Convergence and recovery analysis of the secure distributed control methodology for D-NCS

Distributed control algorithms (e.g., consensus algorithms) are vulnerable to the misbehaving agent compromised by cyber attacks in the Distributed Networked Control Systems (D-NCS). In this paper we continue our work on a proposed secure distributed control methodology that is capable of performing a secure consensus computation in D-NCS in the presence of misbehaving agents. The methodology is introduced first and proven to be effective through convergence analysis. Then, we extend our secure distributed control methodology to the leaderless consensus network by introducing and adding two recovery schemes into the current secure distributed control framework to guarantee accurate convergence in the presence of misbehaving agents. All phases of our method are distributed. That is, at each step of the detection, mitigation, identification, update, and recovery, each agent only uses local and one-hop neighbors´ information. The simulation results are presented to demonstrate the effectiveness of the proposed methods.