Distributed Privacy-preserving Consensus Control Based on Steganography

Multi-agent network (MAN) requires frequent exchange and sharing of information among individuals in collaborative task execution, which poses significant risks to network security. To address the issue of privacy protection of node states in the network, a distributed consensus control strategy based on steganography is proposed. Firstly, an eavesdropper attack model is established, and a distributed privacy-preserving average consensus control algorithm is proposed. Theoretical analysis shows that the proposed algorithm not only effectively protects the privacy of node initial states but also can actively lead eavesdroppers to draw incorrect conclusions through steganographic carrier information. Secondly, a probabilistic indicator is introduced to quantify the privacy leakage index model for MAN, achieving an accurate description of the degree of network privacy leakage. Based on this model, an optimization problem is established from the eavesdropper＇s perspective by balancing the impact on network privacy leakage and the cost of attack, in order to find the optimal benefit attack strategy. Finally, through numerical simulation, the effectiveness and superiority of the proposed method are verified by comparing with existing algorithms. © 2025 Science Press. All rights reserved.