SecurePL: a compiler and toolbox for practical and easy secure multiparty computation

Secure multiparty computation is basically about techniques that allow multiple parties to jointly carry out computations that are based on data from each of the players while the data held by each player remains private to that player. Since the beginning of the notion of secure multiparty computation, many algorithms and methods were introduced on how to achieve this goal. This thesis first introduces different methods to do secure multiparty computation and later focusing on Secret sharing based multiparty computation it explains how efficient and secure multiparty operations can be done. Also while introducing secret sharing based secure multiparty computation we introduce a novel technique which allows to do secure multiparty computation using the Asmuth Bloom secret sharing scheme, which is not possible in the original scheme. The aim of this thesis is the design and implementation of a programming language and libraries for secure multiparty computation, SecurePL. We show that our tool’s ease of use and security allows even a person who has absolutely no knowledge about security or cryptography to write applications that can do secure multiparty computation.