基于混沌序列构造测量矩阵优化算法

发布时间：2018-03-09 23:33

本文选题：压缩感知　切入点：混沌序列　出处：《四川大学学报(工程科学版)》2014年S2期 　论文类型：期刊论文

【摘要】：基于混沌序列具有良好伪随机性的特点来构造测量矩阵,提出了2种基于调整混沌测量矩阵列向量顺序的优化算法。第1种方法是分析信号稀疏向量中系数,对恢复信号贡献值最大的系数,选用混沌自适应测量矩阵最优的向量进行采样。第2种方法是对稀疏向量不均匀采样。通过理论分析和实验仿真结果证明了采用混沌构造测量矩阵的可行性。实验仿真结果还表明了混沌构造测量矩阵的恢复效果比高斯随机测量矩阵和伯努利测量矩阵的恢复效果更好,并且提出的2种优化算法在相同的压缩比下都能显著地提高信号恢复的效果。
[Abstract]:Based on the good pseudorandom of chaotic sequences, two optimization algorithms based on adjusting the sequence of column vectors of chaotic measurement matrices are proposed. The first method is to analyze the coefficients in the sparse vectors of signals. The coefficient with the greatest contribution to the recovery signal, The optimal vector of the chaotic adaptive measurement matrix is selected for sampling. The second method is to sample the sparse vector unevenly. The feasibility of constructing the measurement matrix with chaos is proved by theoretical analysis and experimental simulation. The simulation results also show that the restoration effect of chaotic construction measurement matrix is better than that of Gao Si random measurement matrix and Bernoulli measurement matrix. And the two optimization algorithms can significantly improve the effect of signal recovery under the same compression ratio.
【作者单位】：四川大学电子信息学院;
【分类号】：TN911.7

【相似文献】