Quickest detection of false data injection attack in remote state estimation

Sample Path Generation

FDIGenerateDataNew.ipynb generates sample paths for optimizing the threshold value. Number of samples paths can be set using the variable num_paths, the horizon length using N and the probability of FDI attack at each step using theta.

Optimizing for Threshold Gamma

SingleThresholdSGD.py finds the optimal values of the threshold and subject to different probability of false alarm (PFA) rates through an amalgamation of Simultaneous Pertubation Stochastic Approximation and Two-Timescale Approximation algorithms. Number of iterations for each PFA can be set using num_iters and the batch size using batch_size.

Cite

IEEE ISIT 2021 paper:

A. Gupta, A. Sikdar and A. Chattopadhyay, 
"Quickest detection of false data injection attack in remote state estimation," 
2021 IEEE International Symposium on Information Theory (ISIT), 2021, pp. 3068-3073, 
doi: 10.1109/ISIT45174.2021.9518036.