True Random Number Generator Using PUF and Dynamic Entropy

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional random number generators rely on dynamic entropy sources that may produce output data with poor randomness, and existing true random number generators using physically unclonable functions (PUF) are limited by the predictability of their storage states after enrollment, making them pseudo-random number generators.

Innovation Solution

A true random number generator design utilizing a PUF cell array as a static entropy source, combined with a counting value generator, address generator, and processing circuit, which generates and processes first and second random numbers to produce an output random number with enhanced randomness through logical operations and dynamic reseeding mechanisms.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If a PUF cell array is used as a static entropy source, then uniqueness and security are improved, but the storage state becomes predictable after enrollment, reducing randomness

Engineering Contradiction:
ImprovesecurityVSAvoidpredictability of storage state
Core Design Contradiction:
ReliabilityVSManufacturing precision

Solution Approach 1:

The patent introduces dynamic entropy sources (ring oscillators, metastable circuits) that continuously generate changing random values, transforming the static PUF storage into a dynamic system where the output constantly evolves, preventing predictability while maintaining the unique security properties of the PUF

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent combines static entropy source (PUF cell array) with dynamic entropy sources (ring oscillators, metastable circuits) to create a hybrid random number generator that leverages both the uniqueness/security of PUF and the continuous randomness of dynamic sources, resolving the predictability issue

Inventive Principle:
Principle #5Merging (Combining)

2Manufacturing precision

If dynamic entropy sources are used to generate random numbers, then randomness is improved, but the quality of output data may be poor

Engineering Contradiction:
Improverandomness qualityVSAvoidoutput data quality
Core Design Contradiction:
Manufacturing precisionVSReliability

Solution Approach 1:

The patent implements health test circuits that continuously monitor the quality of random numbers generated by dynamic entropy sources, using feedback mechanisms to detect and correct poor quality output, ensuring high reliability while maintaining randomness

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The patent introduces processing circuits and health test circuits as intermediary components between the dynamic entropy source and the output, which filter, process, and validate the random data to ensure high quality output while preserving the inherent randomness

Inventive Principle:
Principle #24Intermediary (Mediator)

3Manufacturing precision

If processing circuits are added to improve randomness, then output quality is improved, but device complexity increases

Engineering Contradiction:
Improverandomness of output dataVSAvoidcircuit structure
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent designs processing circuits that perform multiple functions simultaneously - generating random numbers, testing health, and ensuring quality - allowing a single circuit component to serve multiple purposes, thereby improving randomness without proportionally increasing complexity

Inventive Principle:
Principle #6Universality (Multi-functionality)

Data Source

PatentEP3709157B1Random number generator
Publication Date: 2022.05.04 EMEMORY TECH INC
  • EP3709157B1 patent drawingFigure 1A~1B
  • EP3709157B1 patent drawingFigure 2~4B
  • EP3709157B1 patent drawingFigure 5~6

AI summary

A random number generator includes a counting value generator, an address generator, a static entropy source and a processing circuit. The counting value generator generates a first random number. The address generator generates an address signal. The static entropy source is connected with the address generator to receive the address signal and generates a second random number. The processing circuit is connected with the static entropy source and the counting value generator to receive the first random number and the second random number. After the first random number and the second random number are processed by the processing circuit, the processing circuit generates an output random number.