Analog Random Sequence Generator Using Resistive-Feedback Noise
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional random bit sequence generators in complex electronic systems, such as GPUs and CPUs, face challenges in generating sufficiently random sequences due to non-random internal operations, which can be exploited by hackers to increase the likelihood of successful hacking.
Innovation Solution
Analog mechanisms using noise sources, including inverters with resistive feedback, are employed to generate random thermal noise that is applied to ring oscillators or sense amplifiers, enhancing noise sources and improving randomness through techniques like resistive feedback, tunable inverter ratios, and differential outputs, while suppressing supply noise to maintain high entropy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional digital logic is used for random sequence generation, then device complexity is reduced and ease of manufacture is improved, but the randomness and entropy of generated sequences deteriorate
Solution Approach 1:
The patent replaces digital logic mechanisms with analog physical mechanisms (ring oscillators, sense amplifiers, noise sources) to generate random sequences. The analog circuitry exploits thermal noise and oscillation variability to produce high-entropy random bits, substituting the conventional digital logic approach that relies on pseudo-random algorithms.
Solution Approach 2:
The patent changes the operating parameters of the analog circuits (oscillator frequencies, sense amplifier thresholds, noise source characteristics) to optimize randomness generation. By tuning these parameters, the system achieves high entropy output while managing the complexity of the analog circuitry.
2Reliability
If noise sources are enhanced to improve randomness, then entropy of generated sequences is improved, but supply noise and interference increase
Solution Approach 1:
The patent extracts and utilizes thermal noise from dedicated noise sources (resistors, transistors) as the primary entropy input, separating this useful noise from harmful supply noise. The design specifically taps into intrinsic thermal noise mechanisms while filtering out power supply interference through differential signaling and synchronous sampling.
Solution Approach 2:
The patent converts the inherently noisy analog nature of oscillators and sense amplifiers into a benefit by exploiting their sensitivity to thermal noise. The natural variability and noise in these analog circuits, which could be considered harmful in digital contexts, becomes the primary source of high-entropy random bits.
3Reliability
If analog mechanisms are used to generate random sequences, then randomness and entropy are improved, but manufacturing precision and device variability increase
Solution Approach 1:
The patent incorporates calibration mechanisms that perform preliminary adjustments during manufacturing or initialization. The calibration circuits measure and compensate for device-specific variations in oscillators and sense amplifiers, storing correction factors that are applied during operation to ensure consistent randomness quality across different manufactured devices.
Solution Approach 2:
The patent employs feedback mechanisms where the output of sense amplifiers is fed back to adjust the operation of ring oscillators and noise sources. This feedback loop continuously optimizes the random sequence generation by compensating for drift and variability in analog parameters, maintaining high entropy output despite manufacturing variations.
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
The proposed solution enhances the randomness and entropy of generated bit sequences, reducing correlation and maintaining high throughput, thus improving the security of electronic systems against hacking.
Implementation Method 1
A noise source comprising an inverter with restive feedback generates random thermal noise
Data Source
AI summary
Random sequence generators utilizing one or more noise generators that include an inverter chain with at least one input stage inverter configured with resistive feedback and additional inverters configured in series with the at least one input stage inverter, wherein an output of the inverter chain is coupled to a bit sequence generator.