Abstract
This article investigates the minimum data rate for achieving consensus of discrete-time linear multiagent systems under weighted-average protocols. The agents transmit information to each other over a network under data rate constraints. The minimum data rate required to enable the multiagent system to achieve consensus is obtained. It is shown that the minimum data rate is an intrinsic property of the multiagent system, which is independent of encoding-decoding schemes, controller gains, and communication topologies (as long as they are connected) chosen. Accordingly, an encoding-decoding scheme is proposed which achieves the minimum data rate. Then, it is proved that the multiagent system can achieve consensus based on the proposed scheme. Further, the minimum data rate for achieving asymptotic consensus with a given asymptotic convergence rate is derived. Finally, the effectiveness of the proposed scheme is verified by simulation examples.