Leveraging Light Intensity Structural Breaks for IoT Context-Awareness

Sensor data serves as a valuable information source that facilitates context-awareness within an Internet of Things (IoT) environment, thereby augmenting its capacity to adapt intelligently to evolving conditions and user requirements. This paper explores the utilization of structural breaks within time series to enhance context-awareness in the realm of a smart IoT environment, particularly within smart homes. By identifying shifts or disruptions in the underlying sequential measurements of light intensity, this research aims to develop a methodology capable of understanding the action of an entity that interacts with a smart environment. Specifically, by leveraging machine learning and statistical techniques for sensor time series, we delve into the detection of structural breaks, elucidating their significance in capturing relevant context information crucial for providing situational awareness. Through empirical validation and a case study with an IoT device, the paper demonstrates the efficacy of incorporating structural break analysis as a pivotal component in enhancing the sensing and awareness of ambient systems deployed in residential settings.