Secure Search over Multi-key Homomorphically Encrypted Data

Abstract

Homomorphic Encryption (HE) is a very attractive solution to ensure privacy when outsourcing confidential data to the cloud, as it enables computation on the data without de-cryption. However HE starts to lose effectiveness when scaled to multiple parties. In this paper, we propose the first multi-key HE search and computation framework. To achieve an efficient set-up for multi-party search and compute, we explore the different approaches to multi-key HE and secure search schemes to reduce rounds of communication. We propose a novel framework to search homomorphically encrypted data outsourced to a semi-honest server and shared with multiple parties dynamically using proxy re-encryption schemes. Our framework performs search with linear search complexity with just one round of communication between the two parties. The protocol provides multi-hop ca-pabilities that enable further computations on the search results.