New Trick Stops Hackers from Mapping Out Power Grids

Large-scale networked systems face growing threats from adversaries who can infer system dynamics from just a few compromised nodes. This paper tackles how to block that inference when only limited state information is available—unlike existing state-feedback methods that require full state access and are impractical for large networks. The authors propose multiple control strategies under partial state measurements, including an output-feedback approach that preserves a subset of open-loop eigenvalues and an observer-based design that reconstructs the system state, preserves all eigenvalues, and offers greater flexibility in sensor placement. They also extend the observer-based design to a distributed framework. Numerical examples demonstrate the scope and validity of the proposed methods.