基于分布式融合的传感器网络拓扑配置

Topology configuration of sensor networks based on distributed fusion

采用2种最优分布式异步航迹融合方法研究异步采样传感器网络的动态拓扑配置策略设计.其主要核心思想是:每一个传感器跟踪节点执行局部Kalman滤波,再将滤波估计结果传输到融合中心;融合中心利用各自不同的融合方式执行预测估计校准和最优递推加权融合,同时利用当前所有传感器信息的全局递推融合估计与系统精度要求的阈值进行实时比较,以决定是否终止或继续进行融合;实现下一时刻网络拓扑的动态配置和网络节能.基于特定的计算准则分析网络能量消耗,并通过计算机仿真验证算法的有效性.结果显示:2种方法都能实现异步采样多传感器网络的动态拓扑配置和节能,且最优异步融合配置方法的效果优于次优异步航迹融合方法.

Dynamic network topology configuration（DNTC） was studied by adopting two optimal distributed asynchronous track fusion（ATF） methods for sensor networks with asynchronous sampling.Firstly,the local Kalman filtering is done in the network node and the estimates are transmitted to the processing center where different fusion operations are performed to realize predict calibration and recursive weighted fusion.Secondly,the real-time recursive fusion estimate is compared with the tracking precision and the decision to stop or continue is given.Finally,the DNTC and energy conservation can be realized.Moreover,the analysis of network energy conservation is also presented and the simulation examples are given to show the validity of the proposed methods.The results show that both the proposed methods can realize the DNTC function and energy conservation,and the method which adopts optimal fusion has better effect than the suboptimal one.