Physical Random Number Generator Entropy Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Physical random number generators are susceptible to environmental variations and entropy fluctuations, leading to unpredictable randomness and inefficiencies in entropy compression, which can compromise security in applications like secret key generation.
Innovation Solution
An apparatus comprising a physical random number generator, a test unit for randomness checking, a minimum entropy estimator, and an entropy control unit that dynamically adjusts the number of input bits for entropy compression based on estimated minimum entropy, ensuring proper entropy compression even with fluctuating entropy outputs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a fixed input length N is used for entropy compression based on minimum entropy estimated at manufacture, then the entropy compression process can be simplified, but processing time increases due to unnecessary processing when actual entropy exceeds the estimated minimum
Solution Approach 1:
The patent applies dynamics by transitioning from a fixed input length N to a variable input length that adapts to the actual entropy of the random number sequence. The entropy control unit dynamically adjusts the number of input bits based on the estimated entropy, allowing the system to optimize processing time while maintaining security. When entropy is high, fewer bits are processed; when entropy is low, more bits are processed to ensure sufficient randomness.
Solution Approach 2:
The patent changes the parameter of input length from a static value N to a dynamic value determined by actual entropy measurements. The entropy control unit modifies the input parameter (number of bits) based on entropy estimates, transforming the entropy compression process from a fixed-parameter operation to a variable-parameter operation that responds to environmental conditions.
2Reliability
If the entropy of the random number sequence drops below the estimated minimum entropy M, then the fixed N-bit input cannot generate sufficient entropy for secure random numbers, but increasing the input length beyond N is not possible with the fixed design
Solution Approach 1:
The patent implements feedback by using the entropy estimator to continuously monitor the actual entropy of the random number sequence and feeding this information back to the entropy control unit. The control unit then adjusts the input length to the entropy compression apparatus based on this feedback, ensuring that sufficient entropy is always provided to the compression process regardless of environmental variations.
Solution Approach 2:
The system dynamically adjusts the input length based on real-time entropy measurements, allowing it to adapt to changing environmental conditions. When entropy drops below the estimated minimum, the system increases the number of input bits to compensate, maintaining the reliability of random number generation throughout the product lifecycle.
3Reliability
If environmental variations such as temperature and voltage changes occur, then the randomness of generated random numbers degrades, but physical random number generators cannot be isolated from environmental factors
Solution Approach 1:
The patent uses feedback to continuously monitor the quality of randomness through entropy estimation and adjusts the system operation accordingly. The entropy estimator detects degradation in randomness quality caused by environmental factors, and the entropy control unit responds by adjusting the input length to maintain sufficient entropy for secure random number generation.
Solution Approach 2:
The system prepares for potential entropy degradation by having the entropy estimator continuously assess randomness quality and the entropy control unit ready to adjust input length. This beforehand cushioning ensures that when environmental variations cause entropy to drop, the system can immediately compensate by processing more input bits, preventing security degradation.
4Ease of manufacture
If entropy compression is performed with fixed N-bit input, then the circuit design is simplified, but processing efficiency decreases when actual entropy exceeds the estimated minimum
Solution Approach 1:
The patent introduces dynamics into the circuit by making the input length variable rather than fixed. The entropy control unit dynamically determines the number of input bits based on entropy estimates, allowing the circuit to maintain simplicity in its basic structure while achieving high efficiency through adaptive operation. The circuit can process fewer bits when entropy is high, improving productivity without complex redesign.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An apparatus for generating physical random numbers includes a physical random number generator configured to generate physical random numbers, a test unit configured to perform a test process to check randomness of the physical random numbers, a minimum entropy estimating unit configured to estimate a minimum entropy based on statistical information generated as a byproduct of the test process, an entropy compressing unit configured to perform an entropy compression process using the physical random numbers as an input, and an entropy control unit configured to control based on the minimum entropy a number of bits of the physical random numbers input into the entropy compression process performed by the entropy compressing unit.