Efficient User Inspection Algorithm Based on Dual Bloom Filters Oriented for Blockchain Data Management Systems

With the increasing application of blockchain technology in decentralized data management systems (DMS), the prevailing problems of poor query performance relative to its conventionally centralized counterparts are in urgent need of reasonable solutions. Bloom filters are efficient data query frameworks that adopt multiple hash functions to map the target database into one-dimensional arrays (one bit per array cell), leading to high-efficiency information extraction and fulfilling the urgent optimization needs of the blockchain's query performance. In light of the above background, dual Bloom filters were proposed herein to accelerate the inspection speed of user information in a blockchain DMS. Taking into account the different scenarios for user registration and user login, a comparative experiment was conducted between the conventional methods and the proposed algorithm, with the latter outperforming the former, as revealed by the results. The proposed algorithm has proven to be capable of swiftly completing not only the duplicate detection of usernames for newly-registered accounts but also the parallel query for the username and password during user login. This functionality has greatly reduced noise from irrelevant information and improved the query performance of blockchain-based hyperledgers.