Semi-Supervised Variational Inference over Nonlinear Channels

David Burshtein*, Eli Bery

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Deep learning methods for communications over unknown nonlinear channels have attracted considerable interest recently. In this paper, we consider semi-supervised learning methods, which are based on variational inference, for decoding unknown nonlinear channels. These methods, which include Monte Carlo expectation maximization and a variational autoencoder, make efficient use of few pilot symbols and the payload data. The best semi-supervised learning results are achieved with a variational autoencoder. For sufficiently many payload symbols, the variational autoencoder also has lower error rate compared to meta learning that uses the pilot data of the present as well as previous transmission blocks.

Original languageEnglish
Title of host publication2023 IEEE 24th International Workshop on Signal Processing Advances in Wireless Communications, SPAWC 2023 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages611-615
Number of pages5
ISBN (Electronic)9781665496261
DOIs
StatePublished - 2023
Event24th IEEE International Workshop on Signal Processing Advances in Wireless Communications, SPAWC 2023 - Shanghai, China
Duration: 25 Sep 202328 Sep 2023

Publication series

NameIEEE Workshop on Signal Processing Advances in Wireless Communications, SPAWC

Conference

Conference24th IEEE International Workshop on Signal Processing Advances in Wireless Communications, SPAWC 2023
Country/TerritoryChina
CityShanghai
Period25/09/2328/09/23

Funding

FundersFunder number
Israel Science Foundation1868/18

    Keywords

    • Channel estimation
    • semi-supervised learning
    • variational autoencoders
    • variational inference

    Fingerprint

    Dive into the research topics of 'Semi-Supervised Variational Inference over Nonlinear Channels'. Together they form a unique fingerprint.

    Cite this