A Review of Detection, Evolution, and Data Reconstruction Strategies for False Data Injection Attacks in Power Cyber-Physical Systems

TL;DR

This paper reviews detection, evolution, and data reconstruction strategies for false data injection attacks in power cyber-physical systems, emphasizing challenges and future directions to improve system resilience and security.

Contribution

It provides a comprehensive overview of current methods, identifies key challenges, and proposes future research directions for enhancing FDIA detection and data reconstruction in power CPS.

Findings

Highlighting cross-domain coordination and multi-temporal evolution in FDIA

Discussing challenges like poor interpretability and data imbalance

Proposing advanced state-aware and action-control reconstruction techniques

Abstract

The integration of information and physical systems in modern power grids has heightened vulnerabilities to False Data Injection Attacks (FDIAs), threatening the secure operation of power cyber-physical systems (CPS). This paper reviews FDIA detection, evolution, and data reconstruction strategies, highlighting cross-domain coordination, multi-temporal evolution, and stealth characteristics. Challenges in existing detection methods, including poor interpretability and data imbalance, are discussed, alongside advanced state-aware and action-control data reconstruction techniques. Key issues, such as modeling FDIA evolution and distinguishing malicious data from regular faults, are identified. Future directions to enhance system resilience and detection accuracy are proposed, contributing to the secure operation of power CPS.