Dr. Mathieu Chagnon

Nokia Bell Labs, Stuttgart, Germany

Mathieu Chagnon obtained his Ph.D. from McGill University, Canada, from the Department of Electrical and Computer Engineering in the Photonics Systems Group, during which he was awarded the IEEE Photonics Society Graduate Student Fellowship, the SPIE Optics and Photonics Education Scholarship, and the Alexander Graham Bell Canada Graduate Scholarship. He conducted his postdoctoral research at the University of Stuttgart, in collaboration with Nokia Bell Labs, for which he was awarded the prestigious Postdoctoral Fellowships from the Natural Sciences and Engineering Research Council of Canada. Dr. Chagnon made seminal contributions in the field of short-reach direct-detect systems, most notably on multi-dimensional self-beating formats, transceivers and digital signal processing. Dr. Chagnon co-authored a book chapter on high-speed interconnects for data center networking and gave Invited and Tutorial talks at major conference venues. He has authored/co-authored a substantial body of peer-reviewed journals and proceedings including several post deadlines, and holds a few active patents. He is currently a Researcher Engineer and Member of technical Staff in the Smart Optical Fabric and Devices Lab, Transmission and DSP, at Nokia Bell Labs in Stuttgart, Germany.

Talk Title 

Pre- and post-transmission digital signal processing using neural networks applied to fiber-optic communication systems relying on power modulation and power detection : Advantages and drawbacks

Abstract

Fiber-optic communication systems relying on power modulation and power detection are the least expensive systems for communication through optical fibers. Moreover, when using basic 2-level High-Low power format over very short distances, they are simple systems exhibiting an almost linear power transfer function. Today, we are facing an urge to drastically increase the binary throughput of fiber-optic transceivers while maintaining the total cost of ownership to a bare minimum. Transceivers based on Power Modulation–Power Detection are perfect candidate to meet the price target. However, these systems become highly nonlinear and exhibit a very complex transfer function when operated at higher spectral efficiencies, higher signaling rate, and over the distances of interest.

In this presentation we explain the nonlinearities and complexity of transceivers relying on power modulation and detection. We detail the challenges for reliable communication through this type of channel when trying to deliver high bitrates over a few tens of kilometers.

We discuss the use of neural networks as digital pre-processors and post-equalizers to apply at the communications endpoints, serving to optimize the transmission directly at the bit-level. We present the multiple advantages as well as the numerous drawbacks of this technology when applied at the physical layer of communication systems.