Secure Multiparty Cloud Computation via Homomorphic Encryption
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In multiparty cloud computation, maintaining data privacy among multiple clients and the cloud is challenging due to security concerns, as existing solutions either impose heavy communication overhead or fail to ensure data privacy and security from misbehavior by clients and the cloud.
Innovation Solution
A system utilizing homomorphic encryption and proxy re-encryption, where clients encrypt their data with unique public/private key pairs, a trusted third party generates re-encryption keys, and the cloud performs computations on re-encrypted data, ensuring that only clients can decrypt the results, thus maintaining data privacy and security.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If clients encrypt their data with unique public/private key pairs and use homomorphic encryption for cloud computation, then data privacy and security are preserved, but communication overhead increases due to key management and re-encryption operations
Solution Approach 1:
The patent introduces a trusted third party as an intermediary that generates key pairs and re-encryption keys. This mediator enables secure multiparty computation by facilitating key distribution and re-encryption operations without requiring direct communication between all clients, thus preserving security while reducing communication overhead compared to fully decentralized approaches
Solution Approach 2:
The patent transforms data from plaintext to encrypted form using homomorphic encryption, changing the parameter state of the data. This allows computations to be performed on encrypted data directly, eliminating the need to decrypt data before processing and thereby maintaining security while reducing communication rounds needed for secure computation protocols
2Reliability
If the cloud performs computations on encrypted data using homomorphic encryption, then data security is maintained, but computation capability and speed are reduced
Solution Approach 1:
The patent segments the computation process into distinct phases: data encryption by clients, re-encryption by the trusted third party, computation on re-encrypted data by the cloud, and result decryption by clients. This segmentation allows each party to perform optimized operations in their respective domain, maintaining security while improving overall computational efficiency
Solution Approach 2:
The trusted third party acts as a mediator that performs re-encryption of client data before it reaches the cloud. This intermediary step transforms the encryption scheme to enable efficient homomorphic operations, allowing the cloud to perform computations on re-encrypted data with reduced computational overhead compared to direct homomorphic encryption
3Reliability
If re-encryption keys are generated and distributed by a trusted third party, then data privacy among clients is preserved, but system complexity increases due to key management infrastructure
Solution Approach 1:
The trusted third party performs multiple functions: generating client key pairs, creating re-encryption keys, and facilitating the secure computation process. This multi-functional approach consolidates key management infrastructure into a single entity, reducing overall system complexity while maintaining client data privacy through centralized but trusted key management
Data Source
AI summary
One embodiment of the present invention provides a system for performing secure multiparty cloud computation. During operation, the system receives multiple encrypted datasets from multiple clients. An encrypted dataset associated with a client is encrypted from a corresponding plaintext dataset using a unique, client-specific encryption key. The system re-encrypts the multiple encrypted datasets to a target format, evaluates a function based on the re-encrypted multiple datasets to produce an evaluation outcome, and sends the evaluation outcome to the multiple clients, which are configured to cooperatively decrypt the evaluation outcome to obtain a plaintext evaluation outcome.