PhD Forum 2022 - Design of False Data Injection Attacks and their Detection and Mitigation in Cyber-Physical Systems

Abstract

False data injection (FDI) attacks are more common in Cyber-Physical Systems (CPSs). In an FDI attack, an attacker can inject an attack sequence at any location in the CPS. Rather than analyzing attacks at a specific location, it is necessary to analyze the system’s behavior in the presence of FDI attacks at multiple locations. Both the cyber and physical systems are vulnerable to FDI attacks. In this work, we study FDI attacks at multiple locations in a CPS. We examine FDI attacks at the sensor, actuator, and physical system. In the worst-case scenario, an attacker could guess the system parameters and remain stealthy by carefully planning the attack sequences. An attacker may inject zero-mean or non-zero-mean attack sequences, which brings the system to an unsafe state. Therefore, it is required to design a suitable detection and mitigation schemes to secure the CPS against FDI attacks. We design an FDI attack scheme in a single sensor and actuator framework and multiple sensors and actuators framework of the CPS. A detection and mitigation strategy against FDI attacks is also proposed