Machine Learning-Enabled IoT Security: Open Issues and Challenges Under Advanced Persistent Threats

Despite its technological benefits, the Internet of Things (IoT) has cyber weaknesses due to vulnerabilities in the wireless medium. Machine Laming (ML)-based methods are widely used against cyber threats in IoT networks with promising performance. An Advanced Persistent Threat (APT) is prominent for cybercriminals to compromise networks, and it is crucial to long-term and harmful characteristics. However, it is difficult to apply ML-based approaches to identify APT attacks to obtain a promising detection performance due to an extremely small percentage among normal traffic. There are limited surveys that fully investigate AFT attacks in IoT networks due to the lack of public datasets with all types of APT attacks. It is worth bridging the state of the art in network attack detection with APT attack detection in a comprehensive review article. This survey article reviews the security challenges in IoT networks and presents well-known attacks, APT attacks, and threat models in IoT systems. Meanwhile, signature-based, anomaly-based, and hybrid intrusion detection systems are summarized for IoT networks. The article highlights statistical insights regarding frequently applied ML-based methods against network intrusion. Finally, open issues and challenges for common network intrusion and APT attacks are presented for future research.

Automated summary

Despite the technological benefits of the Internet of Things (IoT), there are cyber weaknesses due to vulnerabilities in the wireless medium. Machine Learning (ML)-based methods are effective against cyber threats in IoT networks. However, it is challenging to apply ML-based approaches to detect Advanced Persistent Threat (APT) attacks due to their low occurrence in normal traffic. Limited surveys have been conducted on APT attacks in IoT networks, mainly due to the lack of public datasets. This survey article reviews security challenges, well-known attacks, and intrusion detection methods for IoT networks, with a focus on ML-based approaches.