Abstract
Recent works showed that operating amplify-andforward relays in their nonlinear regimes can bring huge performance gains. However, the training of such a nonlinear network requires a knowledge of the channel gains. This work proposes a novel triplet network approach, where we copy the actual network into both an estimation twin and an optimization twin. The channel estimation is based on the transmission of data from a transmitter while taking measurements only at the network output. We show that this novel approach succeeds in estimating channel gains even in deep relay networks where measurements are taken through 12 layers of relays.
| Original language | English |
|---|---|
| Title of host publication | SPAWC 2025 - 2025 IEEE 26th International Workshop on Signal Processing and Artificial Intelligence for Wireless Communications - Proceedings |
| Publisher | Institute of Electrical and Electronics Engineers Inc. |
| ISBN (Electronic) | 9781665477765 |
| DOIs | |
| State | Published - 2025 |
| Event | 26th IEEE International Workshop on Signal Processing and Artificial Intelligence for Wireless Communications, SPAWC 2025 - Surrey, United Kingdom Duration: 7 Jul 2025 → 10 Jul 2025 |
Publication series
| Name | IEEE Workshop on Signal Processing Advances in Wireless Communications, SPAWC |
|---|---|
| ISSN (Print) | 2325-3789 |
Conference
| Conference | 26th IEEE International Workshop on Signal Processing and Artificial Intelligence for Wireless Communications, SPAWC 2025 |
|---|---|
| Country/Territory | United Kingdom |
| City | Surrey |
| Period | 7/07/25 → 10/07/25 |
Bibliographical note
Publisher Copyright:© 2025 IEEE.