Digital Twin IDS Detects Stealthy CAN Bus Attacks in Vehicles

Researchers have developed a novel digital twin-based intrusion detection system (IDS) for vehicle powertrain CAN bus systems. This system models the physical relationships between powertrain signals to predict behavior and identify attacks that manipulate data payloads while maintaining normal communication patterns. The digital twin, trained using a shared-encoder LSTM on Hyundai/Kia CAN logs, demonstrated superior performance over traditional range-and-plausibility baselines, particularly in detecting stealthy payload attacks. AI

IMPACT This research could lead to more robust cybersecurity for connected vehicles by enabling detection of sophisticated, stealthy attacks.

RANK_REASON Academic paper detailing a new method for intrusion detection. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• arXiv
• CAN bus
• digital twin
• Hyundai Motor Company
• Kia
• long short-term memory

• paper
• safety

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →