Design Space Exploration of Current-Starved Ring Oscillators Based on Graphene Nanoribbons

Journal Article (2025)
Author(s)

Florin Silviu Dumitru (Politehnica University of Bucharest)

N. Cucu Laurenciu (TU Delft - Computer Engineering, Radboud Universiteit Nijmegen)

Alexandru Antonescu (Politehnica University of Bucharest)

S.D. Cotofana (TU Delft - Computer Engineering)

Marius Enachescu (Politehnica University of Bucharest)

Research Group
Computer Engineering
DOI related publication
https://doi.org/10.59277/ROMJIST.2025.2.07
More Info
expand_more
Publication Year
2025
Language
English
Research Group
Computer Engineering
Issue number
2
Volume number
28
Pages (from-to)
197-209
Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Abstract

Graphene is well-suited for ultra-low-power (ULP) nano-electronics due to its exceptional characteristics like ballistic transport and its ability to engineer structures fea-turing a geometry-induced bandgap. Identifying the conditions necessary for achieving the maximum level of performance and of power-efficiency frequently requires a design space exploration (DSE). By means of calibration and external regulation of the supply voltage, the ULP graphene-nanoribbon (GNR) ring oscillator presented in this paper is capable of exceeding the performance of its 7 nm FinFET counterpart both in terms of maximum frequency and of power-efficiency. Under nominal supply voltage conditions we achieve a 1.89× higher output frequency while simultaneously reducing the power consumption by 553.8× and achieving a 812× higher power efficiency. After performing a DSE and lever-aging both externally-applied supply voltage modulation and output frequency calibration we achieved a 4.81× higher maximum output frequency operation mode and a 242× higher maximum power-efficiency operation mode when configuring both blocks for peak performance for each mode.