Securing Mobile CPSs against Stealthy Attacks
pdf
ABSTRACT
As Cyber-Physical Systems (CPSs) employing mobile nodes continue to integrate into the physical world, ensuring their safety and security become crucial goals. Due to their mobility, real-time, energy and safety constraints, coupled by their reliance on communication mediums that are subject to interference and intentional jamming, the projected complexities in Mobile CPSs will far exceed those of traditional computing systems. Such increase in complexity widens the malicious opportunities for adversaries and with many components interacting together, distinguishing between normal and abnormal behaviors becomes quite challenging.
The research work in this project falls along two main thrusts: (1) identifying stealthy attacks and (2) developing defense mechanisms. In this presentation, we present a theoretical framework that enables the identification of stealthy attacks in a systematic manner. An adversary solves Markovian Decision Processes problems to identify optimal and suboptimal attack policies in which he/she interferes with a well-chosen subset of signals that are based on the state of the system. We apply approximate policy iteration algorithms to derive potent attack policies. The effects of the attacks are assessed through different instantiations of damage and cost metrics. We expose attack policies on intelligent transportation systems employed in vehicular networks and on target tracking scenarios using autonomous robots.
Award ID: 1149397
Submitted by Katie Dey
on
ABSTRACT
As Cyber-Physical Systems (CPSs) employing mobile nodes continue to integrate into the physical world, ensuring their safety and security become crucial goals. Due to their mobility, real-time, energy and safety constraints, coupled by their reliance on communication mediums that are subject to interference and intentional jamming, the projected complexities in Mobile CPSs will far exceed those of traditional computing systems. Such increase in complexity widens the malicious opportunities for adversaries and with many components interacting together, distinguishing between normal and abnormal behaviors becomes quite challenging.
The research work in this project falls along two main thrusts: (1) identifying stealthy attacks and (2) developing defense mechanisms. In this presentation, we present a theoretical framework that enables the identification of stealthy attacks in a systematic manner. An adversary solves Markovian Decision Processes problems to identify optimal and suboptimal attack policies in which he/she interferes with a well-chosen subset of signals that are based on the state of the system. We apply approximate policy iteration algorithms to derive potent attack policies. The effects of the attacks are assessed through different instantiations of damage and cost metrics. We expose attack policies on intelligent transportation systems employed in vehicular networks and on target tracking scenarios using autonomous robots.
Award ID: 1149397