FD-SOI Ring Oscillator TRNG with Back-Gate Frequency Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing coherent sampling ring oscillator true random number generators (COSO TRNG) face challenges in setting the frequency of ring oscillators, leading to inefficiencies and increased power consumption.
Innovation Solution
A random number generation circuit using two identical ring oscillators implemented in CMOS on fully depleted silicon-on-insulator technology, with a control circuit to adjust the back gates of PMOS and NMOS transistors based on the period difference between the oscillators, ensuring a precise frequency control to minimize thermal noise and quantization noise.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the frequency of ring oscillators is manually set in known COSO TRNG circuits, then the implementation is simpler, but the frequency accuracy is insufficient leading to increased quantization noise
Solution Approach 1:
The patent implements a feedback mechanism where the sampling frequency is automatically adjusted based on the measured period difference between ring oscillators. The control circuit continuously monitors the period difference and adjusts the sampling frequency accordingly, eliminating the need for manual frequency setting while maintaining high accuracy. This resolves the contradiction by automating the frequency adjustment process through feedback control.
Solution Approach 2:
The system performs self-calibration by automatically measuring its own period difference and adjusting its sampling frequency without external intervention. The control circuit uses the inherent period difference measurement to self-regulate the sampling frequency, making the system self-sufficient in maintaining optimal operating conditions while reducing quantization noise.
2Reliability
If the sampling frequency is not precisely matched to ring oscillator frequency, then the implementation is easier, but quantization noise increases reducing random number quality
Solution Approach 1:
The control circuit continuously measures the period difference between ring oscillators and uses this feedback to automatically adjust the sampling frequency. This closed-loop system ensures the sampling frequency remains precisely matched to the ring oscillator frequency, maintaining high random number quality while eliminating the manual frequency matching process.
Solution Approach 2:
The system dynamically changes the sampling frequency parameter based on the measured period difference. By automatically adjusting this critical parameter, the system maintains optimal sampling conditions that maximize random number quality while simplifying operation, as the frequency matching is performed automatically rather than requiring manual intervention.
3Use of energy by moving object
If manual frequency setting is used in known COSO TRNG, then power consumption is higher due to less efficient operation, but automatic control circuit adds complexity
Solution Approach 1:
The automatic control circuit uses feedback from period difference measurements to optimize the sampling frequency, ensuring efficient operation that reduces power consumption. Although the control circuit adds complexity, it enables the system to operate at optimal efficiency points, compensating for the added complexity through improved energy utilization.
Solution Approach 2:
The system automatically adjusts the sampling frequency parameter to optimize power consumption based on the measured period difference. By dynamically changing this parameter to match optimal operating conditions, the system reduces power consumption despite the added control circuitry, as the automatic adjustment prevents inefficient operation.
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 solution achieves a more accurate frequency adjustment, reducing the period difference to a fraction of the oscillator period, thereby enhancing the unpredictability and efficiency of random number generation while minimizing power consumption.
Implementation Method 1
a contribution of the thermal noise is preponderating over a contribution of the quantization noise in the jitter of the output of the second oscillator
Implementation Method 2
a contribution of the thermal noise is preponderating over a contribution of the quantization noise in the jitter of the output of the second oscillator
Implementation Method 3
the randomness then resulting from the clock jitter, that is, over the interval between the theoretical period of a ring oscillator and the real or effective period of this oscillator
Data Source
AI summary
The present description concerns a random number generation circuit (2) of correlated sampling ring oscillator type comprising: two identical ring oscillators (RO1, R02) implemented in CMOS-on-FDSOI technology; a circuit (104) sampling and storing an output (O1) of one of the two oscillators (RO1) at a frequency of the other one of the two oscillators (RO2) and delivering a corresponding binary signal (Beat); and a circuit (200) controlling back gates of PMOS and NMOS transistors of at least one delay element of at least one of the two oscillators (RO1, RO2) based on a period difference between the two oscillators (RO1, RO2).