Beyond the Lens: False Data Injection Attacks on IIoT-Cameras through MQTT Manipulation
Wael Alsabbagh, Chaerin Kim, Nitin Sanjay Patil, Peter Langendörfer · 2024
In the era of Industrial Internet of Things (IIoT), where interconnected devices streamline industrial processes, safeguarding the integrity and security of these systems becomes imperative. This paper delves into the exploration of vulnerabilities within IIoT-based systems, focusing on the manipulation of camera views through MQTT protocol exploitation. Our work unveils a sophisticated attack vector targeting the interaction between IIoT cameras and central control systems, leveraging weaknesses within the MQTT protocol. Through a meticulously designed attack scenario, we demonstrate how attackers can surreptitiously inject and manipulate camera feeds, presenting fabricated imagery to deceive operators and compromise system integrity. By exploiting vulnerabilities in MQTT communication, coupled with strategic manipulation of camera views, adversaries can obscure their activities, posing significant risks to industrial operations. Furthermore, we discuss comprehensive security measures and countermeasures to fortify IIoT systems against such attacks, aiming to mitigate potential threats and ensure the resilience of industrial infrastructures. All our attack codes as well as a proof-of-concept are publicly available for further research.