On state encoding against power analysis attacks for finite state controllers

Abstract

Finite-state controllers are central to the design of numerous small-scale electronic appliances used in home automation, environment/infrastructure monitoring, health care and emerging safety-critical systems such as drones and self-driven cars. It is estimated that there will be 50 billion small-scale IoT devices by 2020. These devices, however, are extremely vulnerable to side-channel attacks, therefore low-cost, low-power defense methods are highly desirable. This paper presents an effective method for secure state encoding of finite-state machine (FSM) based controllers to defend against power analysis attacks. Given a user-defined graded security metric, we derive constrained state encoding for the FSM controllers to mitigate information leakage through the power side-channel, resulting in low-power designs. Experimental results using over 100 FSMs from BenGen and MCNC benchmark suites show a graded increase in encoding length (40–70% for restructured FSMs) depending on the level of security chosen. The mutual information between power side-channel and both Hamming attack models varies between 0 and 2.