Distributed CRS Sampling With Malicious-Secure Setup
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing distributed samplers either require the full power of the random oracle model or achieve only limited security guarantees, and achieving full malicious security in the standard model is impossible, posing challenges in ensuring security and scalability when multiple parties generate a common reference string.
Innovation Solution
New definitions and constructions for distributed samplers that provide meaningful security guarantees in the standard model, utilizing single-message zero knowledge proofs and new notions of security, such as hardness-preserving and indistinguishability-preserving samplers, to ensure security even when parties do not fully trust each other.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple third parties engage in an MPC protocol to generate the reference string, then trust distribution is improved, but communication complexity and logistical burden increase
Solution Approach 1:
The reference string generation is segmented into independent local operations at each third party, with results combined through a deterministic extraction procedure. Each third party runs their own setup algorithm locally and posts messages to a public domain, eliminating the need for complex multi-party communication while maintaining distributed trust.
Solution Approach 2:
Instead of requiring multiple parties to coordinate through MPC, the system uses copies of individual setup messages from each third party as inputs to a deterministic extraction function. This allows any combination of third parties to generate a reference string without requiring them to interact with each other.
2Ease of operation
If a single trusted third party generates the reference string, then ease of operation is improved, but trust concentration and security risks worsen
Solution Approach 1:
The system merges the simplicity of single-party setup with the security benefits of distributed trust by combining individual third party messages through a deterministic extraction function. The reference string is derived from the concatenation of messages from multiple third parties, achieving both ease of operation and distributed trust.
3Adaptability or versatility
If existing distributed sampler constructions are used, then scalability is improved, but security guarantees are limited to semi-honest model
Solution Approach 1:
The system changes the security parameter model from semi-honest to full malicious security by using indistinguishability obfuscation and non-interactive zero-knowledge proofs. This allows the distributed sampler to achieve strong security guarantees against malicious adversaries while maintaining scalability and the ability to incorporate new third parties.
4Reliability
If the random oracle model is used for distributed sampler construction, then security guarantees are improved, but model assumptions and complexity worsen
Solution Approach 1:
The system replaces the random oracle model with standard cryptographic primitives including indistinguishability obfuscation and non-interactive zero-knowledge proofs. This substitution eliminates the need for random oracle assumptions while achieving full malicious security guarantees in the standard model.
Data Source
AI summary
The disclosed technology pertains to a method and system for generating a common reference string for use by multiple parties in cryptographic operations. The method involves executing a generation algorithm that creates encryption and decryption codes, generates non-interactive zero-knowledge proofs, and outputs a message for each party. These messages are published to a public domain and accessed by all parties. A derivation algorithm is then executed at each party to derive a common reference string from the set of messages by combining them. The system comprises a computerized processor configured to execute these algorithms. The common reference string is stored in a computerized storage device.