We propose a Particle Swarm Optimization (PSO) algorithm for the extraction of Vertical-Cavity Surface-Emitting Laser (VCSEL) parameters compatible with a rate equation based model that takes into account the thermal effects. PSO is an evolutionary algorithm that drastically reduces the computational cost and time with respect to traditional brute-force approaches, thanks to the "swarm intelligence" of the agents of the optimization (called "particles"). With an optimal choice of the hyperparameters of the algorithm, the method is shown to predict parameters that accurately reproduce the non-linear behavior of the device, as well as its complicated thermal effects.
Particle swarm optimization hyperparameters tuning for physical-model fitting of VCSEL measurements / Marchisio, Andrea; Ghillino, Enrico; Curri, Vittorio; Carena, Andrea; Bardella, Paolo. - ELETTRONICO. - (2024). (Intervento presentato al convegno SPIE Photonic West tenutosi a San Francisco, California, United States nel 27 January - 1 February 2024) [10.1117/12.3002576].
Particle swarm optimization hyperparameters tuning for physical-model fitting of VCSEL measurements
Marchisio, Andrea;Curri, Vittorio;Carena, Andrea;Bardella, Paolo
2024
Abstract
We propose a Particle Swarm Optimization (PSO) algorithm for the extraction of Vertical-Cavity Surface-Emitting Laser (VCSEL) parameters compatible with a rate equation based model that takes into account the thermal effects. PSO is an evolutionary algorithm that drastically reduces the computational cost and time with respect to traditional brute-force approaches, thanks to the "swarm intelligence" of the agents of the optimization (called "particles"). With an optimal choice of the hyperparameters of the algorithm, the method is shown to predict parameters that accurately reproduce the non-linear behavior of the device, as well as its complicated thermal effects.File | Dimensione | Formato | |
---|---|---|---|
129040N.pdf
accesso aperto
Tipologia:
2a Post-print versione editoriale / Version of Record
Licenza:
PUBBLICO - Tutti i diritti riservati
Dimensione
355.25 kB
Formato
Adobe PDF
|
355.25 kB | Adobe PDF | Visualizza/Apri |
Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/11583/2986950