@article{39,
  author = {Carlos Moreno and Sebastian Fischmeister},
  title = {Non-Intrusive Runtime Monitoring Through Power  Consumption to Enforce Safety and Security Properties in Embedded Systems},
  abstract = {<p>The increasing complexity and connectivity of modern em-<br />
bedded systems highlight the importance of runtime monitoring to en-<br />
sure correctness and security. This poses a significant challenge, since<br />
monitoring tools can break extra-functional requirements such as timing<br />
constraints. Non-intrusive program tracing through side-channel analy-<br />
sis techniques have recently appeared in the literature and constitute<br />
a promising approach. Existing techniques, however, exhibit important<br />
limitations.<br />
In this paper, we present a novel technique for non-intrusive program<br />
tracing from power consumption, based on a signals and system analy-<br />
sis approach: we view the power consumption signal as the output of a<br />
system with the power consumption of training samples as input. Using<br />
spectral analysis, we compute the impulse response to identify the sys-<br />
tem; the intuition is that for the correct training sample, the system will<br />
appear close to a system that outputs a shifted copy of the input signal,<br />
for which the impulse response is an impulse at the position correspond-<br />
ing to the shift. We also use the Control Flow Graph (CFG) from the<br />
source code to constrain the classifier to valid sequences only, leading to<br />
substantial performance improvements over previous works.<br />
Experimental results confirm the effectiveness of our technique and show<br />
its applicability to runtime monitoring. The experiments include tracing<br />
programs that execute randomly generated sequences of functions as well<br />
as tracing a real application developed with SCADE. The experimental<br />
evaluation also includes a case-study as evidence of the usability of our<br />
technique to detect anomalous execution through runtime monitoring.</p>
},
  year = {2017},
  journal = {Formal Methods in Software Design (FMSD)},
}