Hash Vectorization Models for Encrypted Search Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing encrypted search techniques face challenges in balancing security, efficiency, and query expressiveness, often requiring trade-offs that compromise either security or performance when handling large datasets and complex queries.
Innovation Solution
The use of Hash Vectorization (HV) models, which combine a hash filter and a chain vector to facilitate secure and efficient searching of encrypted data through a compression process like Lempel-Ziv parsing, allowing for accurate and expressive queries without decrypting the data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional encrypted search techniques are used, then data security is maintained, but search efficiency and query expressiveness deteriorate
Solution Approach 1:
The patent segments the search process into two distinct phases: an offline indexing phase where encrypted data is processed to create hash vectorization models (HVMs) with filters and chains, and an online query phase where searches are performed efficiently using pre-computed hash structures. This segmentation allows security-critical operations to occur offline while enabling fast online searches without decrypting data.
Solution Approach 2:
The patent performs preliminary actions by pre-computing hash vectorization models during an offline indexing phase. The HVMs include pre-computed filters and chains that capture encrypted data patterns. When online queries arrive, the system leverages these pre-computed structures to rapidly evaluate encrypted search conditions without performing heavy cryptographic operations in real-time.
2Adaptability or versatility
If complex queries are performed on encrypted data, then query expressiveness is improved, but computational overhead and time consumption increase
Solution Approach 1:
The patent introduces a new dimension by transforming encrypted data into a hash-based vectorization space. Instead of operating directly on encrypted byte sequences, the system projects data into hash chains and filters that preserve pattern relationships. This dimensional transformation enables complex pattern matching (e.g., substring searches, wildcard queries) to be evaluated efficiently through hash comparisons rather than exhaustive encrypted data scanning.
Solution Approach 2:
The patent introduces hash vectorization models as an intermediary between the encrypted data and the search query. The HVMs act as a mediator structure that captures encrypted data patterns in a form amenable to efficient query evaluation. The filters and chains in the HVM serve as intermediate representations that enable expressive queries to be resolved through hash-based matching rather than direct encrypted data processing.
3Speed
If hash filters are used to eliminate negative candidates, then search speed is improved, but false positive rates may increase
Solution Approach 1:
The patent implements a nested structure where filters are embedded within hash vectorization models, which in turn contain chains. The filters serve as an outer layer that quickly eliminates non-matching candidates using hash comparisons. Matching candidates then proceed to chain verification, which provides a second layer of validation. This nested arrangement enables rapid filtering while maintaining accuracy through hierarchical verification.
Solution Approach 2:
The patent incorporates feedback mechanisms where the chain verification process provides corrective feedback to filter results. When the filter identifies potential matches, the chain structure verifies these candidates by checking actual encrypted data patterns. This feedback loop eliminates false positives introduced by the filter's hash-based matching while preserving true matches, thereby maintaining both speed and accuracy.
Data Source
AI summary
An encrypted search uses hash vectorization (HV) models, which are secure, one-way hash indices that are produced by a compression process, such as a modeling phase of Lempel-Ziv (LZ) parsing. Each HV model includes a hash filter and a chain vector. The hash filter is a Boolean quotient filter similar to Bloom filters, and allows for the quick elimination of negative query assessments. The hash filter is followed by a chain vector, which provides spatial modeling of hashed elements throughout the compressed and encrypted data, thereby providing increased levels of accuracy, efficiency, and query expressiveness as compared with known techniques.