Presentation
19 June 2024 A photonic neural network for chromatic dispersion compensation in OFDM signals
Emiliano Staffoli, Gianpietro Maddinelli, Pablo R. N. Marciano, Marcelo E. V. Segatto, Lorenzo Pavesi
Author Affiliations +
Abstract
In Intensity Modulation/Direct Detection protocols, chromatic dispersion is a major impairment source, inducing intersymbol interference which leads to BER degradation at the receiver. Fiber Bragg gratings and Dispersion Compensating fibers are nowadays used for equalization, despite the former showing limited performance in compensating all Wavelength Division Multiplexing aggregate channels simultaneously, and the latter being non-tunable devices that introduce latency. In former works, we demonstrated channel equalization via an all-optical delayed complex perceptron, which is implemented as an integrated feed-forward photonic neural network trained for PAM2, PAM4, and PAM8 signals up to 125 km. The network performs an all-optical signal processing with minimized latency, reconfigurability, and low power consumption. This design is now applied for the equalization of complex modulation formats encoded in Orthogonal Frequency Division Multiplexed signals. This allows for an increase in spectral efficiency and enforces the adaptability of the network to different modulation formats.
Conference Presentation
© (2024) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Emiliano Staffoli, Gianpietro Maddinelli, Pablo R. N. Marciano, Marcelo E. V. Segatto, and Lorenzo Pavesi "A photonic neural network for chromatic dispersion compensation in OFDM signals", Proc. SPIE PC13012, Integrated Photonics Platforms III, (19 June 2024); https://doi.org/10.1117/12.3016394
Advertisement
Advertisement
KEYWORDS
Dispersion

Neural networks

Orthogonal frequency division multiplexing

Modulation

Silicon

Signal intensity

Silica

RELATED CONTENT


Back to Top