Privacy-preserved average consensus algorithms with edge-based additive perturbations

In this paper, we consider the privacy preservation problem in continuous-time average consensus algorithms with strongly connected and balanced graphs, against either internal honest-but-curious agents or external eavesdroppers. A novel algorithm is proposed, which adds edge-based perturbation signals to process of computation. Our can be divided into two phases: a coordinated scrambling phase, for preservation, convergence phase. each agent required generate some add them edges leading out it. update their states following normal updating rule. It shown that an obtain target if only no other communicate agent. As eavesdroppers, it proved kind attackers never any agent’s privacy.