通信受限下网络化系统的最优估计

OPTIMAL ESTIMATION FOR NETWORKED SYSTEMS WITH COMMUNICATION CONSTRAINTS

研究了通信受限下网络化系统最优估计问题.由于通信受限,传感器节点无法同时将采样信息传输到远程滤波器.为此,文章提出了集中式最优估计算法和序贯式最优估计算法.前者将观测做扩维处理且具有较好的估计性能,但需要计算高维矩阵的逆,计算负担较大.后者无需计算高维矩阵的逆,具有实时性和灵活性,但损失了估计性能.这两种算法均可推广到测量丢失的情形.最后,通过一个目标跟踪的例子验证了所设计算法的有效性.

The optimal estimation problem is investigated in this paper for networked systems with communication constraints.Due to the time-division multiplexing principle,various sensor nodes cannot transmit sampling data to a remote estimator synchronously.Thus,the traditional kalman filter cannot be applied directly to the networked estimation with the communication constraint.Both the centralized estimation method and the sequential estimation method are proposed to solve the networked estimation problem with communication constraint.The former uses the augmentation technique and the performance is optimal,but it needs to calculate the inverse of high dimension matrix,which involves large amount of computations.The latter can avoid the burden,and has the merits for real-time appHcations,but its performance is inferior to the former.Both of them can be applied to the case with packet losses.Finally,a target tracking example is utilized to show the effectiveness of the proposed methods.