Stacked Intelligent Metasurface-Aided Wave-Domain Signal Processing: From Communications to Sensing and Computing

TL;DR

This paper surveys stacked intelligent metasurfaces (SIM) as a platform for wave-domain neural networks that fuse AI, electromagnetic computing, and programmable metasurfaces. It presents the physical and computational foundations, surveys prototype demonstrations, and analyzes configuration and training strategies for both known and unknown transfer functions. The review covers SIM applications across wireless communications, sensing, and computing, including MIMO precoding, semantic encoding, object recognition, DOA estimation, pattern generation, and logical operations, with experimental evidence and performance comparisons. It concludes by highlighting challenges in channel estimation, calibration, and modeling, and outlines future research directions such as channel multiplexing, structural optimization, and neural-architecture design to unlock the full potential of SIMs in next-generation networks.

Abstract

Neural networks possess incredible capabilities for extracting abstract features from data. Electromagnetic computing harnesses wave propagation to execute computational operations. Metasurfaces, composed of subwavelength meta-atoms, are capable of engineering electromagnetic waves in unprecedented ways. What happens when combining these three cutting-edge technologies? This question has sparked a surge of interest in designing physical neural networks using stacked intelligent metasurface (SIM) technology, with the aim of implementing various computational tasks by directly processing electromagnetic waves. SIMs open up an exciting avenue toward high-speed, massively parallel, and low-power signal processing in the electromagnetic domain. This article provides a comprehensive overview of SIM technology, commencing with its evolutionary development. We subsequently examine its theoretical foundations and existing SIM prototypes in depth. Furthermore, the optimization/training strategies conceived to configure SIMs for achieving the desired functionalities are discussed from two different perspectives. Additionally, we explore the diverse applications of SIM technology across the communication, sensing, and computing domains, presenting experimental evidence that highlights its distinctive advantages in supporting multiple functions within a single device. Finally, we identify critical technical challenges that must be addressed to deploy SIMs in next-generation wireless networks and shed light on promising research directions to unlock their full potential.

Figures (19)

• Figure 1: The evolution timeline of the SIM technology.
• Figure 2: The organization of this paper.
• Figure 3: SIM is the amalgamation of neural networks, electromagnetic computing, and metasurfaces. The advantages of these three technologies complement each other.
• Figure 4: Two SIM configuration methods depending on whether the desired transfer function is known or not.
• Figure 5: Two SIM configuration methods depending on whether there is an accurate numerical model or not.