MAster assignment
Homomorphic Encryption for Secure Computation in the Cloud
TYPE : MASTER CS
Period: Start date: as soon as possible
Student: Unassigned
If you are interested please contact:
Abstract:
This master's thesis aims to explore homomorphic encryption as a solution for secure computation in the cloud. The research will focus on studying different homomorphic encryption schemes, such as partially homomorphic or fully homomorphic encryption, and their applications in secure outsourcing of computations. The objective is to propose efficient and practical homomorphic encryption schemes that enable secure computation on encrypted data while preserving privacy and data confidentiality. The study will contribute to the advancement of secure cloud computing and privacy-preserving data processing.
Objective:
The primary objective of this research is to investigate homomorphic encryption schemes for secure computation in the cloud and propose efficient and practical solutions. The specific objectives include:
- Understanding the principles and concepts of homomorphic encryption, including partially homomorphic and fully homomorphic encryption schemes.
- Analyzing the limitations and challenges of homomorphic encryption in terms of computational overhead, communication complexity, and security assumptions.
- Exploring applications of homomorphic encryption in secure outsourcing of computations, such as private data processing and secure machine learning.
- Designing and implementing efficient and practical homomorphic encryption schemes that address the limitations and challenges.
- Evaluating the performance, security, and scalability of the proposed homomorphic encryption schemes through experiments and analysis.
- Assessing the feasibility and practicality of the proposed solutions for real-world cloud computing scenarios.
Methodology:
- Literature Review: Conduct an extensive review of literature and research papers on homomorphic encryption, secure computation in the cloud, privacy-preserving data processing, and applications of homomorphic encryption in various domains. Identify key research gaps and areas for exploration.
- Homomorphic Encryption Schemes: Study different types of homomorphic encryption schemes, including partially homomorphic and fully homomorphic encryption. Understand their mathematical foundations, security properties, and limitations. Analyze their efficiency and applicability in different computation scenarios.
- Application Scenarios: Identify specific application scenarios in cloud computing where secure computation and privacy-preserving data processing are required. Examples include secure machine learning, private data analysis, and encrypted search. Investigate the requirements and challenges of these scenarios in terms of security, privacy, and performance.
- Scheme Design and Implementation: Design and implement efficient and practical homomorphic encryption schemes that address the limitations and challenges identified in the previous steps. Consider factors such as computational overhead, communication complexity, and security guarantees. Develop algorithms and protocols for secure computation on encrypted data.
- Performance Evaluation: Evaluate the performance of the proposed homomorphic encryption schemes through experiments and analysis. Measure factors such as encryption and decryption time, computational overhead for homomorphic operations, and communication overhead. Compare the performance with existing encryption schemes and assess the practicality of the proposed solutions.
- Security Analysis: Conduct a comprehensive security analysis of the proposed homomorphic encryption schemes. Assess their resistance against known attacks, such as chosen ciphertext attacks or side-channel attacks. Investigate the impact of parameter choices on security and performance trade-offs.
- Feasibility and Practicality Assessment: Assess the feasibility and practicality of the proposed solutions for real-world cloud computing scenarios. Consider factors such as scalability, interoperability, and integration with existing cloud infrastructure. Explore potential deployment challenges and propose strategies for adoption.
Expected Outcome:
The expected outcome of this research is a comprehensive understanding of homomorphic encryption and the proposal of efficient and practical homomorphic encryption schemes for secure computation in the cloud. The thesis will contribute to the advancement of secure cloud computing and privacy-preserving data processing. The findings will aid in the development of cryptographic solutions that enable secure outsourcing of computations while preserving privacy and data confidentiality.
References:
- Gentry, C. (2009). A Fully Homomorphic Encryption Scheme. PhD thesis, Stanford University.
- López-Alt, A., Tromer, E., & Vaikuntanathan, V. (2012). On-the-Fly Multiparty Computation on the Cloud via Multikey Fully Homomorphic Encryption. In Proceedings of the 44th Annual ACM Symposium on Theory of Computing (STOC '12) (pp. 1219-1234). DOI: 10.1145/2213977.2214053
- Brakerski, Z., & Vaikuntanathan, V. (2014). Efficient Fully Homomorphic Encryption from (Standard) LWE. SIAM Journal on Computing, 43(2), 831-871. DOI: 10.1137/120874045
- Ducas, L., Micciancio, D., & Walter, M. (2014). Efficient Lattice (H)IBE in the Standard Model. In Proceedings of the 34th Annual Cryptology Conference on Advances in Cryptology (CRYPTO '14) (pp. 617-636). DOI: 10.1007/978-3-662-44381-1_35