Injecting Control Commands Through Sensory Channel: Attack and Defense

Industrial Control System (ICS) is responsible for controlling and managing critical infrastructures like nuclear plants and power plants. ICS is equipped with various devices like communication media, Programmable Logic Controller (PLC), actuators, and sensors. Sensors are used to measure a physical phenomenon and send these measurements through the sensory channel to the control devices so they can make decisions on the movements of the actuators in the systems. While tampering with the sensor measurements has been the focus of many studies, there is some research that has concentrated on misusing a sensor and sensory channel as an axillary attack device to inject control command into the system. Previous research experiments have demonstrated the feasibility of sensory channel misuse. This thesis has focused on the design and development of attack models and creating a practical detection system. In order to create the detection framework, two signal patterns, which can be used for injecting commands into the system, are created. The patterns use the time and energy between two consecutive rising and falling discrete signals to encode one-bit data. Further, an Intrusion Detection System (IDS) is developed to detect the malicious signals. The developed IDS utilizes the Fourier and Wavelet transformations to build features of examined signals