Adaptive control‐theoretic detection of integrity attacks against cyber‐physical industrial systems

Abstract

The use of control‐theoretic solutions to detect attacks against cyber‐physical industrial systems is a growing area of research. Traditional literature proposes the use of control strategies to retain, eg, satisfactory close‐loop performance, as well as safety properties, when a communication network connects the distributed components of a physical system (eg, sensors, actuators, and controllers). However, the adaptation of these strategies to handle security incidents is an ongoing challenge. In this paper, we survey the advantages of a watermark‐based detector against some integrity attacks as well as the weaknesses against other attacks. To cover these weaknesses, we propose a new control and security strategy that complements the watermark‐based detector. We validate the detection efficiency of the new strategy via numeric simulation. Experimental results are also presented by using a laboratory testbed based on supervisory control and data acquisition industrial protocols.

Geocaching-inspired Resilient Path Planning for Drone Swarms

“…• Zombie drones are drones from the initial swarm which get hacked by a remote adversary (e.g., a remote attacker conducting a cyber-physical attack [25]- [27]). Zombie drones become perpetrators of GPS attacks (i.e., jamming and spoofing attacks) with the objective of disrupting the remaining drones of the swarm.…”

“…These two attacks are used in our simulation experiments to validate the feasibility of our path planning algorithm to correct the intentional disruption of a series of malicious drones in a swarm, under a linear path mobility experiment. The threat model consists of zombie drones, under the control of a remote adversary conducting a cyber-physical attack [25]- [27]. The zombie drones perpetrate GPS attacks on their neighboring drones of the swarm in order to disrupt their mission.…”

The Use of the Blockchain Technology and Digital Watermarking to Provide Data Authenticity on a Mining Enterprise

“…There have appeared a number of papers devoted to the hiding of information in the data of wireless sensor networks, which is the most urgent for the mining sector. As an example, we will note the papers [ 27 , 28 , 29 , 30 , 31 , 32 , 33 , 34 ]. Within the limits of the present research, similar methods can be used for the authentication of mining service machine data.…”

Cyber‐physical architecture assisted by programmable networking

“…Notice that such anomalies can either be unintentional failures, or stealthy attacks disguised by intentional modifications produced by a malicious entity that intends to disrupt the physical system. Assuming a system with appropriate measure to detect both failures and attacks, the PN controller can react to those anomalies, by enforcing some mitigation policies. At the data domain, we have network probes and effectors , conducting data monitoring—if instructed by the control domain.…”

“…Control‐theoretic solutions, combined with network computing security techniques, can lead to powerful solutions to cover both physical and cyber‐physical attacks at the same time. Nevertheless, guaranteeing the resilience of these systems (ie, to keep offering critical functionality under attack conditions) is still an open and critical problem to solve . We argue that the use of Programmable Networking (PN) is a plausible solution to efficiently handle such a problem.…”