Date of Award

5-2021

Document Type

Thesis

Degree Name

Master of Science (MS)

College/School

College of Science and Mathematics

Department/Program

Computer Science

Thesis Sponsor/Dissertation Chair/Project Chair

Bharath K. Samanthula

Committee Member

Boxiang Dong

Committee Member

Jiayin Wang

Abstract

The popularity of cloud computing has increased significantly in the last few years due to scalability, cost efficiency, resiliency, and quality of service. Organizations are more interested in outsourcing the database and DBMS functionalities to the cloud owing to the tremendous growth of big data and on-demand access requirements. As the data is outsourced to untrusted parties, security has become a key consideration to achieve the confidentiality and integrity of data. Therefore, data owners must transform and encrypt the data before outsourcing. In this paper, we focus on a Secure and Verifiable Computation for k-Nearest Neighbor (SVC-kNN) problem. The existing verifiable computation approaches for the kNN problem delegate the verification task solely to a single semi-trusted party. We show that these approaches are unreliable in terms of security, as the verification server could be either dishonest or compromised. To address these issues, we propose a novel solution to the SVC-kNN problem that utilizes the random-splitting approach in conjunction with the homomorphic properties under a two-cloud model. Specifically, the clouds generate and send verification proofs to end-users, allowing them to verify the computation results efficiently. Our solution is highly efficient from the data owner and query issuers’ perspective as it significantly reduces the encryption cost and pre-processing time. Furthermore, we show the correctness of our solution using Proof by Induction methodology to prove the Euclidean Distance Verification. Finally, with a thorough analysis and the empirical results on a real data set, we demonstrate the efficiency and effectiveness of our protocol.

File Format

PDF

Available for download on Friday, December 31, 2021

Share

COinS