Systematic Intrusion Detection Technique for an In-vehicle Network Based on Time-Series Feature Extraction

摘要

In this paper, we propose a systematic intrusion detection algorithm based on time-series feature extraction for an in-vehicle network. Since packet-type valid data are transmitted inside an in-vehicle network periodically, illegal data due to unauthorized intrusion attack can be easily and uniformly detected by using periodical time-series feature of valid data, where recurrent neural network is a key tool to efficiently extract their time-series feature. In fact, through an evaluation using data acquired from actual vehicles, we show that the proposed method can detect typical intrusion attack patterns such as data modification attack and injection attack.