Probabilistic denial of service attack against remote state estimation over a Markov channel in cyber-physical systems

Remote state estimation in which wireless sensors transmit data to a remote device for estimating the state of a physical process is an important issue in cyber-physical systems (CPS). Since the wireless channel is open-access, adversaries can easily launch denial-of-service (DoS) attacks and cause significant estimation performance deterioration, regardless of the use of cryptographic techniques for sensor data security. Most existing studies on DoS attacks in CPS often assume perfect channels. In this paper, we study the impact and strategies of DoS attacks in a more practical scenario where the wireless channel suffer from stochastic background traffics. We propose a probabilistic attack scheme in which the attacker senses the channel state and only executes DoS when the sensing result indicates an idle channel. Specifically, with the proposed probabilistic attack scheme, the attacker launches DoS with a certain probability once the channel is sensed idle. We theoretically analyze the stability of the estimation process and investigate the problem of optimal attack strategy under attackers energy constraint.