Random Number Generator Using Time-Delayed Sampling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing random number generators for cryptographic applications face challenges in producing high-quality random bit streams with low chip area and power requirements, while maintaining high production yield and low design costs, due to limitations in jitter-to-mean-period ratio and bandwidth constraints.
Innovation Solution
The method involves delaying a first signal by integer multiples of a time delay to form time-delayed signals, which are sampled and held by a sample and hold element triggered by a second signal, increasing time resolution and jitter detection without increasing the second signal's frequency, and post-processing the random bits to improve statistical quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the frequency of the second signal is increased to improve time resolution and jitter detection, then the quality of random numbers is improved, but the power consumption and chip area increase
Solution Approach 1:
The patent introduces a time dimension by delaying the first signal by integer multiples of a time delay, creating multiple time-delayed signals. This allows the system to achieve fine time resolution without increasing the frequency of the second signal, thereby avoiding increased power consumption while still improving jitter detection capability.
Solution Approach 2:
The patent segments the first signal into multiple time-delayed versions, where each delayed signal is sampled by the sample and hold element. This segmentation in the time domain allows the system to detect jitter at multiple time points, effectively improving measurement precision without requiring a higher frequency second signal.
2Reliability
If the frequency of the first signal is increased to improve randomness quality, then the entropy of the random stream is improved, but the bandwidth requirements and chip area increase
Solution Approach 1:
Instead of increasing the frequency of the first signal, the patent delays it by integer multiples of a time delay, exploiting the time dimension to create multiple sampled versions. This approach maintains the original signal frequency and bandwidth requirements while improving randomness quality through multi-point sampling.
Solution Approach 2:
The first signal is pre-delayed by integer multiples of a time delay before being sampled by the sample and hold element. This preliminary time-delay action creates multiple time-shifted versions of the signal, allowing the system to capture more entropy information without increasing the signal frequency.
3Reliability
If multiple oscillators are added to improve random stream quality, then the statistical properties are improved, but the device complexity and chip area increase
Solution Approach 1:
The sample and hold element serves multiple functions: it samples the first signal at multiple time points, holds each sampled value, and outputs them as random bits. This multi-functional approach allows the system to achieve improved statistical quality using a single oscillator pair rather than multiple oscillators, reducing device complexity.
Solution Approach 2:
The patent combines the functions of multiple oscillators into a single oscillator pair by using time-delayed sampling of the first signal. Instead of requiring multiple independent oscillators to generate diverse random streams, the system merges their functionality into one oscillator pair with temporal diversity, reducing chip area and complexity.
4Reliability
If the jitter of the second signal is increased to improve randomness, then the entropy of the random stream is improved, but the synchronization stability deteriorates
Solution Approach 1:
The patent shifts from increasing jitter in the time domain to exploiting time delays in the temporal dimension. By delaying the first signal and sampling at multiple time points, the system captures entropy without increasing the jitter of the second signal, thereby maintaining synchronization stability while improving randomness quality.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enhances the quality of the random number generator by ensuring uniform randomness and entropy detection, preventing synchronization, and maintaining low chip area and power consumption, suitable for integrated environments like chip-cards.
Implementation Method 1
a first signal source providing at its output a first signal having a first frequency
Implementation Method 2
a second signal source providing at its output a second signal having a second frequency, where the second frequency is lower than the first frequency
Implementation Method 3
delaying the first signal by integer multiples of a time delay to form at least one time-delayed signal
Implementation Method 4
sampling and holding each of the first signal and the at least one time-delayed signal, where the sampling and holding is triggered by the second signal
Implementation Method 5
Because of jitter, the period of the second signal S2 varies from cycle to cycle, so that the precise moment of the rising edge also varies with each cycle
Data Source
AI summary
Random number generator including a first signal source outputting a first signal having a first frequency, a second signal source outputting a second signal having a second frequency which is lower than the first frequency, and time delay elements, wherein the output signals are delayed by a time interval with respect to the input signal. The output of one time delay element is connected in series to the input of another time delay element. The input of the first time delay element is connected to the output of the first signal source. The output of the first signal source and the output of each of the time delay elements are connected to the data input of a corresponding sample and hold element. Clock signal inputs of the sample and hold elements are each connected to the output of the second signal source. Outputs of the sample and hold elements provide random values.