Transistor Rail Circuit for Gaussian Sampling With Tunable Thermal Noise

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current integrated circuits face challenges in accurately measuring and generating samples from Gaussian distributions due to practical limitations in capacitance and thermal noise, making it difficult to achieve high-resolution voltage fluctuations without introducing significant noise.

Innovation Solution

The use of CMOS-based circuits with transistor rails, where each transistor rail is configured with specific voltage settings and connected by capacitors to create a tunable effective temperature, allowing for the generation of Gaussian distributions by amplifying voltage fluctuations while preserving distribution shape.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If capacitance is increased to improve voltage fluctuation resolution, then measurement precision improves, but thermal noise increases

Engineering Contradiction:
Improvevoltage fluctuation resolutionVSAvoidthermal noise
Core Design Contradiction:
Measurement precisionVSObject-generated harmful factors

Solution Approach 1:

The patent replaces traditional voltage fluctuation generation mechanisms with a mechanical oscillator system. The mechanical oscillator generates physical vibrations that are transduced into electrical signals, substituting the thermal noise-based voltage fluctuations with mechanically-driven fluctuations. This allows high-resolution voltage fluctuations to be generated without the thermal noise penalty that would otherwise require large capacitance values.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Measurement precision

If digital sampling methods are used to generate Gaussian distribution samples, then measurement precision is maintained, but energy consumption increases

Engineering Contradiction:
Improvesampling accuracyVSAvoidenergy consumption
Core Design Contradiction:
Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The system uses the inherent thermal noise present in the mechanical oscillator and surrounding circuitry as a useful resource rather than treating it as waste. The thermal fluctuations naturally drive the mechanical oscillator, which in turn generates the voltage fluctuations needed for sampling. This self-service approach eliminates the need for external energy-intensive digital sampling while maintaining measurement precision.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The patent changes the fundamental parameter from digital voltage levels to mechanical displacement amplitudes. By measuring mechanical oscillator position or velocity rather than digital voltage samples, the system achieves Gaussian distribution sampling through physical phenomena rather than computational methods, dramatically reducing energy consumption while preserving accuracy.

Inventive Principle:
Principle #35Parameter changes

3Power

If transistor rail voltages are increased to amplify voltage fluctuations, then signal strength improves, but distribution shape distortion increases

Engineering Contradiction:
Improvesignal strengthVSAvoiddistribution shape
Core Design Contradiction:
PowerVSStability of the object's composition

Solution Approach 1:

The mechanical oscillator serves as an intermediary between the transistor rail voltage fluctuations and the measurement system. The mechanical system couples to the voltage fluctuations in a way that preserves the Gaussian distribution shape while providing mechanical amplification. The mechanical oscillator's natural resonance and linear response characteristics maintain distribution integrity even when transistor voltages are increased to enhance signal strength.

Inventive Principle:
Principle #24Intermediary (Mediator)

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 enables faster and more energy-efficient generation of samples from Gaussian distributions that accurately reflect physical phenomena, with potential energy savings of up to 10^5 to 10^6 times compared to digital sampling methods.

Implementation Method 1

Current integrated circuits face challenges in accurately measuring and generating samples from Gaussian distributions due to practical limitations in capacitance and thermal noise

Methodology Applied
Scientific EffectThermal noise:

Implementation Method 2

a first capacitor connected to the first node and the second node

Methodology Applied
Scientific EffectCapacitance: Capacitance

Data Source

PatentUS20260044167A1Managing arrangements of transistors in circuits for generating samples from a target distribution
Publication Date: 2026.02.12 EXTROPIC CORP
  • US20260044167A1 patent drawing
  • US20260044167A1 patent drawing
  • US20260044167A1 patent drawing

AI summary

An apparatus comprises: one or more voltage sources; a plurality of transistor rails, each transistor rail of the plurality of transistor rails comprising a first transistor connected to a first node and a second node, a second transistor connected to the first node, and a first capacitor connected to the first node and the second node; wherein the first node of each transistor rail of the plurality of transistor rails is connected to a first node of two different respective transistor rails of the plurality of transistor rails by two different respective capacitors; wherein one or more voltage sources are configured to apply a respective set of voltages to each transistor rail of the plurality of transistor rails.