Sensing or Transmission? Stochastic Scheduling of Energy-Harvesting Sensors Toward Zero-Carbon IoT

Abstract

Internet-of-Things (IoT) has been widely applied in various domains during the past decades. However, battery-powered IoT devices are notorious for the limited lifetime due to the limitation on the battery capability. Lots of methods have been proposed to promote the energy efficiency so as to prolong the network lifetime for sustainable, and even zero-carbon, IoT. Recent developments in hardware design have enabled IoT devices to support themselves by harvesting energy from the environment, forming energy harvesting IoT (EH-IoT). EH-IoT has been recognized as a promising form of sustainable IoT. During the operation of an EH-IoT, one inevitable dilemma on each IoT device is the operation decision on sensing and transmission in each timeslot. The problem is challenging due to the unpredictability of the amount of energy that can be harvested in each timeslot. In this paper, we are motivated to investigate how to dynamically allocate the limited harvested energy for sensing and transmission to maximize the overall network utility in EH-IoT. A perturbation-based Lyapunov technique is applied to tradeoff the optimality gap and system stabilization and an online distributed algorithm is proposed to pursue the close-to-optimal network utility in EH-IoT. Simulation results verify the efficiency of the proposed algorithm and the correctness of our analysis.