Adaptive hybrid intrusion detection system for crowd sourced multimedia internet of things systems

Abstract

The rapidly increasing volume of lightweight devices in Internet of Things (IoT) environment needs a strong Intrusion Detection System (IDS). Conventional IDS cannot be applied directly in IoT networks due to various communication architectures, standards, technologies, and environment specific services. The main problem with current IDS and handling techniques is that they can’t adapt to service changes in real-time. To overcome this open challenge, adaptive hybrid IDS based on timed automata controller approach is proposed in this paper. Proposed Hybrid IDS have additional knowledge in relation to frequent multimedia file formats and use this knowledge to carry out a comprehensive analysis of packets carrying multimedia files. Crowd sourcing online repository for signature based malicious pattern set generation is designed and self-tuning timed automaton is developed to detect the intruder in IoT networks. From the experimental results, it is evident that our proposed method, an adaptive hybrid IDS suit smart city applications and are accurate (99.06%) in detecting Denial of Service (DoS) attacks, control hijacking attacks, zero day attacks, and replay attacks in IoT environments.