Scalable Wi-Fi Intrusion Detection for IoT Systems

The pervasive and resource-constrained nature of Internet of Things (IoT) devices makes them attractive to be targeted by different means of cyber threats. There are a vast amount of botnets being deployed every day that aim to increase their presence on the Internet for realizing malicious activities with the help of the compromised interconnected devices. Therefore, monitoring IoT networks using intrusion detection systems is one of the major countermeasures against such threats. In this work, we present a machine learning based Wi-Fi intrusion detection system developed specifically for IoT devices. We show that a single multi-class classifier, which operates on the encrypted data collected from the wireless data link layer, is able to detect the benign traffic and six types of IoT attacks with an overall accuracy of 96.85%. Our model is a scalable one since there is no need to train different classifiers for different IoT devices. We also present an alternative attack classifier that outperforms the attack classification model which has been developed in an existing study using the same dataset.

[1]  Xianbin Wang,et al.  Machine learning techniques for intrusion detection on public dataset , 2016, 2016 IEEE Canadian Conference on Electrical and Computer Engineering (CCECE).

[2]  Gaël Varoquaux,et al.  Scikit-learn: Machine Learning in Python , 2011, J. Mach. Learn. Res..

[3]  Pierre Geurts,et al.  Extremely randomized trees , 2006, Machine Learning.

[4]  Xingyu Wang,et al.  Distributed intrusion detection system based on data fusion method , 2004, Fifth World Congress on Intelligent Control and Automation (IEEE Cat. No.04EX788).

[5]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[6]  Tianqi Chen,et al.  XGBoost: A Scalable Tree Boosting System , 2016, KDD.

[7]  Ali A. Ghorbani,et al.  A detailed analysis of the KDD CUP 99 data set , 2009, 2009 IEEE Symposium on Computational Intelligence for Security and Defense Applications.

[8]  Yanxia Sun,et al.  A Deep Learning Method With Filter Based Feature Engineering for Wireless Intrusion Detection System , 2019, IEEE Access.

[9]  A. O. Eboka,et al.  Genetic Algorithm Rule-Based Intrusion Detection System (GAIDS) , 2012 .

[10]  Vrizlynn L. L. Thing,et al.  IEEE 802.11 Network Anomaly Detection and Attack Classification: A Deep Learning Approach , 2017, 2017 IEEE Wireless Communications and Networking Conference (WCNC).

[11]  Mousa Al-Akhras,et al.  WSN-DS: A Dataset for Intrusion Detection Systems in Wireless Sensor Networks , 2016, J. Sensors.

[12]  Nils Ole Tippenhauer,et al.  WADAC: Privacy-Preserving Anomaly Detection and Attack Classification on Wireless Traffic , 2018, WISEC.

[13]  Md. Abu Naser Bikas,et al.  An Implementation of Intrusion Detection System Using Genetic Algorithm , 2012, ArXiv.

[14]  Georgios Kambourakis,et al.  Intrusion Detection in 802.11 Networks: Empirical Evaluation of Threats and a Public Dataset , 2016, IEEE Communications Surveys & Tutorials.

[15]  Teresa F. Lunt,et al.  Knowledge-based intrusion detection , 1989, [1989] Proceedings. The Annual AI Systems in Government Conference.

[16]  Pete Burnap,et al.  A Supervised Intrusion Detection System for Smart Home IoT Devices , 2019, IEEE Internet of Things Journal.

[17]  In Lee,et al.  The Internet of Things (IoT): Applications, investments, and challenges for enterprises , 2015 .

[18]  V. Vanitha,et al.  A novel rule based intrusion detection framework for Wireless Sensor Networks , 2013, 2013 International Conference on Information Communication and Embedded Systems (ICICES).

[19]  Taehwan Park,et al.  An Effective Classification for DoS Attacks in Wireless Sensor Networks , 2018, 2018 Tenth International Conference on Ubiquitous and Future Networks (ICUFN).

[20]  Aboubaker Lasebae,et al.  An efficient deep learning model for intrusion classification and prediction in 5G and IoT networks , 2019, 2019 53rd Annual Conference on Information Sciences and Systems (CISS).

[21]  Miad Faezipour,et al.  Enhancing Wireless Intrusion Detection Using Machine Learning Classification with Reduced Attribute Sets , 2018, 2018 14th International Wireless Communications & Mobile Computing Conference (IWCMC).