Movable Antenna Enabled Integrated Sensing and Communication

Wanting Lyu, Songjie Yang, Yue Xiu, Zhongpei Zhang, Chadi Assi, Chau YuenIEEE Trans. Wireless Communications 2025 · DOI

TL;DR

This paper introduces a novel approach to Integrated Sensing and Communication (ISAC) by enabling base station antennas to move. This "movable antenna" (MA) system allows for flexible beamforming, significantly improving both communication rates and sensing accuracy compared to traditional fixed antenna systems. The proposed algorithm optimizes both antenna positions and transmission signals to achieve these gains.

What Problem Does This Paper Address?

Next-generation wireless networks are moving towards Integrated Sensing and Communication (ISAC), where a single system handles both communication and sensing tasks. This integration promises better resource utilization and enhanced performance. However, current ISAC systems typically use fixed antennas, limiting their ability to adapt to dynamic wireless environments and fully exploit spatial diversity. This paper addresses this limitation by proposing an ISAC system where the base station's antennas are movable. This mobility offers an additional degree of freedom to reconfigure the wireless channel and improve both communication and sensing performance, especially in complex, cluttered environments. The core challenge lies in jointly optimizing the antenna positions and the transmission signals, which leads to a complex, non-convex optimization problem.

Methods

The paper proposes a flexible beamforming approach for an ISAC system with a movable antenna (MA) array at the base station (BS). The system aims to maximize a weighted sum of communication rate and sensing mutual information (MI). To tackle the non-convexity of the problem, they employ a fractional programming (FP) framework to transform the objective function. This transformed problem is then solved using an alternating optimization (AO) algorithm, which breaks it down into four sub-problems: updating the beamforming matrix (F), updating antenna positions (x), and updating auxiliary variables (μ, ξ, ξ).

  • Beamforming Matrix Update: This is solved by deriving closed-form expressions using Karush-Kuhn-Tucker (KKT) conditions, leveraging a Lagrangian dual decomposition method.
  • Antenna Position Update: This is the most novel part. Instead of a direct gradient ascent (DGA) which can converge to suboptimal solutions or violate constraints, they propose a 3-stage search-based projected gradient ascent (SPGA) method. This method includes:
  • Initial Point Search: Finding optimal initial antenna positions by discretizing the feasible region.
  • Gradient Ascent Updating: Iteratively updating antenna positions.
  • Feasible Region Projection: Projecting updated positions back into the feasible region to ensure constraints are met.
  • Auxiliary Variable Updates: These are updated using analytical solutions derived from the optimization sub-problems.

Datasets

Not discussed in the paper. The evaluation is based on numerical simulations with synthesized channel models.

Key Insights and Ideas

  • Movable Antennas Unlock New Potential: The ability for antennas to move provides a significant new degree of freedom, allowing the system to actively shape the wireless channel for improved performance in both communication and sensing.
  • SPGA Outperforms DGA for Antenna Positioning: The proposed SPGA method, with its initial search and projection stages, is crucial for overcoming the limitations of simpler gradient ascent methods, leading to better antenna configurations and overall performance.
  • Trade-offs are Manageable: The weighting factors (wc and ws) allow for flexible prioritization between communication and sensing, enabling the system to be tuned for specific application needs.
  • MA Benefits Extend Beyond Communication: While movable antennas have been studied for communication gains, this work demonstrates their significant impact on sensing performance, particularly in clutter suppression and accuracy.
  • Fewer MAs Can Outperform More FPAs: The flexibility of movable antennas means that a system with fewer MAs can achieve better performance than a system with more fixed antennas, offering a path to cost-effective solutions.

Key Results

  • The proposed SPGA-FP MA-enabled ISAC system achieved a 59.8% performance gain compared to conventional ISAC systems with fixed antennas (FPAs).
  • The SPGA method for antenna position optimization showed remarkable performance improvement compared to the DGA method.
  • In high SNR regions, the SPGA-FP flexible beamforming with 8 MAs achieved 18.5% and 37.5% performance gains compared to DGA and FPAs, respectively.
  • The system with 4 MAs and SPGA significantly outperformed 8 FPAs when the feasible moving region for antennas increased.
  • The MA system demonstrated remarkable clutter suppression in sensing beampattern analysis, leading to improved sensing MI.
  • Optimizing antenna positions with SPGA significantly reduced the Cramér-Rao Bound (CRB) for target angle estimation, indicating enhanced sensing accuracy.

Limitations and Future Work

  • Single Target Assumption: The current work focuses on a single sensing target. The authors note that the scheme can be extended to multi-target systems by treating other targets as interference, but this would add complexity.
  • Multi-path Sensing: The paper notes that MAs performed better in multi-path communication scenarios than in single-path sensing. Future work could explore how to leverage MAs for multi-path sensing.
  • Hardware Complexity: While MAs offer performance benefits, the mechanical components for antenna movement add hardware complexity and cost, which is a trade-off to consider.
  • Interference Management: For practical scenarios with multiple sensing targets, advanced techniques for interference suppression will be crucial to further improve system performance.
  • Exploration of Advanced Techniques: The authors suggest exploring advanced techniques for interference suppression to further enhance system performance in complex environments.

Summary updated

BibTeX
@article{lyu2025movable,
  author = {Wanting Lyu and Songjie Yang and Yue Xiu and Zhongpei Zhang and Chadi Assi and Chau Yuen},
  title = {{Movable Antenna Enabled Integrated Sensing and Communication}},
  journal = {IEEE Trans. Wireless Communications},
  year = {2025},
  doi = {10.1109/twc.2025.3525631},
}
Figure