A Deep Learning Approach for False Data Injection Attacks Detection in Smart Water Infrastructure

Cyber-Physical systems (CPS) represent a sophisticated integration of digital technologies with physical processes, particularly vital in critical environments such as smart water infrastructures, which require advanced monitoring and control systems to guarantee safe and resilient operations, especially in the context of attacks. This study introduces a novel unsupervised deep learning approach for detecting false data injection (FDI) attacks in smart water infrastructures. The method employs Long Short-Term Memory (LSTM) networks and Autoencoders to discern the legitimate behavior of time-series water level sensor data. We evaluate this approach using the Mincio River water system in Italy as a case study, employing publicly available data augmented with synthetic—yet realistic—random, replay, and advanced attack scenarios. The experimental results demonstrate the effectiveness of the proposed method in distinguishing anomalies from legitimate data, highlighting its potential for enhancing the security of smart water systems.