TY - GEN
T1 - Implementing Lightweight IoT-IDS on Raspberry Pi Using Correlation-Based Feature Selection and Its Performance Evaluation
AU - Soe, Yan Naung
AU - Feng, Yaokai
AU - Santosa, Paulus Insap
AU - Hartanto, Rudy
AU - Sakurai, Kouichi
N1 - Funding Information:
Acknowledgments. The authors are grateful for the financial support provided by AUN/SEED-Net Project (JICA). This research is also partially supported by Strategic International Research Cooperative Program, Japan Science and Technology Agency (JST), JSPS KAKENHI Grant Numbers JP17K00187 and JP16K00132.
Publisher Copyright:
© 2020, Springer Nature Switzerland AG.
PY - 2020
Y1 - 2020
N2 - The application of many IoT devices is making our world more convenient and efficient. However, it also makes a large number of cyber-attacks possible because most IoT devices have very limited resources and cannot perform ordinary intrusion detection systems. How to implement efficient and lightweight IDS in IoT environments is a critically important and challenging task. Several detection systems have been implemented on Raspberry Pi, but most of them are signature-based and only allow limited rules. In this study, a lightweight IDS based on machine learning is implemented on a Raspberry Pi. To make the system lightweight, a correlation-based feature selection algorithm is applied to significantly reduce the number of features and a lightweight classifier is utilized. The performance of our system is examined in detail and the experimental result indicates that our system is lightweight and has a much higher detection speed with almost no sacrifice of detection accuracy.
AB - The application of many IoT devices is making our world more convenient and efficient. However, it also makes a large number of cyber-attacks possible because most IoT devices have very limited resources and cannot perform ordinary intrusion detection systems. How to implement efficient and lightweight IDS in IoT environments is a critically important and challenging task. Several detection systems have been implemented on Raspberry Pi, but most of them are signature-based and only allow limited rules. In this study, a lightweight IDS based on machine learning is implemented on a Raspberry Pi. To make the system lightweight, a correlation-based feature selection algorithm is applied to significantly reduce the number of features and a lightweight classifier is utilized. The performance of our system is examined in detail and the experimental result indicates that our system is lightweight and has a much higher detection speed with almost no sacrifice of detection accuracy.
UR - http://www.scopus.com/inward/record.url?scp=85064004822&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85064004822&partnerID=8YFLogxK
U2 - 10.1007/978-3-030-15032-7_39
DO - 10.1007/978-3-030-15032-7_39
M3 - Conference contribution
AN - SCOPUS:85064004822
SN - 9783030150310
T3 - Advances in Intelligent Systems and Computing
SP - 458
EP - 469
BT - Advanced Information Networking and Applications - Proceedings of the 33rd International Conference on Advanced Information Networking and Applications AINA-2019
A2 - Takizawa, Makoto
A2 - Enokido, Tomoya
A2 - Barolli, Leonard
A2 - Xhafa, Fatos
PB - Springer Verlag
T2 - 33rd International Conference on Advanced Information Networking and Applications, AINA-2019
Y2 - 27 March 2019 through 29 March 2019
ER -