A New Anomaly-Based Intrusion Detection System for MIL-STD-1553

MIL-STD-1553 is a communication standard developed by the US DoD in 1975 and used in military aircraft, land, and space vehicles. It enables computers and devices to interact with each other. MIL-STD-1553 is used in safety-critical systems due to its dual redundant bus structure, high reliability, and low fault rate. While the standard was considered to be secure when it was developed, MIL-STD-1553-based systems have become vulnerable and they are an easy target for attackers. In this research, we propose a new anomaly-based intrusion detection system for MIL-STD-1553 that uses machine learning for classification purposes. We experimentally evaluated the proposed system against cyber attacks. We observed that Stochastic Gradient Descent provides adequate results for the detection of intrusions on the MIL-STD-1553 bus. Overall experimental results show that the proposed system can be used to detect intrusions on MIL-STD-1553-based communications.