摘要
针对实际中未知稀疏度信号的重建问题,提出了一种自适应的弱选择压缩采样匹配追踪算法。该算法将自适应思想、弱选择思想与Co Sa MP算法相结合,在预选阶段后利用限制性弱选择策略对候选集进行二次筛选,通过双迭代阈值自适应地调整最终支撑集的原子数,并结合若干可靠性验证条件,保证算法的正确有效进行。MATLAB仿真结果表明,在相同的实验条件下,本算法可以有效地重建稀疏信号,同时具有较低的运算量,整体性能较优。
This paper proposed an adaptive weak-selection compressive sampling pursuit algorithm to reconstruct signals with unknown sparsity in practice. The algorithm combines adaptive idea and weak-selection idea with the Co Sa MP algorithm. It adopts limited weak-selection strategy to realize the second selecting of the atoms in the candidate set after the pre-selection stage, and then adaptively adjust the number of atoms in the final support set through double-threshold. We also incorporate some reliability demonstration conditions to the algorithm to ensure the correctness and effectiveness. The simulation results on MATLAB show that our algorithm can get better reconstruction performances and can run fast under the same conditions,which has a better overall performance.
出处
《电子设计工程》
2016年第11期150-153,共4页
Electronic Design Engineering
基金
微系统技术国防科技重点实验室基金项目(9140C18010214XXXX)
关键词
压缩感知
重建算法
自适应
弱选择
压缩采样
compressed sensing
reconstruction algorithms
adaptive
weak selection
compressive sampling
