Biometric Database Search Using Hash-Based Keyword Addressing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Biometric data stored in encrypted form in identification systems complicates database searches, impacting efficiency and security, as decryption is often required for comparison, which can be resource-intensive and vulnerable to security breaches.
Innovation Solution
A method using a family of hash functions and an addressing function to generate keywords from biometric data, deriving multiple database addresses for storing identifiers, allowing secure and efficient identification without decrypting the data, utilizing probabilistic encryption to protect against attacks and ensure confidentiality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If biometric data are stored in encrypted form in the database, then security and confidentiality are improved, but database search efficiency deteriorates
Solution Approach 1:
The patent applies preliminary action by pre-processing biometric data through hash functions during the enrollment phase to generate multiple keywords and corresponding database addresses. This allows the system to store encrypted biometric data in a structured manner where search operations can directly query pre-computed addresses without requiring decryption, thus maintaining both security and search efficiency.
2Measurement precision
If decryption is performed to enable comparison of biometric data, then identification accuracy is improved, but security is worsened due to vulnerability to breaches
Solution Approach 1:
The patent uses copying by creating multiple keyword representations from the original biometric data through different hash functions. These keywords serve as copies or representations that can be used for search operations without revealing the original biometric data. The system stores identifiers at multiple addresses derived from these keywords, enabling accurate identification through comparison while keeping the actual biometric data encrypted and secure.
3Productivity
If multiple hash functions are applied to generate keywords and addresses, then search efficiency is improved, but calculation complexity increases
Solution Approach 1:
The patent applies segmentation by dividing the biometric data processing into multiple independent hash functions, each generating a keyword and corresponding database address. This segmentation allows the search operation to query multiple addresses in parallel and combine results, improving overall search efficiency. The use of multiple independent hash functions distributes the calculation workload and enables more efficient data retrieval compared to single-address storage.
Data Source
AI summary
The invention relates to a database (10) suitable for combining biometric data (b) and an identifier (Id(b)). For this purpose, biometric data (bref) are collected (101). Next, a plurality of keywords is generated (102) by means of applying a family (H) of hash functions to the biometric data. Then, a plurality of addresses (@i) is obtained (103) by means of applying an addressing function (F) to the plurality of keywords and to a secret key. Finally, the identifier is stored (104) at said plurality of addresses.