Optimal jamming power allocation against remote state estimation

This paper investigates a remote state estimation problem, where a smart sensor observes the state of a physical plant, locally estimates the state with Kalman filter, and then sends its local estimation data to a remote estimator through a communication network. There is a jamming attacker in the communication network who intentionally blocks the network with the purpose of deteriorating estimation quality subject to the jamming energy constraint. From the viewpoint of jamming attacker, in order to maximize attack effect, we focus on the questions “when to jam the communication network”, and “how much power the attacker should adopt to jam the network”. After theoretic analysis, we point out that any grouped together attack schedule is optimal for the given jamming power. We provide a close form of jamming power when a sufficient condition holds, and the corresponding jamming effect on remote state estimation is explicitly presented. For a general case, we design a jamming power allocation algorithm and show the computational complexity of the proposed algorithm is not worse than O(T), where T is the length of the time horizon considered.