Accurate Privacy Preserving Average Consensus

Average consensus has significant applications in dynamic load balancing and cooperative control of vehicle formations, where all the agents receive information from neighboring agents via communication network and update their states to achieve an agreement. However, this approach would result in an undesirable disclosure on the initial states of agents to their neighbors. In this brief, we propose an accurate privacy preserving average consensus (APPAC) algorithm, where all the agents independently generate and transmit random numbers based on Paillier cryptosystem to conceal their initial states. Under the proposed APPAC algorithm, the accurate average consensus is indeed achieved. Besides, the necessary and sufficient conditions that initial states can be inferred are also discussed. Extensive simulations are conducted to demonstrate the effectiveness of the proposed algorithm.