pith. sign in

arxiv: 2606.22897 · v1 · pith:ZN2GQ3YWnew · submitted 2026-06-22 · 📡 eess.SY · cs.SY

Adaptive Joint Beamforming and Fluid Antenna System Design for 6G ISAC

Haoyu Quan , Junhui Zhao , Dongming Wang This is my paper

Pith reviewed 2026-06-26 07:25 UTC · model grok-4.3

classification 📡 eess.SY cs.SY
keywords Fluid Antenna SystemsIntegrated Sensing and CommunicationSoft Actor-CriticBeamformingAntenna Reconfiguration6G WirelessReinforcement LearningMobile ISAC
0
0 comments X

The pith

A Soft Actor-Critic framework jointly optimizes fluid antenna positions and beamforming to deliver 4 ms latency and 42% communication gains in mobile ISAC.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper establishes that fluid antenna systems, which dynamically reposition elements, add spatial degrees of freedom that fixed antennas lack, enabling better adaptation in changing wireless channels for integrated sensing and communication. It shows that an end-to-end Soft Actor-Critic algorithm can solve the joint topology and beamforming problem in real time rather than through slow iterative methods. This matters because it reduces online inference to 4 ms while improving performance and cutting hardware needs. A sympathetic reader sees the potential for practical 6G systems that maintain high rates and sensing accuracy under mobility without massive antenna arrays.

Core claim

The central claim is that an end-to-end optimization framework based on the Soft Actor-Critic algorithm solves the joint fluid antenna topology reconfiguration and active beamforming problem for mobile ISAC systems, achieving an online inference latency of only 4 ms, improving communication performance by 42% over alternating optimization, and matching SCA-SDR benchmark performance while requiring 57% fewer antennas.

What carries the argument

End-to-end Soft Actor-Critic optimization that maps channel states directly to fluid antenna positions and beamforming vectors.

If this is right

  • Real-time joint decisions on antenna positions and beams become feasible at 4 ms inference time.
  • Communication rates rise 42% relative to alternating optimization methods under the same conditions.
  • Performance comparable to SCA-SDR is reached with 57% fewer antennas, improving hardware efficiency.
  • Dynamic reconfiguration of fluid antennas supplies extra spatial degrees of freedom for mobile ISAC.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same end-to-end learning structure could be applied to other reconfigurable hardware such as movable antennas or RIS phase profiles.
  • Hardware savings of 57% suggest lower power and cost budgets for handheld or vehicular ISAC nodes.
  • If channel models are extended to include blockage and Doppler, the framework may support higher-mobility scenarios.

Load-bearing premise

The simulation environment and channel models used to train and evaluate the SAC agent accurately capture the real-time dynamics and constraints of mobile ISAC systems with reconfigurable fluid antennas.

What would settle it

A measurement campaign on a physical mobile ISAC testbed with fluid antennas that shows the realized communication rate or sensing accuracy falling more than 20% below the simulated values at the reported antenna counts and latency.

Figures

Figures reproduced from arXiv: 2606.22897 by Dongming Wang, Haoyu Quan, Junhui Zhao.

Figure 1
Figure 1. Figure 1: Architecture of the FAS. As illustrated in [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 3
Figure 3. Figure 3: Communication rate versus the number of UEs [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figure 2
Figure 2. Figure 2: Convergence of the proposed scheme. C. Effectiveness Analysis To evaluate the effectiveness of the proposed scheme, several representative baselines are considered. A Uniform Planar Array (UPA) is adopted as the FPA baseline, while an SCA is employed as the baseline for FAS reconfiguration. For communication beamforming, MRT, Zero-Forcing (ZF), and SDR are adopted for comparison. Furthermore, to jointly ev… view at source ↗
Figure 5
Figure 5. Figure 5: Communication rate versus the target SNR threshold [PITH_FULL_IMAGE:figures/full_fig_p006_5.png] view at source ↗
read the original abstract

Fixed-Position Antennas (FPAs) are constrained by static physical topologies and struggle to adapt to rapidly varying wireless environments. By dynamically reconfiguring the antenna positions, Fluid Antenna Systems (FASs) introduce additional spatial Degrees of Freedom (DoF) for wireless optimization. This paper investigates the joint optimization of Fluid Antenna System (FAS) topology reconfiguration and active beamforming for mobile Integrated Sensing and Communication (ISAC) systems. To enable real-time decision making, an end-to-end optimization framework based on the Soft Actor-Critic (SAC) algorithm is proposed. Simulation results show that the proposed scheme achieves an online inference latency of only 4 ms. Compared to the widely used alternating optimization, it improves communication performance by 42%. Moreover, it achieves performance comparable to the SCA-SDR benchmark while requiring 57% fewer antennas, demonstrating superior hardware efficiency.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The manuscript proposes an end-to-end Soft Actor-Critic (SAC) reinforcement learning framework for the joint optimization of fluid antenna system (FAS) position reconfiguration and active beamforming in mobile integrated sensing and communication (ISAC) systems. It reports simulation results claiming an online inference latency of 4 ms, a 42% improvement in communication performance over alternating optimization, and performance comparable to an SCA-SDR benchmark while using 57% fewer antennas.

Significance. If the simulation results are shown to be robust, the work would demonstrate a practical path to low-latency adaptive ISAC by exploiting the additional spatial degrees of freedom from fluid antennas, potentially reducing hardware complexity while meeting both communication and sensing requirements in dynamic environments.

major comments (2)
  1. [Section IV] Section IV (Simulation Results): The headline metrics (4 ms latency, 42% rate gain, 57% antenna reduction) are presented without any description of the Monte Carlo trial count, error bars, statistical significance testing, exact parameter settings for the geometry-based stochastic channel model, or precise implementations of the alternating-optimization and SCA-SDR baselines. This absence prevents verification that the reported gains are supported by the data.
  2. [Section III] Section III (Proposed SAC Framework): The state vector (instantaneous CSI plus current FAS positions) and reward (weighted sum of achievable rate and sensing SNR) are defined, yet the paper provides no analysis of whether the underlying channel model incorporates Doppler spread due to user mobility or the mechanical repositioning latency of fluid antennas. Without a sensitivity study to model mismatch, the learned policy's claimed superiority remains conditional on simulation fidelity.
minor comments (1)
  1. [Abstract] The abstract states performance numbers but supplies no information on the simulation environment, making it impossible for readers to assess reproducibility from the outset.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive comments, which highlight important aspects for improving reproducibility and robustness. We address each point below and will incorporate the requested details and analyses in the revised manuscript.

read point-by-point responses

  1. Referee: [Section IV] Section IV (Simulation Results): The headline metrics (4 ms latency, 42% rate gain, 57% antenna reduction) are presented without any description of the Monte Carlo trial count, error bars, statistical significance testing, exact parameter settings for the geometry-based stochastic channel model, or precise implementations of the alternating-optimization and SCA-SDR baselines. This absence prevents verification that the reported gains are supported by the data.

    Authors: We agree that these implementation details are required for independent verification. In the revised manuscript we will augment Section IV with the Monte Carlo trial count (1000 independent realizations), error bars showing standard deviation, and t-test results confirming statistical significance (p < 0.01) of the reported gains. We will also tabulate the exact geometry-based stochastic channel parameters (path-loss exponents, shadowing variance, Doppler frequencies) and provide pseudocode plus solver settings (convergence tolerance, regularization) for both the alternating-optimization and SCA-SDR baselines. These additions will directly substantiate the headline metrics. revision: yes

  2. Referee: [Section III] Section III (Proposed SAC Framework): The state vector (instantaneous CSI plus current FAS positions) and reward (weighted sum of achievable rate and sensing SNR) are defined, yet the paper provides no analysis of whether the underlying channel model incorporates Doppler spread due to user mobility or the mechanical repositioning latency of fluid antennas. Without a sensitivity study to model mismatch, the learned policy's claimed superiority remains conditional on simulation fidelity.

    Authors: Section II already specifies that the geometry-based stochastic model incorporates time-varying coefficients with Doppler spread induced by user mobility. Mechanical repositioning latency, however, is not modeled; we implicitly assume reconfiguration completes within the channel coherence time. To address model-mismatch concerns we will add a dedicated sensitivity subsection that varies Doppler spread and introduces artificial repositioning delays, quantifying degradation in both rate and sensing SNR. This will demonstrate the operating regime in which the SAC policy retains its advantage. revision: yes

Circularity Check

0 steps flagged

No circularity: performance metrics are direct simulation outputs from standard SAC training, not reductions by construction

full rationale

The paper proposes an end-to-end SAC-based framework for joint FAS reconfiguration and beamforming. Reported metrics (4 ms latency, 42% rate gain, 57% antenna reduction) are obtained by running the trained policy in the same simulated environment used for training; no equations, parameter fits, or self-citations are shown that would make these outputs equivalent to the inputs by definition. The derivation chain consists of standard RL policy optimization followed by empirical evaluation, which remains self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

No free parameters, axioms, or invented entities are explicitly identifiable from the abstract alone; the work relies on standard reinforcement-learning and wireless-channel assumptions not detailed here.

pith-pipeline@v0.9.1-grok · 5681 in / 1103 out tokens · 34514 ms · 2026-06-26T07:25:53.887534+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

15 extracted references

  1. [1]

    Computation offloading and resource allocation for cloud assisted mobile edge computing in vehicular networks,

    J. Zhao, Q. Li, Y . Gong, and K. Zhang, “Computation offloading and resource allocation for cloud assisted mobile edge computing in vehicular networks,”IEEE Transactions on V ehicular Technology, vol. 68, no. 8, pp. 7944–7956, 2019

    2019

  2. [2]

    Fluid antenna-assisted ISAC systems,

    L. Zhou, J. Yao, M. Jin, T. Wu, and K.-K. Wong, “Fluid antenna-assisted ISAC systems,”IEEE Wireless Communications Letters, vol. 13, no. 12, pp. 3533–3537, 2024

    2024

  3. [3]

    MIMO capacity characterization for movable antenna systems,

    W. Ma, L. Zhu, and R. Zhang, “MIMO capacity characterization for movable antenna systems,”IEEE Transactions on Wireless Communica- tions, vol. 23, no. 4, pp. 3392–3407, 2024

    2024

  4. [4]

    Fluid antenna systems meet low-altitude wireless networks: Fundamentals, opportunities, and future directions,

    W. Liu, X. Zhang, C. Wang, J. Ren, W. Yuan, and C. You, “Fluid antenna systems meet low-altitude wireless networks: Fundamentals, opportunities, and future directions,”IEEE Internet of Things Magazine, vol. 9, no. 3, pp. 55–62, 2026

    2026

  5. [5]

    Beam alignment for MIMO fluid antenna systems,

    H. Jiang, Z. Wang, Y . Liu, A. Nallanathan, and H. Shin, “Beam alignment for MIMO fluid antenna systems,”IEEE Journal on Selected Areas in Communications, vol. 44, pp. 1193–1208, 2026

    2026

  6. [6]

    Antenna positioning and beamforming design for fluid antenna-assisted multi- user downlink communications,

    H. Qin, W. Chen, Z. Li, Q. Wu, N. Cheng, and F. Chen, “Antenna positioning and beamforming design for fluid antenna-assisted multi- user downlink communications,”IEEE Wireless Communications Let- ters, vol. 13, no. 4, pp. 1073–1077, 2024

    2024

  7. [7]

    Fluid-antenna enhanced integrated sensing and communication: Joint antenna positioning and beamforming design,

    T. Hao, C. Shi, Y . Guo, B. Xia, and F. Yang, “Fluid-antenna enhanced integrated sensing and communication: Joint antenna positioning and beamforming design,” in2024 IEEE/CIC International Conference on Communications in China (ICCC). IEEE, 2024, pp. 956–961

    2024

  8. [8]

    Joint beamforming and antenna design for near-field fluid antenna system,

    Y . Chen, M. Chen, H. Xu, Z. Yang, K.-K. Wong, and Z. Zhang, “Joint beamforming and antenna design for near-field fluid antenna system,” IEEE Wireless Communications Letters, vol. 14, no. 2, pp. 415–419, 2025

    2025

  9. [9]

    AI- empowered fluid antenna systems: Opportunities, challenges, and future directions,

    C. Wang, Z. Li, K.-K. Wong, R. Murch, C.-B. Chae, and S. Jin, “AI- empowered fluid antenna systems: Opportunities, challenges, and future directions,”IEEE Wireless Communications, vol. 31, no. 5, pp. 34–41, 2024

    2024

  10. [10]

    Federated learning assisted intelligent IoV mobile edge computing,

    H. Quan, Q. Zhang, and J. Zhao, “Federated learning assisted intelligent IoV mobile edge computing,”IEEE Transactions on Green Communi- cations and Networking, vol. 9, no. 1, pp. 228–241, 2025

    2025

  11. [11]

    Adaptive resource allocation for mobile edge computing in internet of vehicles: A deep reinforcement learning approach,

    J. Zhao, H. Quan, M. Xia, and D. Wang, “Adaptive resource allocation for mobile edge computing in internet of vehicles: A deep reinforcement learning approach,”IEEE Transactions on V ehicular Technology, vol. 73, no. 4, pp. 5834–5848, 2024

    2024

  12. [12]

    Soft actor-critic: Off- policy maximum entropy deep reinforcement learning with a stochastic actor,

    T. Haarnoja, A. Zhou, P. Abbeel, and S. Levine, “Soft actor-critic: Off- policy maximum entropy deep reinforcement learning with a stochastic actor,” inInternational conference on machine learning, vol. 80. PMLR, 2018, pp. 1861–1870

    2018

  13. [13]

    Edge caching and computation management for real-time internet of vehicles: An online and distributed approach,

    J. Zhao, X. Sun, Q. Li, and X. Ma, “Edge caching and computation management for real-time internet of vehicles: An online and distributed approach,”IEEE Transactions on Intelligent Transportation Systems, vol. 22, no. 4, pp. 2183–2197, 2021

    2021

  14. [14]

    Movable-antenna array empowered ISAC systems for low-altitude economy,

    Z. Kuang, W. Liu, C. Wang, Z. Jin, J. Ren, X. Zhang, and Y . Shen, “Movable-antenna array empowered ISAC systems for low-altitude economy,” in2024 IEEE/CIC International Conference on Communi- cations in China (ICCC Workshops). IEEE, 2024, pp. 776–781

    2024

  15. [15]

    Fluid antenna system-enabled integrated sensing, communication, and computing for intelligent IoV,

    H. Quan, J. Zhao, and J. Li, “Fluid antenna system-enabled integrated sensing, communication, and computing for intelligent IoV,”IEEE Transactions on V ehicular Technology, pp. 1–12, 2026

    2026