back to project list |
Full Name:Model Extraction to enable first time right Photonic Integrated Circuit design
- Luceda Photonics
- Caliopa - Huawei
- UGent - INTEC - PRG
- UGent - INTEC - SUMO
- The current design tools for Photonic Integrated Circuit (PIC) designers have fairly primitive models for the photonic components in comparison with the models available to electronic integrated circuit designers. However, the functioning and yield of photonic ICs is very dependent on variations in their fabrication process (more so than in electronic ICs). As a result, PIC design is far from first-time-right and requires many time consuming and costly iterations instead.
- With this project, the consortium wants to help advance the PIC design toolset by addressing the following major challenges:
- Lack of accurate photonic component models which incorporate process variability, enabling reliable simulations of complex and elaborate PICs without unwieldy computational loads.
- Lack of a methodology and software tools for automated extraction of the parameters of such models from test data.
- The goal of this project is to address the challenges by working towards the following main objectives:
- Characterize and model the location dependent aspect and the stochastic correlation length of the manufacturing process variability.
- Develop “white box” models (including non-idealities) and stochastic “black box” models for basic photonic IC components which incorporate the process variability model.
- Develop a methodology and the supporting software tool which can extract the parameters of a PIC component given a simulated or measured dataset and the model description (white or black box) while incorporating the process variability model.
- Develop analysis tools based on Luceda’s existing circuit/ssytem simulator, which can handle process variability (with and without spatial correlation) and corner simulations (Monte-Carlo).
- The circuit simulator will be used for validation of the research in this project and will not yet be optimized for computational time and/or memory usage.
Research topics involved
Publications in the framework of this project (2)
A. Kaintura, D. Spina, I. Couckuyt, L.Knockaert, W. Bogaerts, T. Dhaene,
A Kriging and Stochastic Collocation ensemble for uncertainty quantification in engineering applications, Engineering with Computers, p.1-15 (2017) .
A. Li, Y. Xing, R. Van Laer, R. Baets, W. Bogaerts,
Extreme Spectral Transmission Fluctuations in Silicon Nanowires Induced by Backscattering, IEEE International Conference on Group IV Photonics 2016, China, p.paper FB4 (2 pages) (2016) .
Back to project list