The More You Know: Improving Laser Fault Injection with Prior Knowledge

Abstract

We consider finding as many faults as possible on the target device in the laser fault injection security evaluation. Since the search space is large, we require efficient search methods. Recently, an evolutionary approach using a memetic algorithm was proposed and shown to find more interesting parameter combinations than random search, which is commonly used. Unfortunately, once a variation on the bench or target is introduced, the process must be repeated to find suitable parameter combinations anew.To negate the effect of variation, we propose a novel method combining a memetic algorithm with a machine learning approach called a decision tree. Our approach improves the memetic algorithm by using prior knowledge of the target introduced in the initial phase of the memetic algorithm. In our experiments, the decision tree rules enhance the performance of the memetic algorithm by finding more interesting faults in different samples of the same target. Our approach shows more than two orders of magnitude better performance than random search and up to 60% better performance than previous state-ofthe-art results with a memetic algorithm. Another advantage of our approach is human-readable rules, allowing the first insights into the explainability of target characterization for laser fault injection.

Differential Fault and Algebraic Equation Combined Analysis on PICO

“…In practical experiments, various methods can be employed to inject faults [26][27][28][29][30][31][32][33][34], including variations in the operational voltage, clock, or temperature of the circuit, electromagnetic pulse interference with circuit operation, utilization of laser, UV-ray, X-ray, or focused ion beam (FIB), etc. In addition to this, the rest of the key recovery processes are all accomplished through computer computations.…”

SoK: Assisted Fault Simulation

“…So, in essence, the simulators consider all models indiscriminately or the developer is left to decide. AI models could be used to inform the choice of the parameters for the test cases as done by [7] and [19] which showed that a certain combination of parameters leads to an increase in the number of exploitable faults. A successful AI model would be able to eliminate (to an extent) less probable fault models from the ones which occur more frequently.…”

“…Their solution to narrow down the space of exploitable locations from the vulnerable locations could be beneficial. Krcek et al [19] utilizes genetic algorithms and machine learning to develop a solution that allows them to find more exploitable faults compared to using either of the approaches.…”