ADABCAST: Adaptive broadcast approach for solar Energy Harvesting Wireless Sensor Networks

The problem of message broadcasting from the base station (BS) to sensor nodes (SNs) in solar energy harvesting wireless sensor networks (EHWSN) is considered in this paper. The aim is to ensure fast and reliable broadcasting without interfering with upstream communications (from SNs to BS), whilst taking into account energy harvesting constraints. An adaptive approach is proposed where the BS first selects the broadcast time slots, given a wake-up schedule for the SNs (the time slots where the SN are active and in receiving mode). Hence, the SNs adapt their schedules. This is then iterated seeking optimal selection of the broadcast time slots, so as to minimize broadcast overhead (transmitted messages) and latency. Our approach enables fast broadcast and eliminates the need for adding protocol overhead (redundancy), compared to the existing solutions. Hidden Markov Model (HMM) and Baum-Welch learning algorithm are used for this purpose. Numerical results confirm that our scheme performs the broadcast operation in less time, and by reducing the broadcast overhead, as compared to state-of-the-art approaches.