Energy-Efficient Scheduling for Distributed Estimation in Wireless Sensor and Actuator Networks with Kriging

摘要

Earlier works have investigated the problem of distributed estimation in wireless sensor and actuator networks (WSANs) under packet loss. However, observation correlation was not utilized which can mitigate performance degradation due to packet loss. In this paper, the problem of distributed estimation in WSANs under packet loss and correlated observations is studied. In particular, an energy-efficient scheduling scheme for sensor-actuator and actuator-actuator coordinations is proposed based on the ordinary Kriging method. This method utilizes observation correlation by semivariogram modeling. Considering the tradeoff between estimation performance and energy efficiency of the scheduling scheme, an optimization problem is formulated. As it is difficult to derive the analytical expression of the solution to the optimization problem, a global search algorithm which integrates a genetic algorithm with a simulated annealing algorithm is designed to approximate the solution. The results show that the proposed scheduling scheme outperforms the inverse distance weighting method in terms of estimation performance. Moreover, the tradeoff between estimation performance and energy consumption exists.