A ROOF Computing Architecture based Indoor Positioning System for IoT Applications

In this paper, the design and implementation of an indoor localization and positioning system using four low-cost microphones is proposed. The proposed design leverages architectural advantages of Real-time Onsite Operations Facilitation (ROOF) computing standards. The data collected from deployed sound sensors is used for training the system. Three machine learning algorithms are compared with respect to computational time required for training and testing. Further, an IoT application is demonstrated using the proposed framework which could be integrated with a smart home device control system for providing location based context for device control.