LAUSR

This brief studies energy-constrained confidentiality fusion estimation problem in presence of eavesdroppers, where all data that is transmitted from sensors to fusion center may be collected by eavesdroppers through wireless channels. To prevent eavesdropping, an injection method based on artificial noise is proposed, but the insertion of noises will consume more sensor energy, which can add the design difficulty under energy constraints. In this case, the stochastic sensor data triggers are adopted to reduce the communication rate between the local sensors and the fusion center. Subsequently, a sufficient condition is derived to guarantee the effective noise insertion strategy under certain triggers. Moreover, by selecting appropriate trigger threshold for each sensor with fixed encryption level, it is derived that how to make the eavesdroppers’ local/fusion estimation error covariance unbound while the user’s expected error remains the boundedness. Finally, the single generator infinite power system is employed to show the effectiveness of the proposed methods.