A Lightweight IDS Based on J48 Algorithm for Detecting DoS Attacks on IoT Middleware
Fariz Andri Bakhtiar, Eko Sakti Pramukantoro, Hilman Nihri · 2019 IEEE 1st Global Conference on Life Sciences and Technologies (LifeTech) · 2019
Internet of Things (IoT) is starting to be broadly used in healthcare. The IoT in health comprises health sensors which need to get their data into the cloud through middleware acting as a gateway. Health sensor devices are having issues pertaining interoperability, hence a middleware is needed to resolve them. However, there are Denial of Service (DoS) threats impending the IoT middleware, which can be problems for its availability. Due to DoS attacks, sensors' data would not be able to reach the middleware. An Intrusion Detection System (IDS) can be addressed as a solution for the attack. Since the IoT middleware is resource constrained, an IDS with lightweight algorithm is needed as solution. J48 machine learning algorithm is tested as reliable for use in constrained devices, so at this study we equipped the middleware using a lightweight IDS based on J48 algorithm to overcome the DoS menace. The test results said that the IDS could capture 75% of network packets. It had been 100% successful in detecting the intrusion, though.