TU3.R1.3

Neural Network Equalizers and Successive Interference Cancellation for Bandlimited Channels with a Nonlinearity

Daniel Plabst, Tobias Prinz, Francesca Diedolo, Thomas Wiegart, Technical University of Munich, Germany; Georg Böcherer, Huawei Technologies Düsseldorf GmbH, Germany; Norbert Hanik, Gerhard Kramer, Technical University of Munich, Germany
Session:
Deep Learning in Communications
Track:
8: Deep Learning (such as understanding large language models)
Location:
Ballroom II & III
Presentation Time:
Tue, 9 Jul, 15:05 - 15:25
Session Chair:
Iñaki Esnaola,
Abstract
Neural networks (NNs) inspired by the forward-backward algorithm (FBA) are used as equalizers for bandlimited channels with a memoryless nonlinearity. The NN-equalizers are combined with successive interference cancellation (SIC) to approach the information rates of joint detection and decoding (JDD) with considerably less complexity than JDD and other existing equalizers. Simulations for short-haul optical fiber links with square-law detection illustrate the gains.
Session TU3.R1
TU3.R1.1: PAC Learnability for Reliable Communication over Discrete Memoryless Channels
Jiakun Liu, Wenyi Zhang, University of Science and Technology of China, China; H. Vincent Poor, Princeton University, United States
TU3.R1.2: Neural Estimation of Multi-User Capacity Regions over Discrete Channels
Bashar Huleihel, Dor Tsur, Ziv Aharoni, Ben Gurion University of the Negev, Israel; Oron Sabag, The Hebrew University of Jerusalem, Israel; Haim Permuter, Ben Gurion University of the Negev, Israel
TU3.R1.3: Neural Network Equalizers and Successive Interference Cancellation for Bandlimited Channels with a Nonlinearity
Daniel Plabst, Tobias Prinz, Francesca Diedolo, Thomas Wiegart, Technical University of Munich, Germany; Georg Böcherer, Huawei Technologies Düsseldorf GmbH, Germany; Norbert Hanik, Gerhard Kramer, Technical University of Munich, Germany
TU3.R1.4: Graph Neural Network-based Joint Equalization and Decoding
Jannis Clausius, Marvin Geiselhart, Daniel Tandler, Stephan ten Brink, University of Stuttgart, Germany
Resources
View Manuscript