From f6b68ac4ba4c1da85867c9e3fb175b6e8410a8d0 Mon Sep 17 00:00:00 2001
From: Martin Isaksson <martin@martisak.se>
Date: Wed, 4 Oct 2023 15:31:02 +0200
Subject: [PATCH] Add publication
---
...g-using-personalized-federated-learning.md | 27 +++++++++++++++++++
1 file changed, 27 insertions(+)
create mode 100644 _publications/mmwave-beam-selection-in-analog-beamforming-using-personalized-federated-learning.md
diff --git a/_publications/mmwave-beam-selection-in-analog-beamforming-using-personalized-federated-learning.md b/_publications/mmwave-beam-selection-in-analog-beamforming-using-personalized-federated-learning.md
new file mode 100644
index 0000000..e41ebf4
--- /dev/null
+++ b/_publications/mmwave-beam-selection-in-analog-beamforming-using-personalized-federated-learning.md
@@ -0,0 +1,27 @@
+---
+title: mmWave Beam Selection in Analog Beamforming Using Personalized Federated Learning
+author: M. Isaksson, F. Vannella, D. Sandberg, R. Cöster
+in: To appear in IEEE Future Networks World Forum 2023
+date: 2023-10-03
+tags:
+ - 5G
+ - federated learning
+ - machine learning
+ - beam selection
+ - distributed learning
+ - federated learning
+ - privacy
+online: https://arxiv.org/abs/2310.00406
+abstract: |
+ Using analog beamforming in mmWave frequency bands we can focus the energy towards a receiver to achieve high throughput. However, this requires the network to quickly find the best downlink beam configuration in the face of non-IID data. We propose a personalized Federated Learning (FL) method to address this challenge, where we learn a mapping between uplink Sub-6GHz channel estimates and the best downlink beam in heterogeneous scenarios with non-IID characteristics. We also devise FedLion, a FL implementation of the Lion optimization algorithm. Our approach reduces the signalling overhead and provides superior performance, up to 33.6 % higher accuracy than a single FL model and 6 % higher than a local model.
+bibtex: |
+ @misc{isaksson2023mmwave,
+ title={mmWave Beam Selection in Analog Beamforming Using Personalized Federated Learning},
+ author={Martin Isaksson and Filippo Vannella and David Sandberg and Rickard Cöster},
+ year={2023},
+ eprint={2310.00406},
+ archivePrefix={arXiv},
+ }
+---
+
+{{ page.abstract }}
\ No newline at end of file
--
GitLab