In this paper we introduce a new model for private computation, the Shared Randomness over Broadcast Channel model (SRoBC for short). Following the classical model for private computation [2,4], we consider a set of n computationally unbounded honest but curious players P1,…, Pn with private inputs x1, x2,…, xn. The players wish to compute a function f of their inputs in a private way. Unlike in the classical model, no private channel is available to the players but all the communication takes place using a broadcast channel. Moreover, the only available source of randomness is a shared random string. We show that even in this minimal setting private computation is possible: we present a protocol for computing the sum modulo 2 in a t-private way in the SRoBC model. The protocol uses n(t+1)/2 random bits. We show that this is the optimal randomness complexity in the case each random bit is shared between two players (low-contention protocols). We further show that, in the case t = 1, this protocol is optimal with respect to the randomness complexity regardless of the contention of the protocol.
CITATION STYLE
Galdi, C., & Persiano, P. (2002). Private computation with shared randomness over broadcast channel. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 2288, pp. 244–257). Springer Verlag. https://doi.org/10.1007/3-540-45861-1_19
Mendeley helps you to discover research relevant for your work.