Energy audition based cyber-physical attack detection system in IoT
Yang Meng Lin Shi, Fangyu Li, Wen‐Zhan Song, Xiangyang Li, Jin Ye · Proceedings of the ACM Turing Celebration Conference - China · 2019
In this paper, we propose an attack detection framework in the Internet of Things (IoT) devices. The framework applies a data-centric method to process the energy consumption data and classify the attack status of the monitored device. We implement the framework in real hardware, and emulate common types of attacks to evaluate the performance of the attack detection framework. Due to the characteristic of the energy data, not only cyber attacks but also physical attacks such as heating are also emulated and tested. To shorten the detection time, a two-stage strategy is also proposed to first apply a short time window for a rough detection, then a long time window to the fine detection of anomalies. The accuracy of short-term detection is 90%, while in the long-term detections the accuracy reaches 99.5%. Due to the nature of information from energy consumption data, the framework is more secure in cases the kernel of the device is already compromised.