Achieving Immortality in Wireless Rechargeable Sensor Networks Using Local Learning

Wireless rechargeable sensor networks have been studied in depth after the breakthroughs in wireless power transfer. In this paper, we propose an efficient decentralized recharging mechanism based on the use of a mobile energy charger (MC), restoring the energy supply to nodes in need.Our strategy is composed of an initialization phase, during which the MC moves in the network to learn the sensors’ location and to make an estimate of their consumption rate, followed by charging rounds where the charging schedule is dynamically computed based on the most up-to-date information collected.Our strategy is fully decentralized, uses only local communication between the MC and neighbouring nodes, operates under dynamic network conditions without any a-priori knowledge of the network, is based on simple computations by the MC with little memory requirements, and is highly effective in keeping the network perpetually operating and achieving immortality even in large networks.