Pith. sign in

REVIEW 3 major objections 4 minor 11 references

Towards Intelligent Antenna Positioning: Leveraging DRL for FAS-Aided ISAC Systems

T0 review · 3 major / 4 minor · reviewed 2026-08-10 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read A DRL-BCD framework that alternates convex beamforming with DDPG-based antenna positioning achieves higher communication rates than fixed-position antennas in multi-target ISAC, and scales to more users and targets.

desk verdict A plausible but under-verified DRL recipe for FAS-ISAC; Eq. (12) as printed has sign-reversed penalties, so the central performance claims are not supported as written. read the letter →

arxiv 2501.01281 v1 pith:7HCOIWTS submitted 2025-01-02 eess.SP

Shunxing Yang , Junteng Yao , Jie Tang , Tuo Wu , Maged Elkashlan , Chau Yuen , Merouane Debbah , Hyundong Shin
show 1 more author
Matthew Valenti
This is my paper · ORCID
classification eess.SP
keywords fluidantennasystemsintegratedsensingandcommunicationdeepreinforcementlearningDDPGblockcoordinatedescentbeamformingpositioningmulti-target
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper tries to establish that a dual-function base station with fluid antennas can serve one communication user and sense multiple targets better if antenna positions are optimized by a deep-reinforcement-learning policy rather than fixed. The authors propose a block coordinate descent framework that alternates a convex optimization step for the transmit covariance matrix with a deep deterministic policy gradient (DDPG) step that moves the fluid antennas continuously inside a small region. In simulation, the resulting communication rates exceed those of a fixed-position antenna baseline even when the baseline serves only one target, and the approach extends to multiple users and targets. The intended payoff is a practical, real-time method for balancing sensing and communication in ISAC systems without resorting to intractable joint optimization.

What carries the argument

The load-bearing object is the BCD decomposition of the original problem: for fixed antenna positions, the beamforming subproblem is convex in the transmit covariance matrix $U$ and is solved with a convex solver after relaxing the rank-one constraint, with Gaussian randomization restoring a rank-one solution; for fixed $U$, the position subproblem is recast as a Markov decision process whose state is the antenna coordinate vector plus three beamforming features (trace, largest eigenvalue, and mean eigenvalue of $U$), whose action is incremental position updates bounded within $\pm A/2$ for $A = 4\lambda$, and whose reward is the communication rate minus penalties for sensing-gain violations, power violations, and movement distance. The DDPG actor-critic, trained with Ornstein-Uhlenbeck exploration noise, learns a deterministic continuous positioning policy. This machinery converts an intractable joint optimization into alternating convex solves and policy-gradient updates that can execute in real time once trained.

What would settle it

Inspect the converged DDPG policy on a fixed set of channel realizations at a given SNR: record the achieved sensing gain $\varpi(p^{(k)})$ for each target $k$. If a non-negligible fraction of targets fall below the threshold $\Gamma$ while the reported communication rate remains high, then the reported rate is partly an artifact of the reward penalty, not a genuine sensing-communication trade-off. The same check applies to the rank-one reconstructed beamformer from Gaussian randomization, which may violate the sensing constraint (9f) even when the relaxed convex solution satisfies it.

Watch

Extended reading notes

Core claim

The central claim is that the joint non-convex problem of beamforming and antenna placement in a FAS-aided ISAC system can be decomposed into two alternating blocks: a convex transmit-covariance optimization, solved under a relaxed rank-one constraint, and an antenna-position Markov decision process solved by DDPG. Under a far-field channel model, the paper reports that the proposed DRL-BCD algorithm achieves higher average communication rates than a fixed-position antenna baseline across SNR values, with the gap persisting when multiple sensing targets are added. The paper also reports that adding more users increases maximum rates (for example, 27.6% higher at 30 dB with 12 antennas compared with a single user), and that the DRL policy generalizes to these larger setups.

Load-bearing premise

The paper's central claim depends on the untested assumption that the penalty term in the reward function really forces the DDPG policy to meet the required sensing strength $\varpi(p^{(k)}) \ge \Gamma$ for every target; the paper never reports the achieved sensing strengths or the penalty weights $\alpha_1, \alpha_2, \alpha_3$ and threshold $\Gamma$ it used.

Editorial extensions

If this is right

  • In multi-target ISAC, communication rates improve versus fixed-position antennas even when the sensing workload grows, so antenna mobility is a degree of freedom worth exploiting.
  • The BCD-DRL framework scales to more users and targets without redesigning the algorithm, because the DDPG state and action dimensions grow linearly.
  • After training, antenna positioning decisions can be made in real time, unlike iterative alternating optimization, making the approach usable in fast-changing channel conditions.
  • Continuous antenna repositioning outperforms discrete port selection, suggesting that fine-grained placement freedom is valuable.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A natural next step is to verify sensing constraint satisfaction explicitly; if the penalty weights are large enough, the same reward-shaping recipe could serve as a generic solver for other continuous-placement multi-objective wireless problems.
  • The far-field constant-AoA assumption holds only for small movement regions; testing with near-field or environment-dependent path parameters would reveal how much of the gain survives realistic propagation.
  • Replacing the hand-tuned penalty weights with a constrained reinforcement learning update (for example a Lagrangian method) would remove the need to choose $\alpha_1,\alpha_2,\alpha_3$ and would make the claimed sensing-communication balance more robust.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 4 minor

Summary. The paper considers an integrated sensing and communication (ISAC) system in which a dual-functional base station equipped with fluid antennas (FAs) communicates with a user terminal and senses multiple radar targets. It formulates a joint transmit-covariance and antenna-position optimization problem (Problem (9)) and proposes a block coordinate descent (BCD) framework: for fixed antenna positions, the transmit covariance matrix U is obtained by solving a convex optimization problem (10) with CVX and, if needed, Gaussian randomization; for fixed U, the antenna positions are optimized by a deep deterministic policy gradient (DDPG) agent whose state includes positions and beamforming features, whose action is incremental antenna displacements, and whose reward is given by Eq. (12). Numerical results in Figs. 2 and 3 claim higher average and maximum communication rates than a fixed-position-antenna (FPA) baseline, and the authors further claim scalability to more users and targets. The central claim is that continuous FAS repositioning via DRL provides a practical way to balance communication and sensing in multi-target ISAC.

Significance. The topic is timely: FAS and ISAC are both active 6G research areas, and the specific problem of jointly optimizing beamforming and continuous antenna positions for multi-target sensing is not yet thoroughly explored. The convex beamforming subproblem is formulated and solved in a standard, correct manner, and the use of DDPG to handle the non-convex position subproblem is a plausible design choice that, if validated, could be useful for practical deployment. However, the evidence presented in the manuscript is currently insufficient to support the central claim. The reward function printed in Eq. (12) has penalty terms whose signs are reversed relative to the constraints, no sensing-gain or power-consumption values are reported to verify feasibility of the operating points, and the claimed scalability to multiple users is not backed by a multi-user formulation. These issues are fixable, but the manuscript as written does not substantiate the claimed communication-sensing balance.

major comments (3)
  1. [Section III-B1c, Eq. (12)] The reward function as printed is sign-reversed with respect to the constraints of Problem (9). The first penalty term is −α1 Σ_m max(0, ϖ(p^(m)) − Γ), which is zero when the sensing gain violates the constraint ϖ(p^(m)) ≥ Γ and positive when the constraint is satisfied with margin; the second penalty term is −α2 max(0, Pmax − Tr(U)), which is zero when the power constraint Tr(U) ≤ Pmax is violated and positive when it is satisfied. Thus the penalties vanish exactly on constraint violation, so an agent maximizing this reward is not optimizing the constrained problem (9) and may be encouraged to violate the sensing and power constraints. This contradicts the paper's own description in Section III-B1c, which states that the second term 'penalizes violations of the required sensing gain constraint.' If the implemented reward used the opposite signs, the manuscript must state this explicitly and correct Eq. (12); if the implementation matches the printed formula, the rate gains over FPA in Figs. 2 and 3 cannot be interpreted as evidence of a valid constrained solution.
  2. [Section IV, Tables/Figs. 2 and 3] The manuscript never reports the values of the reward weights α1, α2, α3, the sensing gain threshold Γ, or any achieved sensing gain ϖ(p^(k)) or transmit power Tr(U) at the reported operating points. These quantities are essential because the claimed 'balance' between communication and sensing rests on the assertion that constraint (9f) and (9e) are actually satisfied. Without verifying that the DRL policy meets the sensing-gain requirements, the higher average rates in Figs. 2 and 3 may simply reflect the agent's freedom to violate the sensing and power constraints. Please add a table or figure showing the achieved sensing gains and transmit powers (or their average/feasibility fractions) for the reported configurations, together with the exact values of α1, α2, α3, and Γ used in training.
  3. [Section II and Section III-B, multi-user claim] The system model, channel model, and optimization problem (9) are all formulated for a single user terminal (M = 1). Yet Eq. (12) contains a sum over m = 1, ..., M without defining M, and Fig. 3 reports results for M = 3 user terminals. The multi-user extension is mentioned only in a footnote and in the discussion of Fig. 3; no multi-user channel vector, rate expression, beamforming constraints, or problem formulation is provided. Consequently, the paper's scalability claim for 'additional users' is not mathematically supported. Either provide the full multi-user formulation and explain how the BCD-DRL algorithm is adapted to it, or restrict the scalability claims to the number of targets K rather than the number of users.
minor comments (4)
  1. [Section IV, DDPG hyperparameters] The paper gives some network and training parameters (buffer size 10000, mini-batch 64, target update τ = 0.001) but omits other essential DDPG hyperparameters, including the learning rates for actor and critic, the discount factor, the Ornstein-Uhlenbeck coefficients ξ and ς, and the number of training episodes. Please report the hyperparameter setting to make the results reproducible.
  2. [Figs. 2 and 3] The quantities 'Average Communication Rate' and 'Maximum Communication Rate' are not precisely defined; it is unclear whether the average is over random channel realizations, training seeds, or both, and what the maximum is taken over. Please define these metrics and include error bars or confidence intervals across multiple random seeds, as DRL training is sensitive to seed variability.
  3. [Section III-B1c, Eq. (12)] There is an unbalanced parenthesis in the first max term of Eq. (12): 'max(0, ϖ(p^(m)) − Γ' is missing a closing parenthesis. Also, the variable M is used in Eq. (12) but is not defined in the problem formulation; if it denotes the number of user terminals, a definition should be given before first use.
  4. [Throughout] There are several typographical errors that should be corrected, including 'receiever' in Section II-B, 'constrants' in Section III-B1c, and 'trade-sff' in the title of reference [9]. Additionally, the paper only compares against an FPA baseline; a comparison with at least one other movable-antenna or FAS position-optimization baseline (e.g., the iterative method of [8] or a random search) would strengthen the claim of superiority.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular derivation: the DRL-BCD results are direct simulation outputs against a fixed-antenna baseline, with no fitted parameter masquerading as a prediction.

full rationale

The paper's claimed contribution is algorithmic: a BCD loop in which the transmit covariance U is updated by CVX for fixed antenna positions, and the positions are updated by a DDPG policy trained on the reward in Eq. (12). The reported rate curves in Figs. 2-3 are simulated evaluations of that loop, not quantities derived from the model's inputs by construction. The reward includes the communication rate R, so the agent is explicitly trained to maximize a proxy containing R; this is standard RL-based optimization and does not make the later report of R circular, because the meaningful comparison is to the FPA baseline under the same channel model, which is an independent fixed-position benchmark not encoded in the training reward. No fitted parameter is renamed as a prediction: the only learned quantities are DDPG network weights and the CVX-optimized U, both used as optimization variables. Self-citations [3], [4], and [8] provide background and a comparison baseline, not an imported uniqueness theorem or ansatz, so they are not load-bearing. The principal evidentiary weakness is not circularity: Eq. (12) as printed has max(0, varphi - Gamma) and max(0, Pmax - Tr(U)) terms that are positive when constraints are satisfied and zero when violated, and the paper does not report alpha1, alpha2, alpha3, Gamma, or achieved sensing gains. That is a gap in support for the claim that the reported operating points satisfy (9f), but it is a correctness or validation issue rather than a reduction of a prediction to its inputs.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

The central claim rests on a synthetic channel model, an unverified reward-shaping design, and undisclosed RL hyperparameters. There are three groups of free parameters and four stated assumptions. No new physical entities are introduced.

free parameters (3)
  • Reward weights alpha1, alpha2, alpha3 = not specified
    The reward in Eq. (12) uses these weights to balance sensing gain penalty, power penalty, and movement penalty. They directly determine the found antenna positions and the communication-sensing trade-off, but the paper does not give their values.
  • Sensing gain threshold Gamma = not specified
    The problem (9f) requires detection gain >= Gamma for every target, and the same Gamma appears in the reward penalty. Its simulation value is never stated, so the strictness of the sensing requirement is unknown.
  • DDPG hyperparameters (learning rates, discount factor, OU coefficients xi and sigma) = not specified
    The paper gives buffer size (10000), batch size (64), target update rate tau=0.001, and network sizes, but omits actor/critic learning rates, discount factor, and OU noise parameters, all of which control convergence and solution quality.
assumptions (4)
  • domain assumption Far-field channel model with constant AoAs/AoDs across the antenna movement region.
    Invoked in Section II-B to justify the response vectors in Eqs. (2)-(6). If the movement region is not small relative to distance, the constant-angle assumption breaks down.
  • domain assumption The path response matrix Sigma is diagonal with specified Rician factor tau=1 distribution.
    Section IV assumes Sigma[1,1] ~ CN(0, tau/(tau+1)) and zero off-diagonal entries. This is a synthetic channel model chosen for simulation; the algorithm's reported gains are conditioned on it.
  • standard math Gaussian randomization recovers a near-optimal rank-one solution from the relaxed SDR solution.
    Section III-A uses Gaussian randomization after relaxing the rank-one constraint (9g); the paper does not analyze the approximation ratio.
  • ad hoc to paper The reward shaping in Eq. (12) is an adequate surrogate for the constrained problem (9).
    The penalties with weights alpha1, alpha2, alpha3 replace hard constraints. This is a design choice, not derived from the optimization problem, and the paper provides no evidence that the penalized optimum meets the original constraints.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Towards Intelligent Antenna Positioning: Leveraging DRL for FAS-Aided ISAC Systems." pith.science (2026). https://pith.science/paper/7HCOIWTS

@misc{pith2026250101281,
  author       = {Pith},
  title        = {Pith review of: Towards Intelligent Antenna Positioning: Leveraging DRL for FAS-Aided ISAC Systems},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/7HCOIWTS}},
  note         = {Machine review of arXiv:2501.01281}
}
read the original abstract

Fluid antenna systems (FAS) enable dynamic antenna positioning, offering new opportunities to enhance integrated sensing and communication (ISAC) performance. However, existing studies primarily focus on communication enhancement or single-target sensing, leaving multi-target scenarios underexplored. Additionally, the joint optimization of beamforming and antenna positions poses a highly non-convex problem, with traditional methods becoming impractical as the number of fluid antennas increases. To address these challenges, this letter proposes a block coordinate descent (BCD) framework integrated with a deep reinforcement learning (DRL)-based approach for intelligent antenna positioning. By leveraging the deep deterministic policy gradient (DDPG) algorithm, the proposed framework efficiently balances sensing and communication performance. Simulation results demonstrate the scalability and effectiveness of the proposed approach.

Figures

Figures reproduced from arXiv: 2501.01281 by the authors.

Figure 1
Figure 1. The framework of DDPG. by α1, penalizes violations of the required sensing gain constraint, ensuring that the sensing gain ϖ(p (m) ) for each target m remains greater than Γ. The third term, with weight α2, enforces the total power constraint by penalizing any excess power consumption above the threshold Pmax. The final term, weighted by α3, introduces a movement penalty to reach the objective that is proportional t… view at source ↗
Figure 2
Figure 2. Average Communication rate [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Maximum Communication rate. θ ′ ← τθ + (1 − τ )θ ′ , (14) where τ = 0.001 ensures stable learning. Besides, a buffer of size 10000 stores transitions (st, at, rt, st+1), from which mini-batches of size 64 are sampled for network updates. IV. NUMERICAL RESULTS In our simulation experiments, we consider that the eleva￾tion and azimuth angles θ (t) k , ψ(t) k , θ(r) i , ψ(r) i are all indepen￾dent and identically distr… view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

11 extracted references · 7 canonical work pages

  1. [1]

    A tutorial on fluid antenna system for 6G networks: Encompassing communication theory, optimization methods and hardware designs,

    W. K. New et al. , “A tutorial on fluid antenna system for 6G networks: Encompassing communication theory, optimization methods and hardware designs,” IEEE Commun. Surv. & Tut. , early access, doi:10.1109/COMST.2024.3498855, 2024

    arXiv 2024

  2. [2]

    Integrated sensing and communications: Recent advances and ten open challenges,

    S. Lu et al. , “Integrated sensing and communications: Recent advances and ten open challenges,” IEEE Internet Things J. , vol. 11, no. 11, pp. 19094–19120, Jun. 2024

    work page 2024

  3. [3]

    Approximate capacity-distortion region of joint state sensing and communication in MIMO real Gaussian channels,

    J. Yao et al. ,“Approximate capacity-distortion region of joint state sensing and communication in MIMO real Gaussian channels,” IEEE Trans. Commun., vol. 72, no. 5, pp. 2625–2638, May 2024

    work page 2024

  4. [4]

    Wu, et al

    T. Wu, et al. . ”Fluid antenna systems enabling 6G: Principles, applica- tions, and research directions.” arXiv preprint, arXiv:2412.03839, 2024

    arXiv 2024

  5. [5]

    Fluid antenna systems,

    K.-K. Wong, et al. . “Fluid antenna systems,” IEEE Trans. Wireless Commun., vol. 20, no. 3, pp. 1950–1962, Mar. 2021

    work page 1950

  6. [6]

    MIMO capacity characterization for movable antenna systems,

    W. Ma, L. Zhu, and R. Zhang, “MIMO capacity characterization for movable antenna systems,” IEEE Trans. Wireless Commun. , vol. 23, no. 4, pp. 3392–3407, Apr. 2023

    work page 2023

  7. [7]

    Movable-antenna position optimization: A graph-based approach,

    W. Mei, et al. .“Movable-antenna position optimization: A graph-based approach,” IEEE Wireless Commun. Lett. , vol. 13, no. 7, pp. 1853–1857, Jul. 2024

    work page 2024

  8. [8]

    Fluid antenna-assisted ISAC systems,

    L. Zhou et al. , “Fluid antenna-assisted ISAC systems,” IEEE Wireless Commun. Lett. , vol. 13, no. 12, pp. 3533–3537, Dec. 2024

    2024

Show all 11 references
  1. [9]

    Shifting the ISAC trade-sff with fluid antenna systems,

    J. Zou et al. , “Shifting the ISAC trade-sff with fluid antenna systems,” arXiv preprint, arXiv:2405.05715v1, 2024

    2024 arXiv

  2. [10]

    Fluid antenna system liberating multiuser MIMO for ISAC via deep reinforcement learning,

    C. Wang et al. , “Fluid antenna system liberating multiuser MIMO for ISAC via deep reinforcement learning,” IEEE Trans. Wireless Commun., early access, doi:10.1109/TWC.2024.3376800, 2024

    2024

  3. [11]

    CVX: Matlab software for disciplined convex programming, version 2.1,

    G. Michael et al. , “CVX: Matlab software for disciplined convex programming, version 2.1,” 9 Mar. 2014

    2014

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.