Zhou, Yalin;
Darwazeh, Izzat;
(2025)
An End-to-End Autoencoder-based Non-Orthogonal
Multi-Carrier System for Transmission.
In:
2025 31st International Conference on Telecommunications (ICT).
(pp. pp. 1-5).
IEEE: Budva, Montenegro.
Preview |
Text
Zhou_ICTfinalVersion.pdf Download (478kB) | Preview |
Abstract
Multi-carrier techniques remain prime candidates for high-frequency 6 G and sub-THz wireless transmission, due their resilience to multi-path effects. Spectrally efficient frequency division multiplexing (SEFDM) conserves bandwidth relative to orthogonal frequency division multiplexing (OFDM) but similarly suffers from high PAPR, which affects the power efficiency of mmWave amplifiers. Moreover, SEFDM suffers from high Intercarrier Interference (ICI) as a result of bandwidth compression, which limits SEFDM's applications and practical deployment. To ameliorate the effects of these two problems, this work proposes an end-to-end autoencoder-based SEFDM system that generates optimized constellation mapping at the transmitter side for reducing PAPR. at the receiver side, a neural network joint SEFDM detection and demodulation is implemented to remove much of the ICI. Simulation based results show that the proposed autoencoder-based SEFDM system achieves BER, which outperforms linear detection techniques. Furthermore, the results show that our optimized constellation improves the PAPR, in specific threshold regions, when compared to conventional QPSK-mapped OFDM and SEFDM signals.
| Type: | Proceedings paper |
|---|---|
| Title: | An End-to-End Autoencoder-based Non-Orthogonal Multi-Carrier System for Transmission |
| Event: | 2025 31st International Conference on Telecommunications (ICT) |
| Location: | MONTENEGRO, Budva |
| Dates: | 28 Apr 2025 - 29 Apr 2025 |
| ISBN-13: | 979-8-3315-1447-1 |
| Open access status: | An open access version is available from UCL Discovery |
| DOI: | 10.1109/ICT65093.2025.11046292 |
| Publisher version: | https://doi.org/10.1109/ict65093.2025.11046292 |
| Language: | English |
| Additional information: | This version is the author accepted manuscript. For information on re-use, please refer to the publisher’s terms and conditions. |
| Keywords: | Autoencoder, OFDM, SEFDM, Deep Learning |
| UCL classification: | UCL UCL > Provost and Vice Provost Offices > UCL BEAMS UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science > Dept of Electronic and Electrical Eng |
| URI: | https://discovery.ucl.ac.uk/id/eprint/10219265 |
Archive Staff Only
 |
View Item |
