Window-Based Statistical Analysis Of Timing Subcomponents For Efficient Detection Of Malware In Life-Critical Systems

• Published in: 2019 Spring Simulation Conference (SpringSim) pp. 1 - 12
• Main Authors: Carreon, Nadir, Gilbreath, Allison, Lysecky, Roman
• Format: Conference Proceeding
• Language: English
• Published: SCS 01.04.2019
• Subjects: Anomaly detection; Embedded system security; Embedded systems; Malware; Monitoring; Non-intrusive hardware; Runtime; Timing; Timing-based threat detection
• DOI: 10.23919/SpringSim.2019.8732899

Securing life-critical embedded systems, particularly medical devices, requires both proactive security measures that prevent intrusions and reactive measures that detect intrusions. This paper presents a novel model for specifying the normal timing for operations in software applications using cumulative distribution functions of timing subcomponent within sliding execution windows. We present a probabilistic formulation for estimating the presence of malware for individual operations by monitoring the internal timing of the different components of the system, and we define thresholds to minimize false positives based on training data. Experimental results with a smart connected pacemaker and three sophisticated mimicry malware scenarios demonstrate improved performance and accuracy compared to state-of-the-art timing-based malware detection