## CryptoDB

### Satrajit Ghosh

#### Publications

Year
Venue
Title
2019
EUROCRYPT
Private set intersection (PSI) is an important area of research and has been the focus of many works over the past decades. It describes the problem of finding an intersection between the input sets of at least two parties without revealing anything about the input sets apart from their intersection.In this paper, we present a new approach to compute the intersection between sets based on a primitive called Oblivious Linear Function Evaluation (OLE). On an abstract level, we use this primitive to efficiently add two polynomials in a randomized way while preserving the roots of the added polynomials. Setting the roots of the input polynomials to be the elements of the input sets, this directly yields an intersection protocol with optimal asymptotic communication complexity $O(m\kappa )$. We highlight that the protocol is information-theoretically secure against a malicious adversary assuming OLE.We also present a natural generalization of the 2-party protocol for the fully malicious multi-party case. Our protocol does away with expensive (homomorphic) threshold encryption and zero-knowledge proofs. Instead, we use simple combinatorial techniques to ensure the security. As a result we get a UC-secure protocol with asymptotically optimal communication complexity $O((n^2+nm)\kappa )$, where n is the number of parties, m is the set size and $\kappa$ is the security parameter. Apart from yielding an asymptotic improvement over previous works, our protocols are also conceptually simple and require only simple field arithmetic. Along the way we develop techniques that might be of independent interest.
2019
CRYPTO
Threshold private set intersection enables Alice and Bob who hold sets $S_{\mathsf {A}}$ and $S_{\mathsf {B}}$ of size n to compute the intersection $S_{\mathsf {A}} \cap S_{\mathsf {B}}$ if the sets do not differ by more than some threshold parameter $t$ . In this work, we investigate the communication complexity of this problem and we establish the first upper and lower bounds. We show that any protocol has to have a communication complexity of $\varOmega (t)$ . We show that an almost matching upper bound of $\tilde{\mathcal {O}}(t)$ can be obtained via fully homomorphic encryption. We present a computationally more efficient protocol based on weaker assumptions, namely additively homomorphic encryption, with a communication complexity of $\tilde{\mathcal {O}}(t ^2)$ . For applications like biometric authentication, where a given fingerprint has to have a large intersection with a fingerprint from a database, our protocols may result in significant communication savings.Prior to this work, all previous protocols had a communication complexity of $\varOmega (n)$ . Our protocols are the first ones with communication complexities that mainly depend on the threshold parameter $t$ and only logarithmically on the set size n.
2017
ASIACRYPT
2015
EPRINT

#### Coauthors

Aniket Kate (1)
Jesper Buus Nielsen (1)
Tobias Nilges (2)
Mark Simkin (1)