Fault-Attack Aware Microprocessor Extension (FAME): From Concept to Prototype
Speaker(s) : Patrick Schaumont, Virginia Tech
In the Internet of Things, the cyber-world will use a huge number of small embedded computing elements to control and sense the real world. The integrity and trustworthiness of these embedded systems are crucial, and their manipulation has direct consequences to the safety of the applications they support. In this context, we highlight the use of fault injection as a versatile hacking tool to obtain device control and to extract secret data. The traditional technique, differential fault analysis, dates back almost two decades, and many effective variants have since been proposed. We demonstrate how knowledge of the microprocessor architecture can greatly improve the effectiveness of a fault attack on embedded software.
We discuss a generic countermeasure against fault injection attacks, based on micro-architecture enhancements and software support. The FAME (Fault Aware Microprocessor Extension) architecture uses sensors to capture fault injection and uses software to provide a user-defined fault response. FAME also has specialized support to analyze the effect of fault injection into the microprocessor, leading a better insight into the failing of embedded software under a fault attack. We present a prototype ASIC design and preliminary testing data. We demonstrate that, compared to traditional redundancy based techniques, FAME has a lower overhead in area as well as in performance.