Manchuri, Yeswanth and Khan, Mohammed Zafar Ali
(2018)
Improving the Detection Rate of Cosine Similarity
detector in False Data Injection Attacks in Control
Systems.
Masters thesis, Indian Institute of Technology Hyderabad.
Abstract
Cyber Physical Systems are more vulnerable to attacks than the conventional systems because of
the integrated nature of the cyber as well as physical environment.Replay attacks and False data
injection attacks are in particular harmful because of their deceptive nature to traditional detectors.
A popular traditional detector is Chi Square Detector which detects based on the statistics of devi-
ations of the residual i.e. di�erence of observed measurement and estimated measurement. Since the
statistics is not changed in the attacks mentioned above, Chi square detector fails to detect these.
However, the Cosine detector proposed by [1] also fails in detecting these attacks in control system
scenario. So in this work, we will show why the cosine detector fails to detect them and design a
method to improve the detection rate of the Cosine Detector.
v
Actions (login required)
 |
View Item |
Altmetric
Altmetric