Leaking your engine speed by spectrum analysis of real-Time scheduling sequences

Abstract

Abstract This paper identifies and studies a new security/privacy issue for automobile vehicles. Specifically, attackers can infer the engine speed of a vehicle by observing and analyzing the real-time scheduling sequences on the Engine Control Unit (ECU). First, we present the problem model of engine-triggered task executed on ECU. And then, we introduce two Engine-triggered Task Period Tracing methods (DFT-based ETPT and FRSP-based ETPT) to infer the period variation of engine-triggered task. Finally, simulation experiments are conducted to demonstrate the effect of this new timing side-channel information leakage with our proposed methods.