基于压缩感知的分布式协同估计算法

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。

A distributed collaborative estimation algorithm based on compressed sensing

In order to reduce the amount of calculation and improve the convergence performance,a distributed collaborative estimation algorithm based on compressed sensing and recursive least square(RLS) is proposed. Based on the traditional RLS distributed collaborative estimation algorithm,this algorithm introduces compressed sensing technology. Firstly,the recursive least square method is applied in the compressed domain. And then the estimation of the unknown parameter vector is achieved by compressed sensing reconstruction algorithm. The proposed algorithm can effectively estimate the unknown vector under the incremental strategy and two modes of diffusion strategy. The result of theoretical analysis and simulation shows that the proposed algorithm reduces the amount of calculation of RLS distributed collaborative estimation algorithm and also keeps fast convergence speed and good mean square error performance.