Garbled Circuit Cloud Computing for Mobile Privacy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current cloud computing systems for mobile devices face challenges in providing secure and verifiable computations due to large ciphertext sizes in fully homomorphic encryption schemes and the impracticality of reusable garbled circuits, which compromise data privacy and efficiency.
Innovation Solution
A system utilizing multiple public cloud servers to create and evaluate garbled circuits, with a client generating pseudorandom bit sequences and seed values to construct and verify computations, ensuring privacy and efficiency through secure multiparty computation protocols and cryptographically secure pseudo-random number generation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If fully homomorphic encryption schemes are used to preserve data privacy, then data privacy is improved, but ciphertext size becomes extremely large
Solution Approach 1:
The system segments the encryption task by using garbled circuits instead of fully homomorphic encryption. The computation is divided into gates that operate on encrypted inputs, with each gate producing encrypted outputs. This segmentation avoids the exponential ciphertext expansion of FHE while maintaining security through the garbled circuit model, where only polynomial-size garbled tables are needed.
Solution Approach 2:
The system creates garbled copies of the computation circuit that can be evaluated on encrypted data. Instead of working directly with large FHE ciphertexts, the system generates garbled versions of the circuit gates and tables that represent the computation in an encrypted form, enabling efficient evaluation without the overhead of FHE ciphertext manipulation.
2Quantity of substance
If garbled circuits are used to reduce ciphertext size, then communication efficiency is improved, but the circuit can only be used once and must be recreated for each evaluation
Solution Approach 1:
The system performs preliminary actions by pre-generating garbled circuits and storing them for reuse. Instead of creating new garbled circuits for each computation, the system prepares garbled circuit templates in advance that can be instantiated multiple times with different encrypted inputs, significantly improving computational efficiency while maintaining security.
Solution Approach 2:
The system introduces dynamics by enabling the garbled circuits to be reused across multiple evaluations. The garbled circuit structure is designed to be dynamic and adaptable, allowing the same garbled circuit to process different encrypted inputs through multiple evaluations without degradation of security or performance.
3Reliability
If multiple servers are used to create garbled circuits, then security against collusion is improved, but system complexity increases
Solution Approach 1:
The system segments the garbled circuit creation process across multiple servers, with each server responsible for generating portions of the garbled circuit. This segmentation distributes the computational burden and enables security through diversity, where no single server has complete control over the garbled circuit, preventing collusion attacks while managing complexity through modular architecture.
Solution Approach 2:
The system introduces intermediary components that coordinate between multiple servers participating in garbled circuit creation. These intermediaries manage the complex interactions and communications between servers, orchestrating the distributed generation of garbled circuits while maintaining security protocols and reducing the apparent complexity for individual participants.
Data Source
AI summary
Disclosed are systems and methods for delegating computations of resource-constrained mobile clients, in which multiple servers interact to construct an encrypted program representing a garbled circuit. Implementing the garbled circuit, garbled outputs are returned. Such implementations ensure privacy of each mobile client's data, even if an executing server has been colluded. The garbled circuit provides secure cloud computing for mobile systems by incorporating cryptographically secure pseudo random number generation that enables a mobile client to efficiently retrieve a result of a computation, as well as verify that an evaluator actually performed the computation. Cloud computation and communication complexity are analyzed to demonstrate the feasibility of the proposed system for mobile systems.