On the Use of Artificial Noise for Secure State Estimation in the Presence of Eavesdroppers
Alex S. Leong, Adrian Redder, Daniel E. Quevedo, Subhrakanti Dey · 2018
The problem of remote state estimation in the presence of eavesdroppers has recently been investigated in the literature. For unstable systems it has been shown that one can keep the expected estimation error covariance bounded, while the expected eavesdropper error covariance becomes unbounded in the infinite horizon, using schemes based on transmission scheduling. In this paper we consider an alternative approach to achieve security, namely injecting noise into sensor transmissions, similar to the artificial noise technique used in physical layer security for wireless communications. Numerical results demonstrate significant performance improvements using this approach, with respect to the trade-off between the expected estimation error covariance and expected eavesdropper covariance.