Charging State Aware Optimal Auction Design For Sensor Selection In Crowdsourcing Based Sensor Networks

In this paper, crowdsourcing based wireless sensor networks (WSNs) with rechargeable sensors are used for target localization. For rechargeable sensors, the state of charge (SOC) is one of the key factors that decides the sensors' energy cost for the localization task. To conserve limited resources, the fusion center (FC) employs an optimal sensor selection scheme obtained through an auction design approach. The sensors compete to participate in the target localization task by sending bids based on their energy efficiency (analog data) and SOC (quantized data) to the FC. Aiming at maximizing the expected utility, the FC designs an optimal auction mechanism incorporating both analog information about sensors' energy efficiency and quantized information of the SOC and decides on the winning sensor(s) as well as the payment to the winner( s). Simulation experiments show the effective performance of our framework by investigating the effect of the SOC of the sensors on the utility of the FC and the target localization mean square error (MSE).