Abstract

With the capability of generating sustainable energy by exploiting the ambient environment (e.g., light, heat, vibrations, etc.), the energy harvesting (EH) technology is increasingly used on low-power smart objects, forming self-powered green Internet of Things (IoTs). Despite targeting different application domains, many of these green IoT systems adopt the distributed network paradigm by forming energy harvesting multi-hop wireless networks (EH-MWNs). While distributed networks have good scalability during the deployment phase, they are often complex, inelastic to change, and costly to manage. To this end, LoRa, a promising long-range wireless technology, has been suggested for centralized single-hop network controls in EH-MWNs. However, the signal propagation of LoRa is severely affected by the surrounding environment (e.g., obstructions, terrains, etc.), which not only creates a significant difference in packet delivery rate (PDR) between nodes but also increases the bandwidth cost for network controls. To address this issue, we propose a Cross-interface Data Transfer (CDT) scheme, which leverages the ZigBee interfaces coexisting in EH-MWNs to transfer LoRa data traffic from the nodes with a lower PDR to the nodes with a higher PDR for bandwidth-efficient transmissions. By jointly determining the pairs of nodes performing data transfer and the data transfer rates between them, CDT can reduce the overall LoRa bandwidth consumption while ensuring continuous operation of the EH-MWN. A prototype system is implemented by integrating commercial off-the-shelf LoRa and ZigBee interfaces into an IoT platform. Extensive real-world experiments show that under a moderate data traffic input, the bandwidth consumption of CDT is 25.8% and 48.8% lower than the standard LoRaWAN and a state-of-the-art data transfer scheme, respectively. The average energy cost per node is kept as low as 2.29 mW at the same time. Moreover, CDT also shows its performance advantages in terms of PDR and sustainability.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call