Secured Pseudo-Random Number Generator Using Parallel LFSRs
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional pseudo-random number generators are not suitable for cryptographic applications due to predictability issues, large circuit scale, and complex cryptographic hash function implementation, which affects speed and efficiency.
Innovation Solution
A method utilizing multiple linear feedback shift registers (LFSRs) operating in parallel to generate candidate pseudo-random numbers, with selection based on an independent LFSR, ensuring high-speed, compact, and unpredictable pseudo-random number sequences.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional pseudo-random number generators are used, then generation speed is maintained, but predictability increases and cryptographic security decreases
Solution Approach 1:
The system divides the pseudo-random number generation into multiple independent LFSRs (Linear Feedback Shift Registers), each generating candidate numbers. By segmenting the generation process across multiple independent components, the system achieves both speed (parallel operation) and security (difficulty of prediction without knowing all seed values).
Solution Approach 2:
The patent combines multiple LFSRs with different polynomial configurations to create a composite generation system. Each LFSR contributes unique characteristics, and their combination produces a pseudo-random sequence that inherits the speed of individual LFSRs while gaining enhanced security through the complexity of the composite structure.
2Reliability
If cryptographic hash functions are implemented to improve security, then cryptographic security increases, but circuit scale and computational complexity increase
Solution Approach 1:
Instead of using expensive and complex cryptographic hash functions, the patent employs simpler, more efficient LFSRs that can be implemented with smaller circuitry. The LFSRs provide sufficient security for the application while being computationally lighter and requiring less circuit scale, effectively replacing heavy cryptographic primitives with lighter alternatives.
Solution Approach 2:
The system changes the parameters of the generation process by using multiple LFSRs with different polynomial configurations and seed values rather than relying on cryptographic hash functions. This parameter change maintains security requirements while reducing circuit complexity and improving generation speed.
3Reliability
If multiple LFSRs are used in parallel to improve security, then cryptographic security and unpredictability increase, but device complexity increases
Solution Approach 1:
Each LFSR in the parallel configuration serves multiple functions: generating candidate pseudo-random numbers, contributing to the overall unpredictability, and maintaining its own independent sequence. This multi-functionality justifies the inclusion of multiple components while maximizing their utility.
Solution Approach 2:
The patent merges the outputs of multiple LFSRs into a single pseudo-random number sequence through a selection mechanism. By combining the strengths of multiple independent generators, the system achieves high unpredictability while managing device complexity through efficient integration.
Data Source
AI summary
The present invention provides a method of generating a sequence of pseudo-random numbers which are difficult to predict. The method includes: (i) generating a plurality of candidate pseudo-random numbers by a respectively corresponding plurality of (differently structured) linear feedback shift registers; (ii) generating a “selector number” from one or more additional linear feedback shift registers; and (iii) selecting a candidate number from the plurality of candidate numbers, based on the “selection number” to produce a selected pseudo-random number for output.