Journal article
False data injection attacks with complete stealthiness in cyber–physical systems: A self-generated approach
Automatica (Oxford), Vol.120, p.109117
10/2020
DOI: 10.1016/j.automatica.2020.109117
Abstract
In this paper, we consider the security problem of dynamic state estimations in cyber–physical systems (CPSs) when the sensors are compromised by false data injection (FDI) attacks with complete stealthiness. The FDI attacks with complete stealthiness can completely remove its influences on monitored residuals, which have better stealthy performance against residual-based detectors than existing FDI attacks. Based on self-generated FDI attacks that are independent of real-time data of CPSs, we propose the necessary and sufficient condition of attack parameters such that FDI attacks can achieve complete stealthiness. Furthermore, we introduce the energy stealthiness of FDI attacks, which is a special case of complete stealthiness and makes the accumulated attack energy on residuals is bounded. Then, the existence and design conditions of FDI attacks with energy stealthiness are given. Finally, the superiority of the FDI attacks with complete stealthiness is demonstrated by the IEEE 6 bus power system.
Details
- Title: Subtitle
- False data injection attacks with complete stealthiness in cyber–physical systems: A self-generated approach
- Creators
- Tian-Yu Zhang - College of Information Science and Engineering, Northeastern University, Liaoning, 110819, PR ChinaDan Ye - Northeastern UniversityTianyu Zhang - Computer Science
- Resource Type
- Journal article
- Publication Details
- Automatica (Oxford), Vol.120, p.109117
- Publisher
- Elsevier Ltd
- DOI
- 10.1016/j.automatica.2020.109117
- ISSN
- 0005-1098
- eISSN
- 1873-2836
- Grant note
- 61773097; U1813214; 61621004 / National Natural Science Foundation of China (http://dx.doi.org/10.13039/501100001809) XLYC1907035 / LiaoNing Revitalization Talents Program N2004026; N2004027 / Fundamental Research Funds for the Central Universities (http://dx.doi.org/10.13039/501100012226)
- Language
- English
- Date published
- 10/2020
- Academic Unit
- Computer Science
- Record Identifier
- 9984696564802771
Metrics
1 Record Views