Scalable Data Obfuscation via Domain Size Rank Conversion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional data obfuscation systems in distributed data environments face significant latency and computational overhead due to reliance on deterministic finite state machines, and memory constraints limit the representation of unique sequences, preventing scalable implementation with large amounts of data.
Innovation Solution
A computer-implemented method that determines a domain size and rank score of an input value to generate a unique value, which is then converted into a masked string value, allowing dynamic data masking without using deterministic finite state machines, thereby reducing latency and computational overhead.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If deterministic finite state machines are used for data masking, then data obfuscation can be implemented, but latency and computational overhead increase significantly
Solution Approach 1:
The patent extracts the core obfuscation function from the complex deterministic finite state machine framework. Instead of using full FSMs that process data character by character through multiple states, the invention extracts only the essential masking operation and implements it through a simplified domain-size-based character conversion approach, eliminating the latency-intensive state transition processing while maintaining obfuscation capability
Solution Approach 2:
The patent replaces the mechanical/FSM-based data masking system with a mathematical approach. Instead of using state machine transitions that require multiple processing steps, the invention uses domain size calculations and direct character value conversions to achieve masking, substituting the mechanical processing model with a more efficient mathematical transformation model
2Reliability
If deterministic finite state machines are used for data masking, then data obfuscation can be implemented, but computational overhead increases
Solution Approach 1:
The patent extracts the essential obfuscation function from the computationally intensive deterministic finite state machine framework. By removing the unnecessary state transition logic and retaining only the core character masking operation, the system achieves data obfuscation with significantly reduced computational overhead and energy consumption
Solution Approach 2:
The patent changes the fundamental parameters of the masking approach by using domain size-based character value transformations instead of FSM state transitions. This parameter change fundamentally alters the computational complexity, reducing the number of operations required per character while maintaining the obfuscation function
3Productivity
If character patterns are stored in cache to avoid generating deterministic finite state machines, then some masking operations are avoided, but memory constraints prevent representing all unique sequences
Solution Approach 1:
The patent changes the approach from storing complete character patterns to using domain size-based mathematical transformations. This parameter change allows the system to handle any unique sequence without memory constraints by using calculations rather than stored lookup tables, enabling unlimited sequence representation within memory limits
Solution Approach 2:
The patent introduces domain size as an intermediary parameter that mediates between the input character and the masked output. Instead of directly mapping characters through stored patterns or FSM states, the system uses domain size calculations as an intermediate step to generate masked values, eliminating the need for large character pattern caches
4Reliability
If conventional data obfuscation systems are used, then data privacy can be protected, but scalability with large amounts of data is limited
Solution Approach 1:
The patent replaces the mechanical FSM processing model with a mathematical domain-size-based transformation model. This substitution enables the system to scale efficiently with large data volumes by using O(1) mathematical operations per character rather than the more complex state transition processing, achieving both privacy protection and scalability
Data Source
AI summary
A computer-implemented method, according to one embodiment, includes: receiving a request to mask an input value, and in response to receiving the request, determining a domain size of the input value. A rank score of the input value is also determined. The domain size and the rank score are used to generate a unique value correlated with the input value. Moreover, the unique value is unranked by converting characters of the unique value to create a masked string value. The input value is then replaced with the masked string value.