分布式传感网络的自适应一致性算法

Adaptive consensus algorithm for distributed sensor network

为了提高分布式传感网络的估计精度,提出了一种新的自适应一致性算法。该算法在每次迭代时只需部分节点工作,即进行目标状态的监测。通过节点之间二进制信息的交换来调整每次迭代时的一致性权值,使得每次迭代时工作节点所占的权值更大,进而将该一致性算法与卡尔曼滤波相结合对目标状态进行估计。对该算法进行数值仿真,并与其他一致性加权算法进行比较,验证了该算法的有效性。

To improve the estimation accuracy of distributed sensor networks, this paper proposes a new adaptive consensus algorithm. In each iteration of the algorithm, it requires only some nodes to work, that is monitoring target state. The consensus weight between the nodes is adjusted by exchange of binary information to make the weights connected to the nodes that observe the state greater. It combines the consensus algorithm with Kalman filter to estimate target state. A numerical example is given to illustrate the proposed algorithm and compared with other consistency weighted algorithm, it shows the effectiveness of this approach.