摘要
介绍了压缩感知理论的基础知识,并分析了压缩感知的重建算法。正则化正交匹配追踪算法引入了正则化思想进行原子筛选,使迭代次数减少,但前提是要知道信号的稀疏度。稀疏度自适应匹配追踪算法可以通过设置终止条件来使稀疏度自适应,但迭代次数较多,时间成本较大。在两种方法的基础上提出了一种改进的稀疏度自适应变步长正则化匹配追踪算法,该算法克服了上述两种算法的缺点。仿真结果表明,文中提出的算法较准确地重构出原始信号,且运算时间较低。
This paper introduces the fundamental knowledge of compressed sensing theory, and analyzes the reconstruction algorithm of compressed sensing. Regularized orthogonai matching pursuit algorithm is introduced into the selection of regularization thought of atoms, reduce the number of iterations, but only if you know the signal sparse degree. Sparsity adaptive matching pursuit algorithm can be terrainated by setting the conditions to make adaptive sparse degree, but the number of iterations is more, and time cost is Iarger. This paper puts forward a modified sparsity adaptive variable step regularized matching pursuit algorithm. The algorithm overcomes the drawback of the above two algorithms. The simulation results show that the proposed algorithm can accurately reconstruct the original signal, and less operation time.
出处
《计算机安全》
2014年第1期32-35,共4页
Network & Computer Security
关键词
压缩采样
正则化
自适应
匹配追踪
compression sampling
regularization
adaptive
matching pursuit