Encryption-Decryption-Based Tobit Recursive Filtering for Nonlinear Systems: Handling Censored Measurements

This paper addresses the issue of the secure Tobit recursive filtering (TRF) for nonlinear systems subject to censored measurements. On account of the inherent vulnerability of network-based communication, the measurement data transmitted over networks are susceptible to eavesdropping attacks. In order to prevent information leakage, a novel artificial noise-based encryption-decryption scheme (EDS) is adopted to encrypt the measurement signal before transmission. The focus of the conducted topic is to design an EDS-based TRF algorithm with the aim to ensure that the upper bound of the filtering error covariance is minimized at every sampling instant. Additionally, the uniform boundedness of the filtering error is investigated in the mean-square sense to evaluate the performance of the developed filtering scheme and the corresponding condition is provided. Finally, the efficacy of the presented EDS-based TRF algorithm is demonstrated through a simulation experiment.