Distributed secure estimation for multisensor systems using intermittent encryption

This paper focuses on the secure problem of distributed multisensor estimation for time-varying systems. The communication between sensors and estimators is facilitated through a network. An intermittent encryption-decryption strategy, where measurement data is probabilistically encrypted using dynamic keys before being transmitted by sensors and then decrypted by estimators, is designed to prevent information leakage and conserve resources. Under the effects of the encryption, the optimal estimator parameters are calculated, and the minimum upper bound of the estimation error covariance is recursively determined. Finally, the proposed distributed secure estimation algorithm is validated through a simulation regarding a DC servo motor system. © 2025 The Franklin Institute