scispace - formally typeset
W

Walker J. Turner

Researcher at Nvidia

Publications -  16
Citations -  203

Walker J. Turner is an academic researcher from Nvidia. The author has contributed to research in topics: Computer science & Phase-locked loop. The author has an hindex of 4, co-authored 10 publications receiving 77 citations.

Papers
More filters
Proceedings ArticleDOI

ParaGraph: Layout Parasitics and Device Parameter Prediction using Graph Neural Networks

TL;DR: ParaGraph: a graph neural network model to predict net parasitics and device parameters by converting circuit schematics into graphs and leveraging key modeling techniques based on GraphSage, Relation GCN and Graph Attention Networks increases model accuracy over a large range of prediction values.
Journal ArticleDOI

A 1.17-pJ/b, 25-Gb/s/pin Ground-Referenced Single-Ended Serial Link for Off- and On-Package Communication Using a Process- and Temperature-Adaptive Voltage Regulator

TL;DR: This paper describes a short-reach serial link to connect chips mounted on the same package or on neighboring packages on a printed circuit board (PCB) that employs an energy-efficient, single-ended ground-referenced signaling scheme and employs a novel power supply regulation scheme at both ends.
Proceedings ArticleDOI

A 1.17pJ/b 25Gb/s/pin ground-referenced single-ended serial link for off- and on-package communication in 16nm CMOS using a process- and temperature-adaptive voltage regulator

TL;DR: The combination of signal integrity challenges and production margining requires increased amplitude, equalization, ESD protection, and PVT-tolerant circuit design techniques.
Proceedings ArticleDOI

Ground-referenced signaling for intra-chip and short-reach chip-to-chip interconnects

TL;DR: Various methodologies compatible with GRS are presented, including design considerations and various circuit architectures, which enables simple but efficient signaling across on-chip communication fabrics, off-chip organic packages, and off-package printed circuit boards.
Proceedings ArticleDOI

Parasitic-Aware Analog Circuit Sizing with Graph Neural Networks and Bayesian Optimization

TL;DR: In this article, the authors proposed an improved surrogate performance model using parasitic graph embeddings from the pre-trained parasitic prediction network, which has 20% better R2 prediction score and improves optimization convergence by 3.7 times and 2.1 times compared to conventional Gaussian process regression and neural network based Bayesian linear regression.