Stacked LSTM Based Anomaly Detection in Time-Critical Automotive Networks
Published in Springer Nature, 2023
Recommended citation: V. K. Kukkala, S. V. Thiruloga, and S. Pasricha, "Stacked LSTM Based Anomaly Detection in Time-Critical Automotive Networks," in Machine Learning and Optimization Techniques for Automotive Cyber-Physical Systems, Springer Nature, 2023.
Abstract
Today’s vehicles are increasingly connected with various external systems (e.g., roadside beacons, and other vehicles) to meet the goals of autonomy, which makes them highly vulnerable to multiple cyber-attacks. Moreover, the increased complexity of automotive applications and the in-vehicle networks resulted in poor attack visibility, which makes detecting such attacks particularly challenging in automotive systems. In this chapter, we present a novel anomaly detection framework called LATTE to detect cyber-attacks in Controller Area Network (CAN) bus based automotive systems. Our proposed LATTE framework uses a stacked Long Short Term Memory (LSTM) predictor network with a novel attention mechanism to learn the normal operating behavior at design time. At runtime, LATTE employs a novel detection scheme (also trained at design time) to detect various cyber-attacks (as anomalies). Moreover, we evaluate our proposed LATTE framework under different automotive attack scenarios and present a detailed comparison with the best-known prior works in this area to demonstrate the effectiveness of our approach.