Resilient Cyber-Security Approach For Aviation Cyber-Physical Systems Protection Against Sensor Spoofing Attacks

摘要

The aviation industries are transitioning from conventional aircraft systems to Aviation Cyber-Physical Systems (ACPS) based aircraft. However, like any Cyber-Physical Systems (CPS), the ACPS are vulnerable to cyber-attacks that can be mounted by adversaries through the communication network infrastructure. This paper proposes a novel and resilient security protocol for detecting and defending ACPS against sensor spoofing cyber-attacks. First, a communication environment was developed to establish an aircraft Networked Control System (NCS) using the SimEvents toolbox. Then, a cyber-attack detection algorithm based on the positive selection of the Artificial Immune System (AIS) approach was developed and used to detect and drop suspicious communication packets on the aircraft network traffic. Finally, the NCS and the detection algorithm were integrated and tested on real cyber-security attack scenarios. The algorithm's accuracy was 0.96 based on the True Positive and True Negative algorithm detection rate. For further defending the aircraft against cyber-attacks, Nonlinear Autoregressive Exogenous (NARX) algorithm was developed to reconstruct or estimate the network dropped packets. The estimation accuracy of the NARX reached 0.99 using the coefficient of determination (R-value) based on the linear regression approach. The real-time simulation test results showed that the sensor spoofing cyber-attack was successfully detected. Also, the communication network of the ACPS was defended against the attack because the ACPS was maintaining the normal performance during the course of the cyber-attack.