Secure Biometric Identification Circuit Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for secure biometric identification using Boolean circuits are costly due to the high number of non-XOR gates and circuit depth required to determine multiple minimum distances between biometric data and reference data, making large-scale secure identification inefficient.
Innovation Solution
A method that minimizes the number of non-XOR gates and circuit depth by processing indexed distance data using a marker system to determine the set of minimum indices, allowing each bit of indexed distance data to be read only once and reducing the size of the logic circuit to be evaluated securely.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If secure multi-party computation protocols based on Boolean circuits are used to determine multiple minimum distances, then identification security is improved, but computational cost increases due to high number of non-XOR gates and circuit depth
Solution Approach 1:
The patent segments the determination of multiple minimum distances into iterative single-minimum determinations. Instead of computing all minimum distances simultaneously using a complex Boolean circuit, the system determines one minimum distance at a time through repeated applications of secure comparison protocols, thereby reducing the circuit complexity and non-XOR gate count for each computational step
Solution Approach 2:
The patent applies preliminary filtering to identify and eliminate obviously non-minimal distance values before applying secure comparison protocols. By pre-processing the distance data to remove candidates that cannot be minimum values, the system reduces the number of elements requiring expensive secure computations, thus lowering overall computational cost while maintaining security
2Measurement precision
If all indexed distance data are processed to determine multiple minimum indices, then identification accuracy is improved, but processing time increases
Solution Approach 1:
The patent changes the processing approach by iteratively determining minimum indices rather than processing all data simultaneously. The system modifies the parameter of interest from computing multiple minimums in parallel to computing them sequentially through repeated single-minimum determinations, each with optimized comparison logic that reduces overall processing time while maintaining accuracy
3Use of energy by moving object
If the number of non-XOR gates in the Boolean circuit is reduced, then computational cost is decreased, but the ability to determine multiple minimum distances accurately is compromised
Solution Approach 1:
The patent segments the complex task of determining multiple minimum distances into multiple simpler tasks of determining single minimum distances iteratively. Each segmented task uses a simplified Boolean circuit with fewer non-XOR gates, reducing computational cost per iteration. The overall accuracy is maintained by repeating the process multiple times to determine all required minimum indices
Solution Approach 2:
The patent applies partial action by determining minimum indices iteratively rather than computing all minimum distances simultaneously with a full Boolean circuit. Each iteration performs a partial computation (single minimum determination) with reduced circuit complexity, and the process is repeated sufficient times to achieve the complete result, balancing computational cost and accuracy
Data Source
AI summary
The present invention relates to a method for identifying an entity implemented by an identification system based on indexed distance data (d1,...,dn) corresponding to reference entities and comprising: - a phase of determining a set (I) of minimum indices (index1, ..., indexk) among said binary indexed distance data of length q' comprising an execution step comprising, for each set of the jth bits of the indexed distance data included in a list of data to be processed, j being an integer ranging from q'-1 to 0, starting with the set of most significant bits of the data to be processed and ending with the set of least significant bits of the data to be processed, the search for minimum indices comprising, if a number of indices from a first group of indexed distance data indices (p) is greater than a remaining number of indexed data to be discarded (r), an addition of said indices from a second group to the set of minimum indices; - an identification phase of the entity to be identified among the reference entities corresponding to the stored reference biometric data associated with the minimum distance data indices (index1, ...indexk) determined, the operations on binary integers for the implementation of at least said execution step, being translated into the form of at least one boolean circuit used to implement at least said execution step securely between the control server and the management server using a secure multi-party computation protocol allowing secure evaluation of said boolean circuit.