Optimal DoS Attacks on Remote State Estimation with Continuous Action Spaces

This paper investigates the optimal denial-of-service (DoS) attack against a cyber-physical system (CPS). In this scenario, sensor measurements are transmitted to a remote estimator over a network that can be disrupted by a malicious attacker. We formulate the optimal DoS attack problem as a Markov decision process (MDP) with a continuous action space and analyze the properties of the value function. Furthermore, we provide a condition for the existence of an optimal DoS attack strategy. Our findings demonstrate that an optimal strategy in a continuous action space can achieve superior performance compared to one in a discrete action space.