Adaptive Sampling for Saving Energy: A Case Study on The Libelium-based Environment Monitoring Systems

Abstract

In the plethora of energy saving techniques developed in Internet of Things, adaptive sampling is one of the common methods to reduce the energy consumption of IoT nodes, at the cost of reducing the data accuracy. Additionally, the user cannot define the amount of energy to be saved when performing the adaptive sampling technique. This paper shows a case study applied our developed UDASA – The User-Driven Adaptive Sampling Algorithm for Massive Internet of Things on the Libelium-based environment monitoring systems. The aim of this work is to support users to trade-off between energy consumption on IoT devices versus the data precision. The results show that once applied UDASA in 4 days, the collected data only takes about 10% compared to that of without UDASA, while the system saves 9% of energy, and the data accuracy is about 84% after interpolation.