论文部分内容阅读
针对存在丢包的传感器网络中每个传感器节点对目标估计确信度不同的问题,提出一种基于自适应加权融合的分布式滤波算法.考虑节点在网络中的影响力及其节点属性,将节点重要度与传感器网络节点观测数据间的支持度线性加权,获得每个传感器节点对目标的估计确信度,并将该确信度构成的融合权值引入节点状态估计值的一致性协议中,更新传感器节点对目标的状态估计值,提高分布式滤波算法的估计精度和传感器节点估计值的一致性.仿真结果验证了所提出方法的有效性.
Aiming at the problem that each sensor node in the sensor network with packet loss has different confidence in the target estimation, a distributed filtering algorithm based on adaptive weighted fusion is proposed. Considering the node’s influence in the network and its node attributes, The degree of support and the degree of support of sensor nodes are linearly weighted to obtain the estimated confidence of each sensor node to the target and the fusion weight formed by this confidence is introduced into the consistency protocol of node state estimation to update the sensor The state estimation value of the node to the target increases the consistency between the estimation accuracy of the distributed filtering algorithm and the estimated value of the sensor node.The simulation results verify the effectiveness of the proposed method.