Data Collection in Radio Frequency (RF) Charging Internet of Things Networks

This paper considers minimizing the time required to collect $L$ bits from each Radio Frequency (RF) energy harvesting device serviced by a multi-antenna Hybrid Access Point (HAP). We outline a mixed integer non-linear program (MINLP) to determine the transmit power allocation of the HAP and devices over {\em multiple} time slots. We also outline a Receding Horizon Control (RHC) approach coupled with a Gaussian Mixture Model (GMM) to optimize the transmit powr allocation of the HAP. Our results show that the performance of our approach is within 5% of the optimal solution.