Ring Oscillator TRNG with Full-State Sampling for Faster Entropy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Ring oscillator-based true random number generators (TRNGs) are relatively slow compared to other entropy generation approaches, despite their simplicity and use of digital inverters and logic gates, which limits their speed while maintaining entropy quality.
Innovation Solution
A ring oscillator-based TRNG that samples the complete state of the oscillator, incorporating state compression and using a second ring oscillator for synchronization and meta-stability handling, allowing for faster operation and maintaining entropy quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a ring oscillator-based TRNG samples only a single output state, then the device complexity is reduced, but the productivity (entropy generation speed) is limited
Solution Approach 1:
The patent divides the ring oscillator state sampling into multiple segments by using multiple sampling circuits that sample different nodes (intermediate states) of the ring oscillator simultaneously. This segmentation allows parallel extraction of entropy from multiple points in the oscillator cycle, increasing overall entropy generation speed without requiring a single complex sampling circuit.
Solution Approach 2:
The patent transitions from sampling a single output state (one-dimensional) to sampling multiple intermediate states across different nodes of the ring oscillator (multi-dimensional). By adding the spatial dimension of multiple sampling points along the oscillator cycle, the system extracts more entropy per clock cycle, improving productivity without proportionally increasing complexity.
2Productivity
If a ring oscillator-based TRNG uses multiple sampling circuits to sample complete RO-state, then the productivity increases, but the device complexity increases
Solution Approach 1:
The patent merges the functions of multiple sampling circuits by having them sample different nodes of the same ring oscillator simultaneously. The sampled values from multiple circuits are then combined through logical operations (XOR, AND, OR) to generate the final random output. This merging approach allows parallel entropy extraction while using simple, identical sampling circuit units, scaling productivity without linearly increasing overall complexity.
Solution Approach 2:
Each sampling circuit in the patent is designed as a universal, multi-functional unit that can sample any node of the ring oscillator. The same sampling circuit design is replicated and applied to multiple nodes, making the system scalable. The sampling circuits serve multiple purposes: extracting entropy from different phases of the oscillator, providing redundancy, and enabling flexible configuration for different entropy rates.
3Reliability
If a ring oscillator-based TRNG samples complete RO-state with multiple nodes, then the entropy quality increases, but the loss of time for processing increases
Solution Approach 1:
The patent uses periodic sampling synchronized with the ring oscillator cycle. Multiple sampling circuits sample at different phases of the oscillator period, and the sampling is triggered periodically by a clock signal. This periodic action ensures that complete RO-states are captured efficiently without requiring continuous processing, reducing time loss while maintaining high entropy quality through comprehensive state sampling.
4Ease of manufacture
If a ring oscillator-based TRNG uses digital inverters and logic gates instead of analog circuits, then the ease of manufacture increases, but the speed is reduced
Solution Approach 1:
The patent maintains continuous operation of the ring oscillator and parallel sampling circuits, ensuring that entropy generation is ongoing without interruption. The digital logic circuits continuously sample the oscillator states and generate random bits in real-time. This continuous useful action maximizes the speed of entropy generation while keeping the digital circuit implementation simple and manufacturable, resolving the trade-off between ease of manufacture and speed.
Data Source
AI summary
A true random number generator circuit includes a ring oscillator and a plurality of sampling circuits. The ring oscillator includes a plurality of series-connected stages coupled together in a ring. An output of a last stage of the ring oscillator is coupled to an input of a first stage of the ring oscillator. A sampling circuit of the plurality of sampling circuits has an input coupled to a node located between two adjacent stages of the plurality of series-connected stages. Every node of the ring oscillator is coupled to a corresponding sampling circuit of the plurality of sampling circuits. In another embodiment, a method for generating a random number is provided.