Artificial Neural Network Based Demodulation of Multi-Eigenvalue Modulated Signal

Shohei Yamamoto

Presentation information

Oral Sessions

P2. Photonics in Switching Technologies, Systems, and Architectures for Computing and Big Data

[TuF3] Optical Signal Functionalization III

Tue. Jul 9, 2019 2:00 PM - 3:30 PM Room F (412, 4F, Fukuoka International Congress Center)

Presider: Daniel C. Kilper (University of Arizona, USA)

2:15 PM - 2:30 PM

[TuF3-2] Artificial Neural Network Based Demodulation of Multi-Eigenvalue Modulated Signal

〇Shohei Yamamoto, Ken Mishina, Akihiro Maruta (Osaka Univ., Japan)

Keywords:Advanced modulation and demodulation, Transmission and Propagation , Advanced Modulation, Coding and Multiplexing

We propose a machine learning based demodulation method for optical multi-eigenvalue modulated signal. The proposed method shows 9.2 dB power improvement at BER = 3.8×10^-3 compared with the IST based demodulation method.

Manuscript password authentication.
Password is required to view the manuscript. Please enter a password to authenticate.