Project Title: A New Real-Time IoT Data Analytics Architecture
Team: Celal ÇEKEN,Ph.D Mohammed Al-Hubaishi,Ph.D. Candidate Nur Banu OĞUR, Ph.D. Candidate
Sponsors: Sakarya University
Start Date: 2018
In this project, a new real-time data analytics testbed is proposed for IoT-based systems. In the vertical domain, the proposed platform includes a Wireless Sensor Network (WSN) which was modeled and simulated using Riverbed Modeler. Additionally, the horizontal domain of the testbed has i) a Kafka messaging system that distributes incoming data to different points ii) a web application to visualize the temperature values, in real-time, iii) a Spark platform for real-time data analytics, and v) a NoSQL database that stores incoming data. The study reveals that the vertical domain of any IoT system can be modeled and simulated using Riverbed Modeler. In other words, the proposed architecture combines real data analytics sub-system with a simulation environment and, consequently, can be used as a time- saving experimental environment for IoT-based big data analytics.
In order to investigate the performance of the system developed, the diagnosis of Wolf Parkinson White (WPW) syndrome by logistic regression is discussed as a case study. The results show that the proposed IoT data analytics system can process health data in real-time, successfully, and is capable of handling large volumes of data easily owing to the scalable technologies deployed in it.