A resilient filtering approach to multi-sensor systems with energy harvesters

The focus of this paper is to investigate the resilient filtering problem for multi-sensor systems subject to energy harvesting constraints. A practical scenario is taken into consideration where multiple sensors, after obtaining energy from a battery with an energy harvester, will consume a certain amount of energy to broadcast the measurements to the remote filter. In order to properly schedule the one-to-multiple energy allocation relationship between sensors and a battery, a deterministic energy allocation strategy is employed in this paper. Specifically, the battery supplies energy to the multiple sensors according to their priority order of transmitting measurements from high to low, where the priority order of all sensors is known to be determined. Additionally, the random filter gain variation is also considered for the sake of reflecting the phenomenon of inaccurate filter implementation. In a word, we are committed to develop a resilient filter such that an upper bound on the filtering error covariance is firstly ensured, and then the filter gain matrix is obtained to minimize the upper bound. Finally, we provide a simulation to demonstrate that the proposed resilient filtering strategy is effective.