← Back to all news
arXiv Machine Learning By Kyriakos Stylianopoulos, Mattia Fabiani, Davide Dardari, George C. Alexandropoulos

XL MIMO Extreme Learning Machines: Wave-Domain Inference with Stacked Metasurfaces

Read the original on arXiv Machine Learning →

The paper proposes an XL‑MIMO system with stacked intelligent metasurfaces (SIM) that functions as a two‑layer neural network operating entirely over the air. The first layer is the uncontrollable wireless fading, while the second controllable layer is implemented via SIM operations, requiring only the front‑end metasurface to provide a nonlinear activation. Using the Extreme Learning Machine framework, the authors derive a closed‑form optimal linear SIM combining, enabling fast training and demonstrating classification accuracy comparable to digital models across various datasets.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

Related stories

arXiv Machine Learning
Aug 28

Over-The-Air Extreme Learning Machines with Nonlinear Stacked Intelligent Metasurfaces

The paper proposes an eXtremely Large MIMO system that functions as an Extreme Learning Machine for over‑the‑air binary classification. It uses cascaded metasurfaces, with a front layer providing a fixed nonlinear activation and subsequent tunable linear layers implementing trained weights directly in the wave domain. Numerical results on various datasets show that this low‑complexity, wave‑domain architecture achieves classification accuracy comparable to ideal digital models.

By Kyriakos Stylianopoulos, Mattia Fabiani, Giulia Torcolacci, Davide Dardari, George C. Alexandropoulos
More like this →
arXiv Machine Learning
Aug 7

EqDeepRx: Learning a Scalable and Interference Mitigating MIMO Receiver

arXiv:2602. 11834v2 Announce Type: replace-cross Abstract: While machine learning (ML)-based receiver algorithms have received a great deal of attention in the recent literature, they often suffer from poor scaling with increasing spatial multiplexing order and lack of explainability and generalization.

By Mikko Honkala, Dani Korpi, Elias Raninen, Janne M. J. Huttunen
More like this →
arXiv Machine Learning
Oct 16

Analog RF Computing for Integrated Data and AI Flows Over Wireless Networks

The paper proposes using analog radio frequency (RF) computing to enable edge devices to perform AI inference by receiving a waveform that encodes neural network weights from a base station. Clients then mix this waveform with their own input-encoded signals to perform matrix‑vector multiplications with minimal power and memory. In a multi‑user MIMO downlink setting, the authors formulate a joint beamforming optimization to balance data and AI flows, showing that the aggregate data rate loss is below 5% under realistic 3GPP channel conditions.

By Wentao Yu, Vincent W. S. Wong
More like this →
arXiv Machine Learning
Oct 8

Deep Learning-Based Tri-Hybrid Multi-User MIMO Precoding: The Blessing of EM-Reconfigurable Antennas

The paper introduces Tri-PNet, a deep learning framework that jointly designs electromagnetic (EM), analog, and digital precoding for tri-hybrid multi-user MIMO OFDM systems. By leveraging Conformer architecture, Tri-PNet learns to select radiation patterns and perform hybrid precoding, achieving higher spectral efficiency than random EM precoding and conventional hybrid MIMO while maintaining low online complexity. The study evaluates two radiation-pattern modes—non-regular and 3GPP TR 38.901—showing robust performance even with imperfect channel state information.

By Kaijun Feng, Jiaxin He, Hongrui Yu, Zhen Gao, Anwen Liao, Ziwei Wan, Zhaocheng Wang
llms
More like this →