Secure and efficient outsourcing of large-scale nonlinear programming

Abstract

Cloud computing is attracting increasing attention since it enables clients with limited computing resources to perform and complete large-scale computations. However, it also comes up with some security and privacy concerns and challenges, such as the input and output privacy of the client, and cheating behaviors of the cloud. Motivated by these issues and focused on engineering optimization tasks, we study secure outsourcing of large-scale nonlinear programming, which has not been investigated before. Specifically, a secure and efficient transformation scheme is employed to protect both input and output privacy of the client, and corresponding detailed proofs and analysis are also provided. We apply the reduced gradient method to solve the encrypted nonlinear programming problem in the cloud side. We conduct experiments to measure performance of the designed outsourcing protocol, and the results show the practicability of the proposed mechanism.