In addition, the recovery accuracy is also improved.