Thesis
A machine learning execution framework for microwave transistor behavior prediction
Washington State University
Master of Science (MS), Washington State University
05/2021
DOI:
https://doi.org/10.7273/000000048
Handle:
https://hdl.handle.net/2376/112458
Abstract
In this thesis, multiple machine learning models are applied to predict the performance of a Gallium Nitride Radio Frequency power transistor during a nested
fundamental and second harmonic load pull. Taking the resource constraint into consideration, we propose an execution framework to make a model selection, so that
the prediction performance meets the accuracy goal but runs with relatively low computational resources. We show how this framework works to select a suitable model
and produce a relatively accurate result under resource constraints. In addition, we
also applied a greedy algorithm based on q-learning to reduce the size of the training
dataset.
Metrics
27 File views/ downloads
38 Record Views
Details
- Title
- A machine learning execution framework for microwave transistor behavior prediction
- Creators
- WEI LI
- Contributors
- XINGHUI ZHAO (Degree Supervisor) - Washington State University, Engineering and Computer Science (VANC), School ofSCOTT WALLACE (Committee Member) - Washington State University, Engineering and Computer Science (VANC), School ofBEN MCCAMISH (Committee Member) - Washington State University, Engineering and Computer Science (VANC), School of
- Awarding Institution
- Washington State University
- Academic Unit
- Engineering and Computer Science (VANC), School of
- Theses and Dissertations
- Master of Science (MS), Washington State University
- Publisher
- Washington State University
- Format
- pdf
- Number of pages
- 82
- Identifiers
- 99900586063701842
- Resource Type
- Thesis