Secure detection using binary sensors

Cyber-physical systems employing remote sensors and actuators and sparse communication networks are pervading the infrastructure. In this paper we consider a prototypical problem of estimating a binary state using measurements provided by binary sensors. We propose a new approach to estimate the states based on sensor measurements that may have been corrupted by an attacker. The problem is formulated as a minimax problem in which a detector attempts to maximize the probability of detection in case of the worst case attempt by the attacker to minimize this probability. A fixed form of the detector is proposed in the case where the sensors are of equivalent specifications, along with a method to find the optimal detector parameters.