TY - GEN
T1 - Scalable multi-party private set-intersection
AU - Hazay, Carmit
AU - Venkitasubramaniam, Muthuramakrishnan
N1 - Publisher Copyright: © International Association for Cryptologic Research 2017.
PY - 2017
Y1 - 2017
N2 - In this work we study the problem of private set-intersection in the multi-party setting and design two protocols with the following improvements compared to prior work. First, our protocols are designed in the so-called star network topology, where a designated party communicates with everyone else, and take a new approach of leveraging the 2PC protocol of [FNP04]. This approach minimizes the usage of a broadcast channel, where our semi-honest protocol does not make any use of such a channel and all communication is via point-to-point channels. In addition, the communication complexity of our protocols scales with the number of parties. More concretely, (1) our first semi-honest secure protocol implies communication complexity that is linear in the input sizes, namely O((∑n i=1 mi)·κ) bits of communication where κ is the security parameter and mi is the size of Pi‘s input set, whereas overall computational overhead is quadratic in the input sizes only for a designated party, and linear for the rest. We further reduce this overhead by employing two types of hashing schemes. (2) Our second protocol is proven secure in the malicious setting. This protocol induces communication complexity O((n2+nmMAX+nmMIN log mMAX)κ) bits of communication where mMIN (resp. mMAX) is the minimum (resp. maximum) over all input sets sizes and n is the number of parties.
AB - In this work we study the problem of private set-intersection in the multi-party setting and design two protocols with the following improvements compared to prior work. First, our protocols are designed in the so-called star network topology, where a designated party communicates with everyone else, and take a new approach of leveraging the 2PC protocol of [FNP04]. This approach minimizes the usage of a broadcast channel, where our semi-honest protocol does not make any use of such a channel and all communication is via point-to-point channels. In addition, the communication complexity of our protocols scales with the number of parties. More concretely, (1) our first semi-honest secure protocol implies communication complexity that is linear in the input sizes, namely O((∑n i=1 mi)·κ) bits of communication where κ is the security parameter and mi is the size of Pi‘s input set, whereas overall computational overhead is quadratic in the input sizes only for a designated party, and linear for the rest. We further reduce this overhead by employing two types of hashing schemes. (2) Our second protocol is proven secure in the malicious setting. This protocol induces communication complexity O((n2+nmMAX+nmMIN log mMAX)κ) bits of communication where mMIN (resp. mMAX) is the minimum (resp. maximum) over all input sets sizes and n is the number of parties.
KW - Private set-intersection
KW - Scalable multi-party computation
UR - http://www.scopus.com/inward/record.url?scp=85014496165&partnerID=8YFLogxK
U2 - 10.1007/978-3-662-54365-8_8
DO - 10.1007/978-3-662-54365-8_8
M3 - منشور من مؤتمر
SN - 9783662543641
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 175
EP - 203
BT - Public-Key Cryptography – PKC 2017 - 20th IACR International Conference on Practice and Theory in Public-Key Cryptography, Proceedings
A2 - Fehr, Serge
PB - Springer Verlag
T2 - 20th IACR International Conference on Practice and Theory of Public-Key Cryptography, PKC 2017
Y2 - 28 March 2017 through 31 March 2017
ER -