Pipeline-Friendly Aggregate Signatures for Parallel Verification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing blockchain technologies face inefficiencies in transaction processing due to sequential submission of transactions, leading to lagged processing speeds, especially when many transactions are generated simultaneously, and the use of BLS-based threshold signatures consumes significant CPU time, limiting the application of blockchain systems.
Innovation Solution
A pipeline-friendly signature method is developed, utilizing an aggregate signature with a linear verification structure and threshold signatures, allowing parallel processing and reducing verification pressure by avoiding pipeline interruptions and optimizing network communication.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If sequential transaction submission is used, then verification accuracy is maintained, but transaction processing speed deteriorates
Solution Approach 1:
The transaction processing is divided into multiple independent stages (transaction generation, validation, packaging, verification, commitment) that can be processed in parallel pipelines. Each stage handles specific tasks independently, allowing multiple transactions to be processed simultaneously at different stages, thereby improving throughput while maintaining verification accuracy through staged validation.
Solution Approach 2:
Transactions are pre-validated and packaged before being submitted for verification. The system performs preliminary checks on transaction format, signature validity, and consensus requirements in advance, so that when transactions reach the verification stage, they are ready for rapid parallel processing, improving overall processing speed without sacrificing accuracy.
2Reliability
If BLS-based threshold signatures are used, then security and power distribution are improved, but CPU time consumption increases
Solution Approach 1:
Multiple individual signatures from different nodes are merged into a single aggregate signature using BLS threshold signature technology. This allows the system to maintain high security requirements (needing T out of N signatures) while reducing the number of verification operations needed, as one aggregate signature verification replaces multiple individual verifications, thereby lowering CPU time consumption.
Solution Approach 2:
The system changes the verification parameter from verifying N individual signatures to verifying one aggregate signature. This parameter transformation maintains the security level (since the aggregate signature still requires T valid signatures to generate) but dramatically reduces computational overhead, addressing the CPU time consumption issue.
3Productivity
If pipeline processing is implemented, then transaction processing efficiency is improved, but pipeline interruptions occur due to failures
Solution Approach 1:
The system implements checkpoint mechanisms and error buffering at each pipeline stage. When a failure occurs in one pipeline, the buffered data and checkpoint states allow the system to recover and restart the affected pipeline without stopping the entire processing system, thereby maintaining pipeline continuity and reliability while preserving processing efficiency.
Solution Approach 2:
Error handling intermediaries are introduced between pipeline stages to catch and manage failures. These intermediaries act as buffers that isolate pipeline failures, preventing them from propagating and interrupting the entire pipeline. This allows individual pipelines to be restarted independently, maintaining overall system reliability and continuity.
Data Source
AI summary
The present application provides a pipeline-friendly signature and verification method, a device and a storage medium. The method includes the following steps: generating, by the Witness node, a public-private key pair including a private key and a public key, through a basic signature algorithm, and sending the public key to other Witness nodes; selecting, by the Witness node, a random number, and sending V to a Leader node; receiving C sent by the Leader node; calculating a message digest through the basic signature algorithm and a message hash algorithm, according to the C and the public key, and then calculating ps, and sending the ps to the Leader node.


