AI Signal Processing Paradigm for Movable Antenna: From Spatial Position Optimization to Electromagnetic Reconfigurability

TL;DR

The paper tackles the inefficiency of fixed antennas in dynamic 6G scenarios by introducing MARA, a unified paradigm that combines SMA's spatial movement with ERA's electromagnetic reconfigurability. It develops a unified wideband mMIMO-OFDM channel model and SE optimization for SMA, ERA, and MARA, demonstrating that MARA yields substantial gains over traditional TFA and individual SMA/ERA schemes. A comprehensive AI-driven survey covers channel estimation/prediction, beamforming with port/mode selection, and ISAC, discussing method classes, trade-offs, and practical deployment considerations. The work further outlines future directions, including novel waveforms, embodied intelligence, and multi-stage beamforming with emerging architectures, highlighting AI as a core enabler for scalable, adaptive 6G MARA systems.

Abstract

As 6G wireless communication systems evolve toward intelligence, high reconfigurability, and space-air-ground integration \cite{liu2025toward, liu2024near}, the limitations of traditional fixed antenna (TFA) have become increasingly prominent. As a remedy, spatially movable antenna (SMA) and electromagnetically reconfigurable antenna (ERA) have respectively emerged as key technologies to break through this bottleneck. SMA activates spatial degree of freedom (DoF) by dynamically adjusting antenna positions, ERA regulates radiation characteristics using tunable metamaterials, thereby introducing DoF in the electromagnetic domain. However, the ``spatial-electromagnetic dual reconfiguration" paradigm formed by their integration poses severe challenges of high-dimensional hybrid optimization to signal processing. To address this issue, we integrate the spatial optimization of SMA and the electromagnetic reconfiguration of ERA, propose a unified modeling framework termed movable and reconfigurable antenna (MARA) and investigate the channel modeling and spectral efficiency (SE) optimization for MARA. Besides, we systematically review artificial intelligence (AI)-based solutions, focusing on analyzing the advantages of AI over traditional algorithms in solving high-dimensional non-convex optimization problems. This paper fills the gap in existing literature regarding the lack of a comprehensive review on the AI-driven signal processing paradigm under spatial-electromagnetic dual reconfiguration and provides theoretical guidance for the design and optimization of 6G wireless systems with advanced MARA.

Figures (12)

• Figure 1: Schematic of a SMA and its operation in a multipath channel.
• Figure 2: Two primary implementation methodologies for ERA.
• Figure 3: Application scenarios of MARA-aided wireless networks.
• Figure 4: Flowchart of the organization of this paper.
• Figure 5: Mechanical slide-based SMA communication prototype developed by Southeast University in fig1.