Cover Image

Spiking neural network nonlinear demapping on neuromorphic hardware for IM/DD optical communication

Neuromorphic computing implementing spiking neural networks (SNN) is a promising technology for reducing the footprint of optical transceivers, as required by the fast-paced growth of data center traffic. In this work, an SNN nonlinear demapper is designed and evaluated on a simulated intensity-modu...

Full description

Saved in:
Bibliographic Details
Main Authors: Arnold, Elias (Author) , Böcherer, Georg (Author) , Strasser, Florian (Author) , Müller, Eric (Author) , Spilger, Philipp (Author) , Billaudelle, Sebastian (Author) , Weis, Johannes (Author) , Schemmel, Johannes (Author) , Calabrò, Stefano (Author) , Kuschnerov, Maxim (Author)
Format: Article (Journal)
Language:English
Published: 01 June 2023
In: Journal of lightwave technology
Year: 2023, Volume: 41, Issue: 11, Pages: 3424-3431
ISSN:1558-2213
DOI:10.1109/JLT.2023.3252819
Online Access:Verlag, lizenzpflichtig, Volltext: https://dx.doi.org/10.1109/JLT.2023.3252819
Get full text
Author Notes:Elias Arnold, Georg Böcherer, Florian Strasser, Eric Müller, Philipp Spilger, Sebastian Billaudelle, Johannes Weis, Johannes Schemmel, Stefano Calabrò, Maxim Kuschnerov
Description
Summary:Neuromorphic computing implementing spiking neural networks (SNN) is a promising technology for reducing the footprint of optical transceivers, as required by the fast-paced growth of data center traffic. In this work, an SNN nonlinear demapper is designed and evaluated on a simulated intensity-modulation direct-detection link with chromatic dispersion. The SNN demapper is implemented in software and on the analog neuromorphic hardware system BrainScaleS-2 (BSS-2). For comparison, linear equalization (LE), Volterra nonlinear equalization (VNLE), and nonlinear demapping by an artificial neural network (ANN) implemented in software are considered. At a pre-forward error correction bit error rate of 2×10−3 , the software SNN outperforms LE by 1.5 dB, VNLE by 0.3 dB and the ANN by 0.5 dB. The hardware penalty of the SNN on BSS-2 is only 0.2 dB, i.e., also on hardware, the SNN performs better than all software implementations of the reference approaches. Hence, this work demonstrates that SNN demappers implemented on electrical analog hardware can realize powerful and accurate signal processing fulfilling the strict requirements of optical communications.
Item Description:Veröffentlicht: 6. März 2023
Gesehen am 17.08.2023
Physical Description:Online Resource
ISSN:1558-2213
DOI:10.1109/JLT.2023.3252819

Similar Items