Ring Random Number Generator Auto-Tuning Sampling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current random number generators face challenges in detecting pseudo-periodic sequences and ensuring sufficient randomness, especially when sampling rates are too high, leading to low entropy and inefficient operation, particularly in environments not considered during their design.
Innovation Solution
An auto-tuning circuit that adjusts the sampling rate based on real-time randomness testing using multiple bit-producing circuits and test circuits to detect non-random behavior, ensuring optimal sampling frequency and robust randomness generation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the sampling rate is increased to improve throughput, then productivity increases, but the randomness quality deteriorates because samples are taken too frequently for the jitter to cause sufficient uncertainty
Solution Approach 1:
The patent implements dynamic adjustment of the sampling rate based on real-time monitoring of jitter characteristics. The system continuously measures the phase jitter of the drifting oscillator and adapts the sampling frequency accordingly, allowing the sampling rate to vary within a range rather than remaining fixed. This dynamic approach enables the system to maintain optimal randomness quality while maximizing throughput by sampling as fast as the jitter conditions permit.
2Reliability
If the sampling rate is decreased to improve randomness quality, then reliability improves, but productivity decreases due to reduced throughput
Solution Approach 1:
The patent changes the sampling rate parameter dynamically based on measured jitter characteristics. Instead of using a conservative fixed sampling rate, the system adjusts this parameter in real-time according to the actual phase jitter conditions, allowing faster sampling when jitter is sufficient and slower sampling when jitter is low, thereby optimizing both randomness quality and throughput.
Solution Approach 2:
The patent implements a feedback mechanism where the randomness quality (phase jitter) is continuously monitored and used to adjust the sampling rate. The system measures the jitter characteristics and feeds this information back to the sampling rate controller, which then adapts the sampling frequency to maintain optimal randomness while maximizing throughput.
3Reliability
If conservative design parameters are used to ensure sufficient entropy under all conditions, then reliability improves, but productivity decreases due to reduced speed under normal conditions
Solution Approach 1:
The patent replaces static conservative design parameters with dynamic parameters that adapt to actual operating conditions. The sampling rate is continuously adjusted based on measured jitter characteristics, allowing the system to operate at higher speeds when conditions permit while maintaining entropy sufficiency. This eliminates the need for overly conservative fixed parameters that limit performance under normal conditions.
4Device complexity
If a fixed sampling rate is used to simplify circuit design, then device complexity is reduced, but adaptability decreases when used in environments not considered during design
Solution Approach 1:
The patent implements a self-adjusting system where the random number generator automatically monitors its own jitter characteristics and adapts its sampling rate without external intervention. The system includes onboard measurement circuitry that continuously assesses the phase jitter and automatically adjusts the sampling frequency, enabling the circuit to adapt to different environmental conditions while maintaining relatively simple overall architecture.
Data Source
AI summary
An apparatus includes: a plurality of bit producing circuits; a controller setting a sample frequency at which bits from the bit producing circuits are sampled; and a plurality of test circuits determining if bits sampled from each of the bit producing circuits are random, wherein the controller adjusts the sample frequency if the test circuits determine that the sampled bits are not random. A method performed by the apparatus is also included.