Channel charting based beamforming

Luc Le Magoarou (1, 2, 3, 4, 5) , Taha Yassine (1, 2, 3, 4, 5) , Stephane Paquelet (1, 2, 3, 5) , Matthieu Crussière (1, 2, 3, 4, 5)

1 Artificial Intelligence
2 IRT b-com - Institut de Recherche Technologique b-com
3 Hypermedia - Domaine Hypermedia (IRT b<>com)
4 INSA Rennes - Institut National des Sciences Appliquées - Rennes
5 IETR - Institut d'Électronique et des Technologies du numéRique

Channel charting (CC) is an unsupervised learning method allowing to locate users relative to each other without reference. From a broader perspective, it can be viewed as a way to discover a low-dimensional latent space charting the channel manifold. In this paper, this latent modeling vision is leveraged together with a recently proposed location-based beamforming (LBB) method to show that channel charting can be used for mapping channels in space or frequency. Combining CC and LBB yields a neural network resembling an autoencoder. The proposed method is empirically assessed on a channel mapping task whose objective is to predict downlink channels from uplink channels.

Domains: Networking and Internet Architecture [cs.NI]; Machine Learning [cs.LG]; Signal and Image processing

Luc Le Magoarou, Taha Yassine, Stephane Paquelet, Matthieu Crussière. Channel charting based beamforming. Asilomar Conference on Signals, Systems, and Computers, Oct 2022, Pacific Grove, United States. ⟨hal-03882861v2⟩