基于斜投影算子的压缩采样匹配追踪算法

CSMP Algorithm of Compressed Sensing Based on Oblique Projection Operator

斜投影匹配追踪算法较正交匹配追踪算法针对非正交字典的重构能力有了大幅提高,但受噪声影响大。该文通过投影矩阵迭代公式导出斜投影匹配追踪算法,并引入回溯的思想,提出基于斜投影算子的压缩采样匹配追踪算法,有效降低了噪声的影响和非正交投影产生的误差,提高了重构精度。仿真结果表明,算法的重构效果优于其它同类算法。

Oblique projection matching pursuit algorithm （ OPMP） voids the mismatch phenomenon caused by the non-orthogonality of over-complete dictionary in the orthogonal matching pursuit algorithm （ OMP） , but it is affected by noise .Using the iteration formula of projection matrix , the OPMP is derived in this paper .By introducing the idea of backtracking , a compressive sampling matching pursuit algorithm based on oblique projection operator（OPCSMP） is proposed.OPCSMP effectively reduces the effect of noise and errors which generated by non-orthogonal projection , and improves the accuracy of the reconstruction .Experimental results show that the effect of the proposed algorithm is far superior to that of the OMP .