Towards Internet of Things for event-driven low-power gas sensing using carbon nanotubes

Abstract

One of most important applications of sensing devices under the Internet of Things paradigm is air quality monitoring, which is particularly useful in urban and industrial environments where air pollution is an increasing public health problem. As these sensing systems are usually battery-powered and gas sensors are power-hungry, energy-efficient design and power management are required to extend the device's lifetime. In this paper, we present a two-stage concept where a novel low-power carbon nanotube is used as a gas detector for an energy-consuming metal-oxide (MOX) semiconductor gas sensor. We propose a design of a heterogeneous sensor node where we exploit the low-power nanotube gas sensor and the more accurate MOX sensor. This work performs energy consumption simulations for three event-driven scenarios to evaluate the power consumption reduction, as well as the limitations of carbon nanotubes. Our results show the benefits of the proposed approach over the scenarios with adaptive duty-cycling with only MOX gas sensors, proved with 20%-35% node lifetime prolongation. The delay introduced due to the nanotube recovery time can be overcome by radio duty-cycled activity for detecting alarm messages from the neighbour nodes.