Resilient Randomized Quantized Consensus
Seyed Mehran Dibaji, Hideaki Ishii, Roberto Tempo · IEEE Transactions on Automatic Control · 2016
We study the problem of quantized consensus where some agents are subject to faults and attacks and might make updates arbitrarily. The network consists of integer-valued agents under directed communication links. The misbehaving agents are assumed to be omniscient and may try to prevent the normal agents to reach consensus. We show that consensus can be achieved among the normal agents via a simple update scheme under which each normal agent ignores some of its neighbors' states taking unsafe values. We employ the connectivity notion of graph robustness and provide necessary and sufficient conditions for resilient consensus in a network with a known bounded number of faulty agents. It will be shown that randomization is essential both in quantization and in the updating times when normal agents interact in an asynchronous manner.