Pith. sign in

REVIEW 3 major objections 2 minor 19 references

A bi-level DRL framework positions movable antennas and designs symbol-level waveforms to raise the minimum radar SINR in DFRC systems.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

T0 review · grok-4.3

2026-06-29 11:03 UTC pith:PB6YLZX4

load-bearing objection Incremental bi-level TD3-plus-CCP setup for MA placement and symbol-level waveforms in DFRC, but the simulation gains rest on unexamined TD3 behavior in non-convex space. the 3 major comments →

arxiv 2605.27839 v1 pith:PB6YLZX4 submitted 2026-05-27 eess.SP

Movable Antenna Enhanced Dual-Functional Radar-Communication: A Symbol-Level Precoding Approach

Ran Yang , Ning Wei , Zheng Dong , Chadi Assi , You Li , Fei Xu , Yue Xiu This is my paper
classification eess.SP
keywords movable antennadual-functional radar-communicationsymbol-level precodingdeep reinforcement learningradar SINRbi-level optimization
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

The paper sets out to maximize the lowest radar signal-to-interference-plus-noise ratio across multiple targets by jointly choosing antenna locations, transmitted waveforms, and receive filters in a movable-antenna DFRC setup. Direct optimization is blocked by waveform constraints and the nonlinear dependence of channel coefficients on antenna coordinates. A two-layer method therefore uses the TD3 reinforcement-learning algorithm to search for good placements while an inner loop applies penalty convex-concave and majorization-minimization steps to produce feasible waveforms. Simulations indicate that the resulting placements deliver higher radar SINR and a more favorable sensing-communication balance than fixed-antenna baselines.

Core claim

The bi-level optimization framework, with the twin delayed deep deterministic policy gradient algorithm in the outer loop selecting antenna positions and penalty convex-concave procedure together with majorization-minimization in the inner loop regularizing the symbol-level precoder and filters, yields improved minimum radar SINR values and a superior sensing-communication trade-off relative to benchmark schemes in cluttered environments.

What carries the argument

Bi-level optimization in which TD3 searches antenna positions to maximize min radar SINR while an inner CCP/MM loop produces compliant space-time waveforms and receive filters, handling the nonlinear position-to-channel mapping.

Load-bearing premise

The TD3 algorithm can locate antenna positions that meaningfully raise the minimum radar SINR without converging to placements that leave the objective unimproved.

What would settle it

A set of Monte-Carlo trials in which randomly chosen antenna positions produce equal or higher min radar SINR than the TD3-derived positions would show that the placement optimization step adds no value.

Watch this falsifier — get emailed when new claim-graph text bears on it.

If this is right

  • The minimum radar SINR across targets increases relative to fixed-antenna designs.
  • The sensing-communication performance trade-off curve lies above those of the benchmark schemes.
  • The approach supplies a tractable surrogate for an otherwise non-convex joint placement-and-waveform problem.
  • The non-linear mapping from positions to channels is navigated sufficiently well for the reported SINR gains to appear.

Where Pith is reading between the lines

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

  • The same outer-loop placement search could be paired with other inner-loop solvers if the waveform constraints change.
  • Real-time re-optimization of antenna positions would become feasible once the computational cost of the inner CCP/MM iterations is reduced.
  • Extension to scenarios with moving targets would require only that the channel model inside the inner loop be updated at each time step.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 2 minor

Summary. The paper proposes a bi-level optimization framework for symbol-level precoding in movable-antenna enhanced dual-functional radar-communication (DFRC) systems. The outer loop uses the twin delayed deep deterministic policy gradient (TD3) algorithm to optimize antenna positions while the inner loop applies penalty convex-concave procedure (CCP) and majorization-minimization (MM) to design space-time waveforms and receive filters, with the goal of maximizing the minimum radar SINR across multiple targets in clutter. Simulations are reported to show improved radar SINR and a better sensing-communication trade-off relative to benchmark schemes.

Significance. If the TD3 outer loop reliably identifies antenna placements that improve the min-SINR objective beyond fixed-position baselines, the framework would offer a practical heuristic for a non-convex joint design problem that arises in MA-DFRC. The combination of DRL with established inner-loop convexification techniques is a reasonable engineering approach, but the absence of convergence analysis or robustness checks on the reported gains limits the result's immediate theoretical or practical impact.

major comments (3)
  1. [Abstract] Abstract and simulation results section: the headline claim that the proposed method 'significantly improves radar SINR' rests on TD3 successfully navigating the non-linear position-to-channel mapping in a multi-target cluttered environment, yet no convergence guarantees, ablation studies on TD3 hyperparameters, random seeds, or comparisons against exhaustive/convex-relaxation position search are provided; without these, the reported superiority could be an artifact of initialization rather than a robust property of the bi-level scheme.
  2. [Proposed Method] Problem formulation and proposed method sections: the optimization problem is stated to be intractable due to waveform constraints and the non-linear antenna-position mapping, but the manuscript supplies no analysis of how the TD3 actor-critic updates interact with the inner CCP/MM loop or whether the composite objective remains stable under realistic clutter models; this directly affects whether the claimed min-SINR gains are load-bearing or merely simulation-specific.
  3. [Simulation Results] Simulation results: no error bars, multiple independent runs, or explicit baseline definitions (e.g., fixed-position arrays, random MA placement, or convex-relaxation alternatives) are described, making it impossible to assess whether the reported sensing-communication trade-off improvements exceed statistical variation or post-hoc tuning effects.
minor comments (2)
  1. [System Model] Notation for the channel coefficients as functions of antenna positions should be introduced earlier and kept consistent between the system model and the TD3 state representation.
  2. [Abstract] The abstract mentions 'practical waveform constraints' but does not list them explicitly; a short enumerated list would improve readability.

Simulated Author's Rebuttal

3 responses · 1 unresolved

We thank the referee for the constructive feedback highlighting the need for stronger empirical validation and clearer presentation of the bi-level framework. We will revise the manuscript to incorporate multiple runs, error bars, explicit baseline definitions, and additional discussion on empirical stability. However, theoretical convergence analysis for the TD3 outer loop remains outside the scope of this letter.

read point-by-point responses

  1. Referee: [Abstract] Abstract and simulation results section: the headline claim that the proposed method 'significantly improves radar SINR' rests on TD3 successfully navigating the non-linear position-to-channel mapping in a multi-target cluttered environment, yet no convergence guarantees, ablation studies on TD3 hyperparameters, random seeds, or comparisons against exhaustive/convex-relaxation position search are provided; without these, the reported superiority could be an artifact of initialization rather than a robust property of the bi-level scheme.

    Authors: We agree that robustness evidence can be strengthened. In revision we will add ablation results on TD3 hyperparameters (actor/critic learning rates and exploration noise) and report performance averaged over 5 independent random seeds with different initializations to show consistency of the min-SINR gains. We will also note that exhaustive search over continuous positions is computationally prohibitive and that no convex relaxation for the position subproblem is currently available. Theoretical convergence guarantees for TD3 in this setting are not provided, as they are generally unavailable for DRL heuristics and lie beyond the letter's scope. revision: partial

  2. Referee: [Proposed Method] Problem formulation and proposed method sections: the optimization problem is stated to be intractable due to waveform constraints and the non-linear antenna-position mapping, but the manuscript supplies no analysis of how the TD3 actor-critic updates interact with the inner CCP/MM loop or whether the composite objective remains stable under realistic clutter models; this directly affects whether the claimed min-SINR gains are load-bearing or merely simulation-specific.

    Authors: The outer TD3 treats the inner CCP/MM solver as a black-box reward evaluator that returns the achieved min-SINR for each candidate position vector. We will insert a short paragraph describing the observed training behavior, including that reward curves remain stable across the simulated clutter realizations without divergence. A rigorous analysis of the composite dynamics is intractable because the inner loop is non-differentiable; such analysis is left for future work. The clutter model follows the standard point-target-plus-clutter formulation used in prior DFRC literature. revision: partial

  3. Referee: [Simulation Results] Simulation results: no error bars, multiple independent runs, or explicit baseline definitions (e.g., fixed-position arrays, random MA placement, or convex-relaxation alternatives) are described, making it impossible to assess whether the reported sensing-communication trade-off improvements exceed statistical variation or post-hoc tuning effects.

    Authors: We will expand the simulation section to (i) explicitly list all baselines (fixed-position ULA, random MA placement within the feasible region, and a non-MA symbol-level precoding benchmark), (ii) average all curves over 10 independent runs that vary both TD3 random seeds and channel realizations, and (iii) include error bars showing one standard deviation. These additions will allow readers to judge whether the reported trade-off gains exceed statistical variation. revision: yes

standing simulated objections not resolved
  • Theoretical convergence guarantees or a complete stability analysis of the TD3 actor-critic updates interacting with the non-differentiable inner CCP/MM loop under general clutter models

Circularity Check

0 steps flagged

No circularity: bi-level optimization framework is independent of simulation outcomes

full rationale

The paper formulates a bi-level optimization problem to maximize minimum radar SINR by jointly designing waveforms, filters, and MA positions, then solves the intractable non-linear problem via an outer TD3 layer for placement and inner CCP/MM layers for waveforms. This structure is presented as a direct algorithmic response to the stated intractability, with no equations or claims reducing the SINR objective or its solution to a fitted parameter, self-citation chain, or renamed input by construction. Simulation results are reported as empirical outcomes of applying this framework, not as derivations that presuppose the claimed gains. The derivation chain therefore remains self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim rests on the domain assumption that the joint optimization is intractable and that DRL plus convex approximations can produce practically useful solutions; no free parameters or invented entities are explicitly introduced in the abstract.

axioms (1)
  • domain assumption The optimization problem is intractable due to practical waveform constraints and the non-linear mapping from antenna positions to channel coefficients.
    Directly stated in the abstract as the motivation for the bi-level framework.

pith-pipeline@v0.9.1-grok · 5730 in / 1215 out tokens · 42603 ms · 2026-06-29T11:03:51.022402+00:00 · methodology

0 comments
read the original abstract

This letter investigates a symbol-level precoder design for movable antenna (MA)-enhanced dual-functional radar-communication (DFRC) systems. To enhance radar sensing capabilities, we formulate an optimization problem aimed at maximizing the minimum radar signal-to-interference-plus-noise ratio (SINR) across multiple targets in a cluttered environment. Our approach jointly designs the space-time transmitted waveforms, receiving filters, and antenna placement. However, the resulting problem is intractable to solve due to practical waveform constraints and the non-linear mapping from antenna positions to the corresponding channel coefficients. To address these challenges, we develop a bi-level optimization framework by leveraging deep reinforcement learning (DRL). Specifically, the twin delayed deep deterministic policy gradient (TD3) algorithm is employed in the outer layer to optimize antenna placement, while penalty convex-concave procedure (CCP) and majorization-minimization (MM) techniques are incorporated in the inner layer for regularizing waveform design. Simulation results demonstrate that the proposed method significantly improves radar SINR and achieves a superior sensing-communication trade-off compared to benchmark schemes.

Figures

Figures reproduced from arXiv: 2605.27839 by Chadi Assi, Fei Xu, Ning Wei, Ran Yang, You Li, Yue Xiu, Zheng Dong.

Figure 1
Figure 1. Figure 1: (a) Convergence behaviour of the proposed algorithm [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: (a) The minimum radar SINR versus QoS γ. (b) The minimum radar SINR versus normalized moving region. spaced between intervals of λ/2, Algorithm 2 is directly employed to optimize transmission waveform x; 2) BLP￾MA: The BLP and transceiver antenna positions are jointly optimized to enhance the radar performance [13]; 3) BLP￾FPA: The BLP method is performed with FPAs to maximize the radar SINR performance. I… view at source ↗

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

19 extracted references · 1 canonical work pages

  1. [1]

    A vision of 6G wireless sy stems: Applications, trends, technologies, and open research pro blems,

    W. Saad, M. Bennis, and M. Chen, “A vision of 6G wireless sy stems: Applications, trends, technologies, and open research pro blems,” IEEE Network, vol. 34, no. 3, pp. 134–142, 2020

    2020

  2. [2]

    A survey of recent advances in optimization methods for wireless communications,

    Y .-F. Liu et al. , “A survey of recent advances in optimization methods for wireless communications,” IEEE J. Sel. Areas Commun. , vol. 42, no. 11, pp. 2992–3031, 2024

    2024

  3. [3]

    Enabling joint communication and radar sensing in mobile networks—a survey,

    J. A. Zhang et al. , “Enabling joint communication and radar sensing in mobile networks—a survey,” IEEE Commun. Surv. Tutorials , vol. 24, no. 1, pp. 306–345, 2022

    2022

  4. [4]

    A tutorial on interference exploitation via symbol- level precoding: Overview, state-of-the-art and future di rections,

    A. Li et al. , “A tutorial on interference exploitation via symbol- level precoding: Overview, state-of-the-art and future di rections,” IEEE Commun. Surv. Tutorials , vol. 22, no. 2, pp. 796–839, 2020

    2020

  5. [5]

    Joint transmit waveform and passive beamforming design for RIS-aided DFRC systems,

    R. Liu et al., “Joint transmit waveform and passive beamforming design for RIS-aided DFRC systems,” IEEE J. Sel. Topics Signal Process. , vol. 16, no. 5, pp. 995–1010, 2022

    2022

  6. [6]

    SLP-based dual-functional waveform design for ISAC systems: A deep learning approach,

    P . Jiang et al. , “SLP-based dual-functional waveform design for ISAC systems: A deep learning approach,” IEEE Trans. V eh. Technol., vol. 74, no. 7, pp. 11 105–11 119, 2025

    2025

  7. [7]

    Secure transceiver design for discrete RIS enhanced dual-functional radar-communication: A symbol-level pre coding ap- proach,

    R. Y ang et al. , “Secure transceiver design for discrete RIS enhanced dual-functional radar-communication: A symbol-level pre coding ap- proach,” IEEE Wireless Commun. Lett. , vol. 14, no. 4, pp. 1034–1038, 2025

    2025

  8. [8]

    Movable antenna enhanced wir eless sensing via antenna position optimization,

    W. Ma, L. Zhu, and R. Zhang, “Movable antenna enhanced wir eless sensing via antenna position optimization,” IEEE Trans. Wireless Com- mun., vol. 23, no. 11, pp. 16 575–16 589, 2024

    2024

  9. [9]

    Movable antennas for wireles s commu- nication: Opportunities and challenges,

    L. Zhu, W. Ma, and R. Zhang, “Movable antennas for wireles s commu- nication: Opportunities and challenges,” IEEE Commun. Mag. , vol. 62, no. 6, pp. 114–120, 2024

    2024

  10. [10]

    Movable-antenna en hanced multiuser communication via antenna position optimizatio n,

    L. Zhu, W. Ma, B. Ning, and R. Zhang, “Movable-antenna en hanced multiuser communication via antenna position optimizatio n,” IEEE Trans. Wireless Commun. , vol. 23, no. 7, pp. 7214–7229, 2024

    2024

  11. [11]

    Movable antenna for wireless communications: Pro- totyping and experimental results,

    Z. Dong et al. , “Movable antenna for wireless communications: Pro- totyping and experimental results,” IEEE Trans. Wireless Commun. , vol. 25, pp. 6586–6599, 2026

    2026

  12. [12]

    Modeling and performance an alysis for movable antenna enabled wireless communications,

    L. Zhu, W. Ma, and R. Zhang, “Modeling and performance an alysis for movable antenna enabled wireless communications,” IEEE Trans. Wireless Commun., vol. 23, no. 6, pp. 6234–6250, 2024

    2024

  13. [13]

    Robust transceiver design for RIS enhanced dual- functional radar-communication with movable antenna,

    R. Y ang et al. , “Robust transceiver design for RIS enhanced dual- functional radar-communication with movable antenna,” IEEE Trans. V eh. Technol., pp. 1–15, 2026

    2026

  14. [14]

    Joint discrete antenna positioning and beamforming optimization in movable antenna enabled full-duplex ISAC n etworks,

    Z. Li et al. , “Joint discrete antenna positioning and beamforming optimization in movable antenna enabled full-duplex ISAC n etworks,” IEEE Trans. Wireless Commun. , vol. 25, pp. 7220–7234, 2026. JOURNAL OF LATEX CLASS FILES, VOL. 18, NO. 9, SEPTEMBER 2020 6

    2026

  15. [15]

    Movable Antenna Empow- ered Covert Dual-Functional Radar-Communication,

    R. Y ang, N. Wei, Z. Dong et al. , “Movable Antenna Empow- ered Covert Dual-Functional Radar-Communication,” arXiv e-prints , p. arXiv:2601.14868, Jan. 2026

    work page arXiv 2026

  16. [16]

    Meta-reinforcement learning optimization for mov able antenna- aided full-duplex cf-dfrc systems with carrier frequency o ffset,

    Y . Xiu, W. Lyu, Y . Li, R. Y ang, P . L. Y eoh, W. Zhang, G. Liu, and N. Wei, “Meta-reinforcement learning optimization for mov able antenna- aided full-duplex cf-dfrc systems with carrier frequency o ffset,” IEEE Transactions on Communications , vol. 74, pp. 5803–5819, 2026

    2026

  17. [17]

    Robust optimization for movable antenna-aided cell-free isac with time synchronization errors,

    Y . Xiu, Y . Zhao, R. Y ang, W. Lyu, D. Niyato, D. In Kim, G. Li u, and N. Wei, “Robust optimization for movable antenna-aided cell-free isac with time synchronization errors,” IEEE Transactions on Wireless Communications, vol. 25, pp. 10 082–10 097, 2026

    2026

  18. [18]

    Mova ble antenna enabled isac beamforming design for low-altitude a irborne vehicles,

    Y . Xiu, S. Y ang, W. Lyu, P . Lep Y eoh, Y . Li, and Y . Ai, “Mova ble antenna enabled isac beamforming design for low-altitude a irborne vehicles,” IEEE Wireless Communications Letters , vol. 14, no. 5, pp. 1311–1315, 2025

    2025

  19. [19]

    Convex optimization,

    S. Boyd, “Convex optimization,” Cambridge UP , Mar. 2004

    2004