A Learnable SIM Paradigm: Fundamentals, Training Techniques, and Applications

Hetong Wang; Yashuai Cao; Tiejun Lv

eess.SPMar 13, 2026

A Learnable SIM Paradigm: Fundamentals, Training Techniques, and Applications

Authors: Hetong Wang, Yashuai Cao, Tiejun Lv

Organizations: School of Information and Communication Engineering, Beijing University of Posts and Telecommunications (BUPT), Beijing 100876, China · School of Intelligence Science and Technology, University of Science and Technology Beijing, Beijing 100083, China

Abstract

Stacked intelligent metasurfaces (SIMs) represent a breakthrough in wireless hardware by comprising multilayer, programmable metasurfaces capable of analog computing in the electromagnetic (EM) wave domain. By examining their architectural analogies, this article reveals a deeper connection between SIMs and artificial neural networks (ANNs). Leveraging this profound structural similarity, this work introduces a learnable SIM architecture and proposes a learnable SIM-based machine learning (ML) paradigm for sixth-generation (6G)-andbeyond systems. Then, we develop two SIM-empowered wireless signal processing schemes to effectively achieve multi-user signal separation and distinguish communication signals from jamming signals. The use cases highlight that the proposed SIM-enabled signal processing system can significantly enhance spectrum utilization efficiency and anti-jamming capability in a lightweight manner and pave the way for ultra-efficient and intelligent wireless infrastructures.

Figures & tables

Fig. 1: ML paradigm with the SIM-enabled learnable architecture including: (a) The SIM provides a physical platform for implementing a learnable architecture inspired by ANNs; (b) The SIM is integrated with the BS as the combiner to effectively distinguish multi-user transmission signals in MU-MISO uplink systems; (c) The learnable SIM can be further designed to separate communication and jamming signals, and defend against Mallory’s jamming attack.

Fig. 2: Visualization of end-to-end quasi-orthogonal subchannel formation via SIM-based wavefront processing, with $K=4$ users, and $N=100$ meta-atoms, compared with conventional RISs.

Fig. 3: Average loss functions vs. episodes for different

\eta_{0}

and

\beta

; constellation diagrams for different values of

N

, and invariant angles and ranged distance.

Fig. 4: Constellation diagrams with jamming-agnostic and jamming-aware SIM configurations during the training period.

Fig. 5: (a) SER and SR comparison with jamming-agnostic and jamming-aware SIM configurations across various SNR, with

K=4,N=100,L=6

; (b) SER performance under hardware imperfections.

Hetong Wang received the B.S. and M.S. degrees from the School of Telecommunications Engineering, Xidian University, Xi’an, China, in 2020 and 2023, respectively. She is currently pursuing the Ph.D. degree at the School of Information and Communication Engineering, Beijing University of Posts and Telecommunications (BUPT), Beijing, China. Her current research interests include physical layer security, reconfigurable intelligent surface, stacked intelligent metasurface, and machine learning. She has served as a reviewer for many high-impact IEEE journals, such as IEEE WCM, IoT, WCL, CL, OJCOMS.

Table 6

Yashuai Cao received the B.E. and Ph.D. degrees in communication engineering from Chongqing University of Posts and Telecommunications (CQUPT) and Beijing University of Posts and Telecommunications (BUPT), China, in 2017 and 2022, respectively. From 2022 to 2023, he was a lecturer in the Department of Electronics and Communication Engineering, North China Electric Power University (NCEPU), Baoding. From 2023 to 2025, he was a Postdoctoral Research Fellow with the Department of Electronic Engineering, Tsinghua University, Beijing, China. He is currently a Distinguished Associate Professor with the School of Intelligence Science and Technology, University of Science and Technology Beijing, Beijing, China. His research interests include Stacked Intelligent Metasurface, Environment-Aware Communications, and Channel Knowledge Map.

Table 7

Tiejun Lv received the M.S. and Ph.D. degrees in electronic engineering from the University of Electronic Science and Technology of China (UESTC), Chengdu, China, in 1997 and 2000, respectively. From January 2001 to January 2003, he was a Postdoctoral Fellow at Tsinghua University, Beijing, China. In 2005, he was promoted to Full Professor at the School of Information and Communication Engineering, Beijing University of Posts and Telecommunications (BUPT). From September 2008 to March 2009, he was a Visiting Professor with the Department of Electrical Engineering at Stanford University, Stanford, CA, USA. He is the author of four books, one book chapter, more than 160 published journal papers and 200 conference papers on the physical layer of wireless mobile communications. His current research interests include signal processing, communications theory and networking. He was the recipient of the Program for New Century Excellent Talents in University Award from the Ministry of Education, China, in 2006. He received the Nature Science Award from the Ministry of Education of China for the hierarchical cooperative communication theory and technologies in 2015 and the Shaanxi Higher Education Institutions Outstanding Scientific Research Achievement Award in 2025.

Table 8

Explore similar work

Over-The-Air Extreme Learning Machines with Nonlinear Stacked Intelligent Metasurfaces
Aug 27, 2026Kyriakos Stylianopoulos, Mattia Fabiani, Giulia Torcolacci +2Efficient Neural Network Inference Reconfigurable Intelligent Surfaces
Department of Informatics and Telecommunications, National and Kapodistrian University of Athens, Greece · DEI, University of Bologna, Italy · National Laboratory of Wireless Communications (WiLab), CNIT, Italy
Relay-Assisted Activation-Integrated SIM for Wireless Physical Neural Networks
Apr 5, 2026Meng Hua, Deniz GündüzReconfigurable Intelligent Surfaces
Department of Electrical and Electronic Engineering, Imperial College London, United Kingdom
Metasurfaces-Integrated Wireless Neural Networks for Lightweight Over-The-Air Edge Inference
Feb 22, 2026Kyriakos Stylianopoulos, Mario Edoardo Pandolfo, Paolo Di Lorenzo +1Wireless Communications Edge Inference
Department of Informatics and Telecommunications, National and Kapodistrian University of Athens, 16122 Athens, Greece · National Inter-University Consortium for Telecommunications (CNIT), Parma, Italy · DIAG Department, Sapienza University of Rome, via Ariosto 25, Rome, Italy +1

Aug 27, 2026J/K move · Enter open · S save

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

Kyriakos Stylianopoulos, Mattia Fabiani, Giulia Torcolacci, Davide Dardari+1

Department of Informatics and Telecommunications, National and Kapodistrian University of Athens, Greece · DEI, University of Bologna, Italy · National Laboratory of Wireless Communications (WiLab), CNIT, Italy

Efficient Neural Network Inference Reconfigurable Intelligent Surfaces

Open paper AskPDF arXiv

The recently envisioned goal-oriented communications paradigm requires machine learning inference to be performed directly on wirelessly transferred data. This paper presents an eXtremely Large (XL) Multiple-Input Multiple-Output (MIMO) system that operates as an Extreme Learning Machine (ELM) to execute Over-The-Air (OTA) binary classification. To reduce hardware complexity, the receiver is equipped with cascaded metasurfaces terminating in a single radio-frequency chain. A front metasurface layer applies a fixed nonlinear response to the incoming signal, acting as the ELM's activation function. Subsequent tunable linear metasurface layers physically approximate the trained network weights directly in the wave domain. Numerical evaluations across diverse datasets showcase that our XL MIMO architecture achieves classification accuracy comparable to idealized digital models, thereby proving the viability of low-complexity, wave-domain OTA learning.

Abstract

Explore similar work

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

Relay-Assisted Activation-Integrated SIM for Wireless Physical Neural Networks

Metasurfaces-Integrated Wireless Neural Networks for Lightweight Over-The-Air Edge Inference