Optical communication is the only viable solution to respond to the demand for a high bit rate and long transmission distance. Directly modulated lasers (DMLs) are a cheap solution for modulating the light in optical fibre. Moreover, their hardware is simpler than externally modulated lasers. However, DML is inherently chirped and the transmission length with high bit rate is limited. This work explores and implements neural networks based signal predistortion schemes to create transmitters.

Machine Learning based Signal Generation Strategies for High-Speed Optical Transmitters

Rossi, Andrea
2019/2020

Abstract

Optical communication is the only viable solution to respond to the demand for a high bit rate and long transmission distance. Directly modulated lasers (DMLs) are a cheap solution for modulating the light in optical fibre. Moreover, their hardware is simpler than externally modulated lasers. However, DML is inherently chirped and the transmission length with high bit rate is limited. This work explores and implements neural networks based signal predistortion schemes to create transmitters.
2019-10-15
neural networks, DML, DSP
File in questo prodotto:
File Dimensione Formato  
andrea_rossi_tesi.pdf

accesso aperto

Dimensione 3.06 MB
Formato Adobe PDF
3.06 MB Adobe PDF Visualizza/Apri

The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.12608/24608