Department of Automation, University of Science and Technology of China, Auhui 230027, Hefei, China.
School of Energy and Power, Jiangsu University of Science and Technology, Zhenjiang 212003, Jiangsu, China.
Sensors (Basel). 2020 Feb 16;20(4):1073. doi: 10.3390/s20041073.
We study the sensor and relay nodes' power scheduling problem for the remote state estimation in a Wireless Sensor Network (WSN) with relay nodes over a finite period of time given limited communication energy. We also explain why the optimal infinite time and energy case does not exist. Previous work applied a predefined threshold for the error covariance gap of two contiguous nodes in the WSN to adjust the trade-off between energy consumption and estimation accuracy. However, instead of adjusting the trade-off, we employ an algorithm to find the optimal sensor and relay nodes' scheduling strategy that achieves the smallest estimation error within the given energy limit under our model assumptions. Our core idea is to unify the sensor-to-relay-node way of error covariance update with the relay-node-to-relay-node way by converting the former way of the update into the latter, which enables us to compare the average error covariances of different scheduling sequences with analytical methods and thus finding the strategy with the minimal estimation error. Examples are utilized to demonstrate the feasibility of converting. Meanwhile, we prove the optimality of our scheduling algorithm. Finally, we use MATLAB to run our algorithm and compute the average estimation error covariance of the optimal strategy. By comparing the average error covariance of our strategy with other strategies, we find that the performance of our strategy is better than the others in the simulation.
我们研究了带有中继节点的无线传感器网络（WSN）中远程状态估计的传感器和中继节点的功率调度问题，该问题是在给定有限的通信能量的情况下，在有限的时间段内完成的。我们还解释了为什么最优的无限时间和能量情况不存在。以前的工作应用了一个预定义的阈值，用于 WSN 中两个连续节点的误差协方差间隙，以调整能量消耗和估计精度之间的权衡。然而，我们没有调整权衡，而是采用一种算法来找到最佳的传感器和中继节点调度策略，该策略在我们的模型假设下，在给定的能量限制内，以最小的估计误差实现。我们的核心思想是通过将前一种更新方式转换为后一种方式，将传感器到中继节点的误差协方差更新方式与中继节点到中继节点的方式统一起来，从而可以使用分析方法比较不同调度序列的平均误差协方差，从而找到具有最小估计误差的策略。示例用于演示转换的可行性。同时，我们证明了我们调度算法的最优性。最后，我们使用 MATLAB 运行我们的算法，并计算最优策略的平均估计误差协方差。通过将我们的策略的平均误差协方差与其他策略进行比较，我们发现我们的策略在模拟中的性能优于其他策略。
