Secure detection with correlated binary sensors

Sensor networks use binary measurements and state estimations for several reasons, including communication and processing overheads. Such a state estimator is vulnerable to attackers that can hijack a subset of the sensors in an effort to change the state estimate. After exhibiting a simulation that demonstrates the possible effect of integrity cyberphysical systems, this paper extends the authors´ methodology for designing the detectors resilient to integrity attacks, using the concept of invariant sets, to systems where the sensor measurements are not independent. In cyberphysical systems, the sensors in question monitor a system constrained to obey physical laws, so that physical quantities measured by and the noise in each sensor will be correlated to the sensors close to it. Further increase in the confidence of the estimate can be achieved by considering these correlations. This paper focuses on modeling the correlation between the sensors and its ramifications on the worst-case probability of detection.