Energy-efficient sensor selection strategies based on sensing model clustering
Xiaole Wang, Tianshe Yang, Henghai Fan, Bo Qin, Jiansheng Zhu · International Journal of Sensor Networks · 2016
It can effectively save and balance energy consumption by sensor selection technology. This paper had studied the relationship between the similarity of sensing model and the system state evaluation, and proposed a sensor selection strategy based on sensing model clustering. This strategy has taken full account of the sensing model sensitivity and the residual energy of sensor, so it can quickly solve the energy-efficient sensor selection problem. The sensor selection objective function has been established by using the Fisher information matrix (FIM), and it is proved that the covariance of error would be smaller when the similarity between any two sensors' sensing model in the selection subset was larger. Finally, by taking the target tracking as the simulation scenarios, the experimental results show that the proposed strategy can maintain high efficiency under the premise of providing a more accurate system state estimation.