Machine Learning assisted fitting of VCSEL parameters
keywords LASER A SEMICONDUTTORE, MACHINE LEARNING
Reference persons PAOLO BARDELLA
Description The goal the thesis is to identify a procedure for the automatic extraction of fitting parameters (cavity properties, material characteristics, …) for Vertical Cavity lasers, able to reproduce experimental evidences (on L-I curves, optical spectra, RIN, …) using the VCSEL model implemented in Synopsys Optsim®.
The idea behind this thesis is to use the implemented model (available both in Optsim and as a MATLAB script) performing a large number of simulations varying the parameters and over reasonable ranges and training a Neural Network (NN). Once trained, the NN should be able to return the best combination of parameters able to fit any provided experimental data.
No prerequisites on ML are required and no previous experience with Optsim is needed.
Deadline 26/10/2023 PROPONI LA TUA CANDIDATURA