Optimal Innovation-Based Attacks: The Tradeoff Between Alarm Rate and Attack Performance

In this paper, the problem of designing uniformly optimal innovation-based attacks against \(\chi ^2\) detectors is investigated. In the existing researches, the attack performance is usually improved by extending the attack form. Differently, this paper focuses on improving the attack performance by optimizing the bound of the stealthiness constraints. Then the optimal attack scheme is obtained by solving the non-convex optimization problem. Accordingly, the tradeoff between the attack performance and the alarm rate has been provided and the uniformly optimal attack scheme has been derived. Finally, simulation results are provided to demonstrate the feasibility and effectiveness of the proposed strategy.