Behavior-Rule Specification-based IDS for Safety-Related Embedded Devices in Smart Home

The rapid development of Information & Communication Technology (ICT) has accelerated the development of Internet of Things (IoT) devices, which in turn increases the application of these devices to build a safe home environment for all. However, there are security concerns that need to be resolved to overcome the possible invasion of privacy (e.g., IoT camera) and malicious access to safety-related devices. If the protection against external intrusion is weak, it can pose a fatal threat. Hence, this paper researched the security dedicated to the smart home environment. To date, as a representative security solution suitable for the limited-resource nature of IoT, there exist an Intrusion Detection System (IDS). But, by careful analysis of this existing IDS, it was found that there are limitations in supporting security in resource-constrained IoT-assisted smart homes. In this paper, we designed a behavior-rule specification-based IDS to provide security suitable for the said environment. The proposed approach does not only copes with unknown threats, which is a disadvantage of the existing IDS but also lowers the false positive rate. Finally, the IDS performance was validated in MATLAB tool, then compared with the detection accuracy of contemporary classification algorithms, namely Support Vector Machine and Decision Tree.