Private Aggregation with Custom Collusion Tolerance

摘要: While multiparty computations are becoming more and efficient, their performance has not yet reached the required level for wide adoption. Nevertheless, many applications need this functionality, while others it simpler computations; operations such as multiplication or addition might be sufficient. In work we extend well-known computation protocol (MPC) summation of Kurswave et al. More precisely, introduce two extensions one which bases its security on Decisional Diffie-Hellman hypothesis does use pairings, that significantly reduces pairings original. Both protocols proven secure in semi-honest model. Like original, entirely broadcast-based self-bootstrapping, but provide a significant boost, allowing them to adopted by devices with low processing power can also extended naturally achieve \(t\)-privacy malicious model, remaining practical. Finally, further improve if users decide decrease collusion tolerance.