An Optimal Management Modelling of Energy Harvesting and Transfer for IoT-based RF-enabled Sensor Networks

Abstract

[EN] A crucial conduct norm for a sensor network is to avoid network failures and packet drop. One of the other essential requirements is to effectively manage the energy levels of the nodes according to the states of the operation required for an application. This paper focuses to propose an energy management model with the aim of allowing energy optimization of Radio Frequency(RF)-enabled Sensor Networks (RSN) during the process of Energy Harvesting (EH) and Energy Transfer (ET) through controlled optimization. Primarily, energy harvesting of sensor networks through RF signals is focussed in this research to address the drawback of frequent replacement of batteries, persistent recharge request, dead state of nodes and periodical eradication of batteries. Secondly, this paper focuses on mathematical modelling of the RF sensor nodes within the proposed Energy Harvesting RSN (EHRSN) and Energy Transfer RSN (ETRSN) framework of Energy Management RSN model (EMRSN) where the nodes are characterized as Semi Markov Decision Process (SMDP) and optimal policies are computed for numerically evaluating and analysing the issue of higher energy consumption. The most optimal state transitions are computed and mathematically formulated based upon stochastic dynamic programming to carry out the numerical analysis. It has been found that through controlled optimization, the sensor networks when energized through RF energy for EH process, the probability of 0.8 or more works best at the lower power level. On the other hand, for ET, the sensors tend to work more when the probability is either 0.8 or more at higher power levels. The results obtained are further employed to program the sensors accordingly in the Internet of Things (IoT) contexts during EH and ET processes to achieve maximum throughput, network lifetime and energy efficiency