1
|
Hutchins J, Alam S, Rampini DS, Oripov BG, McCaughan AN, Aziz A. Machine learning-powered compact modeling of stochastic electronic devices using mixture density networks. Sci Rep 2024; 14:6383. [PMID: 38493250 PMCID: PMC10944466 DOI: 10.1038/s41598-024-56779-8] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/08/2023] [Accepted: 03/11/2024] [Indexed: 03/18/2024] Open
Abstract
The relentless pursuit of miniaturization and performance enhancement in electronic devices has led to a fundamental challenge in the field of circuit design and simulation-how to accurately account for the inherent stochastic nature of certain devices. While conventional deterministic models have served as indispensable tools for circuit designers, they fall short when it comes to capturing the subtle yet critical variability exhibited by many electronic components. In this paper, we present an innovative approach that transcends the limitations of traditional modeling techniques by harnessing the power of machine learning, specifically Mixture Density Networks (MDNs), to faithfully represent and simulate the stochastic behavior of electronic devices. We demonstrate our approach to model heater cryotrons, where the model is able to capture the stochastic switching dynamics observed in the experiment. Our model shows 0.82% mean absolute error for switching probability. This paper marks a significant step forward in the quest for accurate and versatile compact models, poised to drive innovation in the realm of electronic circuits.
Collapse
Affiliation(s)
- Jack Hutchins
- Department of Electrical Engineering & Computer Science, University of Tennessee, Knoxville, TN, 37996, USA
| | - Shamiul Alam
- Department of Electrical Engineering & Computer Science, University of Tennessee, Knoxville, TN, 37996, USA
| | - Dana S Rampini
- National Institute of Standards and Technology, Boulder, Co, 80305, USA
| | - Bakhrom G Oripov
- National Institute of Standards and Technology, Boulder, Co, 80305, USA
| | - Adam N McCaughan
- National Institute of Standards and Technology, Boulder, Co, 80305, USA
| | - Ahmedullah Aziz
- Department of Electrical Engineering & Computer Science, University of Tennessee, Knoxville, TN, 37996, USA.
| |
Collapse
|
2
|
Oripov BG, Rampini DS, Allmaras J, Shaw MD, Nam SW, Korzh B, McCaughan AN. A superconducting nanowire single-photon camera with 400,000 pixels. Nature 2023; 622:730-734. [PMID: 37880435 DOI: 10.1038/s41586-023-06550-2] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/29/2023] [Accepted: 08/17/2023] [Indexed: 10/27/2023]
Abstract
For the past 50 years, superconducting detectors have offered exceptional sensitivity and speed for detecting faint electromagnetic signals in a wide range of applications. These detectors operate at very low temperatures and generate a minimum of excess noise, making them ideal for testing the non-local nature of reality1,2, investigating dark matter3,4, mapping the early universe5-7 and performing quantum computation8-10 and communication11-14. Despite their appealing properties, however, there are at present no large-scale superconducting cameras-even the largest demonstrations have never exceeded 20,000 pixels15. This is especially true for superconducting nanowire single-photon detectors (SNSPDs)16-18. These detectors have been demonstrated with system detection efficiencies of 98.0% (ref. 19), sub-3-ps timing jitter20, sensitivity from the ultraviolet21 to the mid-infrared22 and microhertz dark-count rates3, but have never achieved an array size larger than a kilopixel23,24. Here we report on the development of a 400,000-pixel SNSPD camera, a factor of 400 improvement over the state of the art. The array spanned an area of 4 × 2.5 mm with 5 × 5-μm resolution, reached unity quantum efficiency at wavelengths of 370 nm and 635 nm, counted at a rate of 1.1 × 105 counts per second (cps) and had a dark-count rate of 1.0 × 10-4 cps per detector (corresponding to 0.13 cps over the whole array). The imaging area contains no ancillary circuitry and the architecture is scalable well beyond the present demonstration, paving the way for large-format superconducting cameras with near-unity detection efficiencies across a wide range of the electromagnetic spectrum.
Collapse
Affiliation(s)
- B G Oripov
- National Institute of Standards and Technology, Boulder, CO, USA.
- Department of Physics, University of Colorado Boulder, Boulder, CO, USA.
| | - D S Rampini
- National Institute of Standards and Technology, Boulder, CO, USA
- Department of Physics, University of Colorado Boulder, Boulder, CO, USA
| | - J Allmaras
- Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA, USA
| | - M D Shaw
- Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA, USA
| | - S W Nam
- National Institute of Standards and Technology, Boulder, CO, USA
| | - B Korzh
- Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA, USA
| | - A N McCaughan
- National Institute of Standards and Technology, Boulder, CO, USA
| |
Collapse
|