Communication Capacity and Reliability in Green Internet of Things for Critical Applications under Extreme Environments

Abstract

The Internet of Things (IoT) deployments in remote outdoor environments are subject to the unavailability of an adequate power source. Hence the IoT nodes must adapt to green energy generated from renewable resources like solar, wind, etc. newlineThe nodes also need energy harvesting mechanisms to operate when there is a shortage of external energy. One such application we see in this research work is the Green IoT (GIoT) aided landslide sensing system that uses ambient energy like solar and transmits the data in real-time. The GIoT environment has dynamic weather conditions that cause frequent network disconnections, erratic packet losses, changing energy harvesting rates, and varying energy consumption levels. The landslides are more susceptible during monsoons and rough weather conditions. During this time, the green energy generation decreases, and IoT network connectivity also becomes unstable. It results in difficulty in providing early warning for landslides. Hence there is a need for an reliable end-to-end real-time data transmission architecture from the field sensors to the Data Management Center (DMC). The proposed system is capable of real-time streaming of heterogeneous sensor readings through heterogeneous wireless networks at a rate up-to 1700/s within a latency of 10 s and less than a minute to re-establish the connection. We developed a state machine called A3 continuum for GIoT nodes that can handle Adaptive QoS, Adaptive Energy consumption, and Adaptive Transmission. A3 continuum can handle the dynamicity of packet loss, energy consumption, and transmission in each node. Evaluation results show A3 continuum increases the QoS by 51% and saves energy consumption by 50% in the nodes. We also designed a security mechanism called Voice Response Internet of THIngs (VRITHI) for GIoT nodes that reduces DoS attacks from 82-65% to 28%. Both A3 continuum and VRITHI require only a few kilobytes of memory in ultra-low power hardware of GIoT that are powered by ambient energy. newline newline