A new Intrusion Detection System for Secured IoT/IIoT Networks based on LGBM

Authors : İlhan Fırat KILINÇER, Oğuzhan KATAR

Pages : 321-328

Doi:10.29109/gujsc.1173286

View : 107 | Download : 91

Publication Date : 2023-06-23

Article Type : Research Paper

Abstract :The Internet of Things insert ignore into journalissuearticles values(IoT); is one of the technologies used in many fields today. Cyber attacks against IoT/Industrial IoT insert ignore into journalissuearticles values(IIoT); networks, which are increasingly used thanks to the convenience it provides, are constantly increasing. Detection of attacks against IoT/IIoT networks is one of the popular topics recently. The development of a dataset for IoT applications is essential for the intrusion detection in IoT networks. In this context, the ToN_IoT dataset created in the laboratory of UNSW Canberra insert ignore into journalissuearticles values(Australia); is one of the most comprehensive datasets that can be used to detect cyber attacks on IoT networks. In this study, fridge, garage door, GPS tracker, modbus, motion light, weather, thermostat datasets related to IoT sensors from ToN_IoT datasets were used. The datasets used were subjected to multi-class classification with the Light Gradient Boosting Machine insert ignore into journalissuearticles values(LGBM); classifier proposed in the study. The obtained results were compared with the literature and it was seen that the proposed method provided the highest classification performance in the literature. It has been determined that the proposed method is effective in preventing cyber attacks on IoT/IIoT networks.
Keywords : Internet of Things, Light GBM, ToN IoT, cyber security