P11 - Indirect Measurement Method Using Reconfigurable Nonintrusive Sensors for Integrated Sensory Electronics
- Event
- SMSI 2023
2023-05-08 - 2023-05-11
Nürnberg - Band
- Poster
- Chapter
- Poster
- Author(s)
- Q. Zaman, S. Alraho, A. König - Technical University of Kaiserslautern, Kaiserslautern (Germany)
- Pages
- 302 - 303
- DOI
- 10.5162/SMSI2023/P11
- ISBN
- 978-3-9819376-8-8
- Price
- free
Abstract
This paper presents a new approach for indirect measurement methods (IMs) by using reconfigurable non-intrusive sensors (NS) for the infield optimization of the reconfigurable integrated circuit with self-x properties. The typical IMs approach of using the regression model for the device under test (DUT) performance prediction is integrated with a metaheuristic optimization algorithm for the reconfigurable non-intrusive sensors. The novelty of this work comes from running the optimization algorithm on the NS by copying the same tuning knobs of the DUT, which allows for indirectly optimizing the DUT performance without interrupting its operation. Additionally, the infield optimization will be based on lowcost measurement of the embedded sensors. The achieved correlation performance metrics for the regression task is 90.13%. The DUT circuit is designed using XFAB 0.35 μm technology.