Abstract
This paper combines Beaver triples from secret-sharing theory with communication obfuscation to protect agents’ initial values against passive adversaries in a multi-agent system. The proposed algorithm is proved to achieve average consensus and privacy preservation simultaneously. Compared with encryption-based methods, it reduces online computation and communication overhead, while requiring less restrictive privacy conditions than several noise-obfuscation approaches.
Ning LI
Tenured Professor
Tenured Professor at the School of Automation and Sensing, Shanghai Jiao Tong University. Research areas include distributed optimization and control, prognostic and health management, active perception and intelligent decision-making of unmanned systems, and artificial intelligence.