论文部分内容阅读
以应用于隧道结构健康监测的无线传感器网络为基础,针对长线形的隧道结构和分布式的节点布置,提出了超长线状多跳非均匀分簇结构.通过考虑节点剩余能量和优化簇头分簇半径,降低并平衡节点能耗.针对传感器数据冗余量大的问题,提出了基于超长线状分簇结构的分布式卡尔曼滤波融合算法.利用单节点不同时刻的数据,通过卡尔曼滤波器得到局部估计值,降低数据时间冗余度.在簇头节点端和汇聚节点端分别实现分布式卡尔曼滤波融合算法,降低数据空间冗余度,达到具有一致性的网络数据估计值.实验结果表明:该方法能有效实现超长线状分簇结构下的分布式数据融合,具有高可靠性和准确性.
Based on WSN applied to tunnel structure health monitoring, an ultra-long linear multi-hop non-uniform clustering architecture is proposed for long-line tunnel structure and distributed node layout. By considering the residual energy of nodes and optimizing the cluster head Cluster radius to reduce and balance node energy consumption.Aiming at the problem of large sensor data redundancy, a distributed Kalman filter fusion algorithm based on ultra-long linear clustering structure is proposed.Using the data of single node at different time and using Kalman filter Get local estimates and reduce the time redundancy of data.Furthermore, a distributed Kalman filter fusion algorithm is implemented at the cluster head node and the sink node to reduce the data space redundancy and achieve a consistent network data estimation. The results show that this method can effectively realize the distributed data fusion under the ultra-long linear clustering structure with high reliability and accuracy.