基于压缩感知的分布式协同估计算法

发布时间：2019-02-14 23:03

【摘要】：为了降低分布式协同估计算法的计算量并改善其收敛性能,提出了基于压缩感知(CS)和递归最小二乘(RLS)的分布式协同估计算法。该算法在传统RLS分布式协同估计算法的基础上引入压缩感知技术,首先在压缩域中进行递归最小二乘运算,然后利用压缩感知重构算法得到未知参数向量的估计值。提出的算法能够在增量式策略和两种模式的扩散式策略下实现对未知向量的有效估计。理论分析和仿真结果表明,该算法一方面降低了RLS分布式协同估计算法的计算量,另一方面保持较快的收敛速度与良好的均方误差性能。
[Abstract]:In order to reduce the computational complexity and improve the convergence performance of the distributed cooperative estimation algorithm, a distributed cooperative estimation algorithm based on compression-aware (CS) and recursive least square (RLS) is proposed. Based on the traditional RLS distributed cooperative estimation algorithm, the compressed sensing technique is introduced. Firstly, the recursive least square operation is performed in the compressed domain, and then the estimated value of unknown parameter vector is obtained by using the compressed perceptual reconstruction algorithm. The proposed algorithm can effectively estimate unknown vectors under incremental strategy and diffusion strategy of two modes. Theoretical analysis and simulation results show that the proposed algorithm can reduce the computational complexity of the RLS distributed cooperative estimation algorithm on the one hand and maintain a faster convergence rate and good mean square error performance on the other hand.
【作者单位】：北京信息科技大学信息与通信工程学院;
【基金】：国家自然科学基金资助项目(61302073) 北京市自然科学基金面上项目(4172021);北京市自然科学基金资助项目(Z160002) 北京市教委面上项目(KM201711232010)
【分类号】：TN911.7

【相似文献】