OPTIMUM SENSOR SCHEDULING FOR CHEMICAL DIFFUSIVE SOURCE LOCALISATION USING DISTRIBUTED ESTIMATION

Yong Zhang,Liyi Zhang,Li Wang,Jigong Li,Zhe Ban
2016-01-01
Abstract:Chemical diffusive source localisation using probability estimation method in sensor networks has an important significance in the environmental monitoring, security protection and pollution control. Sensor scheduling, which is also means sensor selection, is the most important problem in such sensor systems. In this paper, a distributed particle filter (DPF) estimation method with sensor scheduling optimisation to determine the position and emission rate of a diffusive source is studied. We gave a composite objective function which combines the information utility measure and the communication energy consumption between sensor nodes, and the sensor scheduling scheme is designed based on the optimisation of the objective function. At each time instant, each selected sensor implement the parameters estimation with a dispersion physical model and concentration signals measured by the selected sensor and its neighbours. In order to compare the performance of proposed DPF algorithm, the unscented Kalman filter (UKF) was also studied. Simulation results show that both the UKF and DPF could localise the diffusive source, but the DPF is computationally effective and it can get a significant reduction in the required number of processing nodes to achieve the desired performance with more robust to parameter perturbation.
What problem does this paper attempt to address?