Quantifying Differential Privacy of Gossip Protocols in General Networks

In this work, we generalize the study of quantifying the differential privacy guarantees of gossip protocols. Firstly, The lower bound of differential privacy of all gossip protocols is estimated for general networks in both synchronous and asynchronous settings. Secondly, in wireless networks, considering unreliable links for both node's communications and attacker's detection, a fail probability is adopted. The tradeoff between spreading time and differential privacy of standard and private gossips is investigated and it is suggested that noise can serve as a useful privacy protection tool. Finally, the delayed monitoring, a more realistic monitoring model, is considered, and the corresponding differential privacy is evaluated.