Pinching-Antenna Systems (PASS) Meet Multiple Access: NOMA or OMA?

Qiao Ren; Siyu Lin; Xidong Mu; Yuanwei Liu

REVIEW 3 major objections 6 minor 10 references

Pinching-Antenna Systems (PASS) Meet Multiple Access: NOMA or OMA?

T0 review · 3 major / 6 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read For two equal-rate users, TDMA with per-slot antenna placement uses less transmit power than NOMA once four or more pinching antennas are deployed.

desk verdict A useful three-way comparison in pinching-antenna systems, but the TDMA-over-NOMA headline rests on a provably optimal TDMA baseline versus a heuristic NOMA baseline, so the central quantitative claim needs a stronger NOMA benchmark before it's fully convincing. read the letter →

arxiv 2506.13490 v1 pith:KILUO6NG submitted 2025-06-16 eess.SP

Qiao Ren , Xidong Mu , Siyu Lin , Yuanwei Liu This is my paper · ORCID

classification eess.SP

keywords pinching-antennasystemsmultipleaccessNOMApinchingbeamformingtransmitpowerminimizationtime-divisionsuccessiveconvexapproximationflexibleantennas

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 asks which multiple-access scheme a two-user pinching-antenna system (PASS) should use: non-orthogonal multiple access (NOMA), frequency-division multiple access (FDMA), or time-division multiple access (TDMA). It sets up power-minimization problems in which pinching-antenna positions along a dielectric waveguide are optimized so that each user's target rate is met with the least transmit power. The central finding is that when the two users demand the same rate, TDMA can beat NOMA once at least four pinching antennas are active, because TDMA lets each time slot place its own antenna cluster around the user being served. The paper also finds that NOMA always needs less power than FDMA and that PASS greatly outperforms conventional fixed-antenna baselines.

What carries the argument

The central mechanism is the time-switching feature of PASS: under TDMA, each time slot can use its own antenna-position vector $\mathbf{x}_{p,k}$, so the array is redeployed to serve each user in turn. For a single-user slot, the optimal placement is to cluster all $N$ pinching antennas symmetrically around the user's $x$-coordinate with equal spacing $\Delta$, a closed-form result the paper takes from prior work. For NOMA and FDMA, by contrast, one shared placement must balance the two users' channels, and the paper handles the resulting non-convex problem with a two-stage algorithm: a successive-convex-approximation stage that fixes coarse positions by minimizing large-scale path loss, followed by a fine-tuning stage that shifts antennas on a wavelength scale so their phases combine constructively at both users. The phase-alignment constraint $\phi_k^n - \phi_k^{n-1} = 2m_k\pi$ is what makes constructive beamforming possible, and the first-order Taylor approximation of the phase around neighboring antennas is what allows the fine-tuning step to be computed.

What would settle it

Run the same two-user power-minimization problem with an exact global solver for NOMA and FDMA at $N=6$ with equal target rates; if the resulting NOMA transmit power falls below the TDMA value, the reported ordering is an artifact of the heuristic. Alternatively, measure the energy and switching time required to move the pinching antennas between the two TDMA slot positions and add them to the power budget; a large enough overhead invalidates the practical ranking.

Watch

Extended reading notes

Core claim

For a single-waveguide PASS with N pinching antennas serving two users, the paper argues that the best multiple-access choice is not fixed: it depends on how symmetric the users' rate demands are. When the rate targets are equal or close, time-division multiple access with per-slot pinching-antenna placement is more power-efficient than NOMA for N≥4, because each TDMA slot can position the full antenna array to maximize the channel gain of exactly one user. When rate targets are asymmetric, NOMA regains the advantage, since its shared resource block exploits the disparity in channel conditions. The paper also establishes that NOMA consistently requires less transmit power than FDMA under the same PASS setup, and that PASS itself yields large power savings relative to a conventional single antenna or a fixed-position array.

Load-bearing premise

The comparison's conclusion depends on treating the TDMA solution as optimal while the NOMA and FDMA solutions come from a heuristic without a proven gap, and on assuming that reconfiguring the antennas between time slots costs no power or time; if the heuristic is far from optimal or reconfiguration overhead is material, the TDMA-over-NOMA result could reverse.

Editorial extensions

If this is right

For symmetric user rate requirements in a two-user PASS downlink, orthogonal access via TDMA can be more power-efficient than NOMA once $N \ge 4$, so NOMA is not automatically the best multiple-access scheme for pinching-antenna systems.
NOMA still beats FDMA in this setting, so among the two orthogonal schemes the choice matters: time-splitting exploits PASS's reconfigurability, while frequency-splitting does not.
PASS-based beamforming can sharply reduce required transmit power relative to a conventional single antenna or a fixed-position array with the same number of elements.
When user rate targets are asymmetric, NOMA's spectrum-efficiency advantage over orthogonal access reappears, so the multiple-access choice should adapt to traffic symmetry.
The optimal TDMA deployment guideline is simple: in each slot, place the pinching antennas symmetrically around the served user with equal spacing, then refine the phases.

Reading between the lines

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

A fair reader might infer that the NOMA and FDMA power curves are upper bounds, since the two-stage algorithm is heuristic; a better NOMA solver could narrow or close the reported gap at $N \ge 4$.
Because the TDMA analysis ignores the cost of moving pinching antennas between slots, an energy-aware comparison with switching overhead included could favor NOMA for short time slots or slow reconfiguration.
The results suggest a practical hybrid: use TDMA with per-slot antenna clustering for symmetric traffic and switch to NOMA with a shared asymmetric placement when user rates diverge.
A testable extension would be to derive an optimality gap or a global-search benchmark for the NOMA phase-alignment problem, turning the numerical ordering into a provable one.

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.

Desk Editor's Note

read the letter

Dear colleague,

Two things to know. First, the paper actually delivers a new result: a three-way comparison between NOMA, FDMA, and TDMA in a two-user pinching-antenna system, with the specific finding that TDMA beats NOMA when users demand symmetric rates, because each time slot can place all PAs optimally for one user. This is a genuinely useful design insight for the PASS subfield. Second, the headline is built on an asymmetric baseline: TDMA uses the provably optimal PA placement from prior work [10], while NOMA and FDMA use a two-stage SCA plus local phase fine-tuning with no optimality gap established. The crossover in Fig. 2 could be an artifact of a suboptimal NOMA solver rather than a fundamental property. That concern is the main soft spot, and it is addressable.

What the paper does well: the system model is clean, the SIC power derivation is careful, the SCA subproblem is convex, the complexity is stated, and the PA deployment visualizations in Fig. 4 give intuitive design rules — NOMA wants asymmetric PAs, FDMA splits clusters, TDMA focuses per slot. The paper also explicitly acknowledges the implementation overhead of TDMA. Credit is given appropriately to [8] and [10] for the phase alignment and the TDMA optimal placement.

Where it is soft. The NOMA/FDMA algorithm is heuristic, and the paper never benchmarks it — no exhaustive search for small N, no grid search, no optimality gap. Because the comparison treats the TDMA solution as globally optimal and the NOMA solution as 'high-quality,' the quantitative crossover claims are not like-for-like. The abstract states the TDMA-over-NOMA result without this caveat. Second, the numerical figures show single simulation runs; there are no error bars or multiple channel realizations, so we don't know if the crossover is stable across user positions. Third, the fine-tuning explicitly relaxes the exact phase alignment constraint, so the channel gain achieved is not guaranteed to be near the coherent-combining upper bound. None of these are fatal; all are fixable with additional experiments and caveats.

Who this is for: researchers working on pinching-antenna systems, particularly those interested in multiple access. It deserves a serious referee. I'd send it to review with a specific request: add a benchmark for the NOMA solver (even a small exact search) and report statistics over user drops. With those, the TDMA-over-NOMA result would be either confirmed or qualified in a meaningful way.

Referee Report

3 major / 6 minor

Summary. The paper considers a fundamental two-user pinching-antenna system (PASS) under three multiple-access schemes: NOMA, FDMA, and TDMA. For each scheme, a transmit-power minimization problem is formulated subject to user rate requirements. For NOMA and FDMA, the paper develops a two-stage algorithm that first solves a large-scale path-loss surrogate via successive convex approximation and then fine-tunes PA positions for phase coherence. For TDMA, it invokes a globally optimal PA placement from prior work that deploys PAs symmetrically around each user with equal spacing. Numerical results show that PASS significantly outperforms conventional antenna systems, NOMA consistently outperforms FDMA, and TDMA outperforms NOMA for symmetric user rate requirements when the number of PAs N≥4.

Significance. If the conclusions are substantiated, the finding that time-switched per-slot PA reconfiguration (TDMA) can beat power-domain NOMA in a two-user PASS under symmetric rate targets is non-obvious and useful for system design; it would also add a new data point to the ongoing discussion of NOMA versus OMA in flexible-antenna systems. The problem formulation is clean, the power expressions are derived carefully, and the SCA subproblem is convex and solvable. The paper provides reproducible numerical experiments and a clear deployment guideline from Fig. 4. However, the central comparison is currently undermined by an asymmetry in solution quality between the proven-optimal TDMA baseline and the heuristic NOMA/FDMA baseline.

major comments (3)

[Sec. IV, Figs. 2-3] The central conclusion that TDMA requires less transmit power than NOMA for N≥4 (Fig. 2) and at equal rate targets (Fig. 3) is obtained by comparing the globally optimal TDMA solution of Lemma 1 (from [10]) with NOMA and FDMA solutions from the heuristic two-stage algorithm of Sec. III-A, for which no optimality gap or convergence certificate is given. Because the NOMA solver may be substantially suboptimal, the reported crossover could be an artifact of the solver rather than a fundamental property of PASS with NOMA. The authors should either quantify the performance of the NOMA/FDMA algorithm (e.g., by comparing against an exhaustive grid search or a global optimization method in the small-N regime) or explicitly frame the conclusion as "TDMA outperforms the proposed NOMA algorithm" in the abstract and in Sec. IV.
[Sec. III-A2, Eqs. (15)-(24)] The fine-tuning stage does not enforce the exact phase-alignment constraint (15) for both users; as the paper itself states, a single uniform PA spacing cannot maximize channel gains for both users simultaneously. The algorithm searches a restricted segment to minimize the surrogate objective (20), but it does not verify whether the refined positions satisfy (15) for either user or whether the actual objective (10) is improved. Consequently, the transmit powers reported for NOMA and FDMA in Figs. 2 and 3 are not certified to correspond to achievable channel gains |v_k|^2 used in the objective, so the comparison with TDMA is not an apples-to-apples comparison of achievable performance.
[Sec. II-A, Eq. (10)] The NOMA problem (10) is solved for a fixed SIC order (λ1=0, λ2=1) without including the corresponding ordering condition |v1|^2 ≥ |v2|^2. Since the PA placement is part of the optimization, the solution may violate this condition, in which case the power expressions in (9) do not represent a valid NOMA transmission. The exhaustive search over the two SIC orders mentioned in the text does not resolve this unless each subproblem enforces the channel-gain inequality. Please add the ordering constraint or explicitly verify the ordering for the reported optimized solutions.

minor comments (6)

[Abstract and Sec. III-A] The word "fine-tuning" is misspelled as "fine-turning" in the abstract, in the header of Sec. III-A, and several times in the body text.
[Sec. IV] The symbol δ is used both for the minimum PA spacing (λ/2) and for the relative offset x_p^n - x_p^c in the fine-tuning procedure; this dual use is confusing and should be resolved with distinct symbols.
[Eq. (17)] In problem (17), the constraint list "(8b),(8c)" appears to be a typo; the rate constraint (8b) is no longer present after the transformation, so it should be "(8c),(8d)" as in (20d).
[Fig. 4 caption] The labels T1 and T2 are not defined in the caption; please define them as the two TDMA time slots.
[Sec. II-A and Eqs. (23)-(24)] The sentence "both of them yield similar characteristics" is vague; please clarify what is similar between the two SIC orders and why an exhaustive search over the two orders is sufficient. Also, the hat notation in bΔx and bΔx′ (Eqs. (23)-(24)) is not defined.
[Sec. IV] The parameter "n_neff" should be typeset as n_eff for consistency with the notation introduced in Sec. II.

Circularity Check

0 steps flagged · score 2.0 of 10

No circularity: TDMA-vs-NOMA comparison comes from direct simulation of independently formulated optimization problems; only a minor non-load-bearing self-citation appears.

full rationale

No step in the paper reduces by construction to its own inputs. The central claim that TDMA requires less power than NOMA for symmetric rates is obtained by simulating the formulated power-minimization problems (8), (12), and (14), with no parameter fitted to a subset of the reported results and then renamed as a prediction. The TDMA optimality is imported from Lemma 1 with proof cited to [10, Appendix B], a reference whose author list (Xu, Ding, Karagiannidis) is disjoint from the present authors, so this is not a self-citation chain. The only self-citations are [8] (Zhu/Mu/Liu) for the Euclidean-distance expression and the first-order phase Taylor expansion used in fine-tuning, and [4] (Liu/Mu) for PASS background. These support the phase-coherent fine-tuning stage, but the approximation is written out explicitly in Eqs. (21)-(22) and is not the source of the TDMA-over-NOMA conclusion. The NOMA/FDMA two-stage SCA-plus-fine-tuning algorithm does lack an optimality certificate, so the comparison is an algorithmic benchmark rather than a proof of a fundamental ordering; that is a correctness or optimality-gap concern, not circularity. No fitted input is called a prediction, no self-citation is load-bearing, no uniqueness theorem is imported from the authors' own prior work, and no known result is merely renamed. The derivation is therefore not circular, and the minor self-citation does not affect the independence of the central numerical comparison.

Assumptions & free parameters 2 free parameters · 5 assumptions · 0 invented entities

The paper introduces no new physical entities. The main assumed inputs are the PASS channel model, the time-switching capability, and the optimal TDMA placement from prior work. The algorithm has two hand-chosen fine-tuning parameters that affect the numerical comparison.

free parameters (2)

fine-tuning segment thresholds = 3 Delta, 4 Delta from central PA
The heuristic fine-tuning step in Section III-A2 restricts the search for each PA to intervals defined by 3 Delta and 4 Delta offsets; these are hand-chosen and affect the NOMA/FDMA power results and hence the comparison.
phase-alignment integer choices m_k = unspecified
The phase alignment constraints (22) involve integers m_k; the fine-tuning searches over locations but the paper does not specify how m_k are selected, effectively a hand-tuned degree of freedom in the algorithm.

assumptions (5)

standard math SCA convergence and Taylor linearization in (19) yield a convex subproblem (20) solvable by CVX.
The SCA method is standard and the linearization is valid, but convergence to a stationary point is not proven for this specific non-convex problem.
domain assumption PASS channel model (1)-(2): in-waveguide amplitudes 1/sqrt(N) independent of position, no waveguide loss, free-space path loss eta/d, and continuous PA activation with minimum spacing Delta.
This is the physical model for pinching antennas used throughout; the paper states it as an ideal assumption in Section II.
domain assumption The time-switching feature of PASS: each TDMA time slot can use a different PA configuration without inter-slot overhead.
Introduced in Section II-B2 and used to decouple the TDMA problem into per-user optimizations; the paper acknowledges overhead only in text, not in the rate model.
domain assumption Lemma 1 from [10]: for each user, the optimal PA placement is symmetric around the user with equal spacing Delta, valid only when D_k >> Delta.
The TDMA solution is imported from prior work [10] and its validity condition D_k >> Delta is assumed to hold in the simulations.
ad hoc to paper The fine-tuning 'balance' procedure does not satisfy the exact phase alignment (15) for both users simultaneously; instead it searches a restricted interval to minimize the SCA objective.
The authors admit that a single uniform PA phase adjustment cannot maximize both users' gains (Section III-A2), so the heuristic search is an ad hoc compromise.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Pinching-Antenna Systems (PASS) Meet Multiple Access: NOMA or OMA?." pith.science (2026). https://pith.science/paper/KILUO6NG

@misc{pith2026250613490,
  author       = {Pith},
  title        = {Pith review of: Pinching-Antenna Systems (PASS) Meet Multiple Access: NOMA or OMA?},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/KILUO6NG}},
  note         = {Machine review of arXiv:2506.13490}
}

read the original abstract

A fundamental two-user PASS-based communication system is considered under three MA schemes, namely non-orthogonal multiple access (NOMA), frequency division multiple access (FDMA), and time division multiple access (TDMA). For each MA scheme, a pinching beamforming optimization problem is formulated to minimize the required transmit power for satisfying users' rate requirements. For NOMA and FDMA, a two-stage algorithm is proposed, where the locations of PAs are derived sequentially by using the successive convex approximation (SCA) method and fine-turning phase adjustment. For TDMA, by leveraging the time-switching feature of PASS, the optimal pinching beamforming of each time slot is derived to maximize the served user channel gain. Numerical results are provided to show that: 1) PASS can achieve a significant performance gain over conventional antenna systems, and 2) NOMA consistently outperforms FDMA, while TDMA provides superior performance than NOMA for symmetric user rate requirements.

Figures

Figures reproduced from arXiv: 2506.13490 by the authors.

**Figure 2.** The transmit power changes with the number of PAs N. 1 1.5 2 2.5 3 3.5 4 4.5 5 -10 -5 0 5 10 15 20 [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗

Review history (2 revisions) →

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

10 extracted references · 1 canonical work pages

[10]

Rate maximization for downlink pinching-antenna systems,

Y . Xu, Z. Ding, and G. K. Karagiannidis, “Rate maximization for downlink pinching-antenna systems,”IEEE Wireless Commun. Lett., vol. 14, no. 5, pp. 1431–1435, May 2025

2025
[8]

Pinching- antenna systems (PASS)-enabled secure wireless communications,

G. Zhu, X. Mu, L. Guo, S. Xu, Y . Liu, and N. Al-Dhahir, “Pinching- antenna systems (PASS)-enabled secure wireless communications,” arXiv preprint arXiv:2504.13670, 2025

arXiv 2025
[1]

Fluid antenna systems,

K.-K. Wong, A. Shojaeifard, K.-F. Tong, and Y . Zhang, “Fluid antenna systems,”IEEE Trans. Wireless Commun., vol. 20, no. 3, pp. 1950–1962, Mar. 2021

1950
[2]

Modeling and performance analysis for movable antenna enabled wireless communications,

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

2024
[3]

Flexible-antenna systems: A pinching-antenna perspective,

Z. Ding, R. Schober, and H. Vincent Poor, “Flexible-antenna systems: A pinching-antenna perspective,”IEEE Trans. Commun., early access, doi:10.1109/TCOMM.2025.3555866

arXiv 2025
[4]

Pinching- antenna systems (PASS): Architecture designs, opportunities, and out- look,

Y . Liu, Z. Wang, X. Mu, C. Ouyang, X. Xu, and Z. Ding, “Pinching- antenna systems (PASS): Architecture designs, opportunities, and out- look,”arXiv preprint arXiv:2501.18409, 2025

arXiv 2025
[5]

Antenna activation for NOMA as- sisted pinching-antenna systems,

K. Wang, Z. Ding, and R. Schober, “Antenna activation for NOMA as- sisted pinching-antenna systems,”IEEE Wireless Commun. Lett., vol. 14, no. 5, pp. 1526–1530, May 2025

2025
[6]

Minimum data rate maximization for uplink pinching-antenna systems,

S. A. Tegos, P. D. Diamantoulakis, Z. Ding, and G. K. Karagiannidis, “Minimum data rate maximization for uplink pinching-antenna systems,” IEEE Wireless Commun. Lett., vol. 14, no. 5, pp. 1516–1520, May 2025

2025

Show all 10 references

[7]

Performance analysis of pinching- antenna systems,

D. Tyrovolas, S. A. Tegos, P. D. Diamantoulakis, S. Ioannidis, C. K. Liaskos, and G. K. Karagiannidis, “Performance analysis of pinching- antenna systems,”IEEE Trans. Cognit. Commun. Netw., to appear, 2025

2025
[9]

Boyd and L

S. Boyd and L. Vandenberghe,Convex Optimization. Cambridge, U.K.: Cambridge Univ. Press, 2004

2004

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.

[1] [10]

Rate maximization for downlink pinching-antenna systems,

Y . Xu, Z. Ding, and G. K. Karagiannidis, “Rate maximization for downlink pinching-antenna systems,”IEEE Wireless Commun. Lett., vol. 14, no. 5, pp. 1431–1435, May 2025

2025

[2] [8]

Pinching- antenna systems (PASS)-enabled secure wireless communications,

G. Zhu, X. Mu, L. Guo, S. Xu, Y . Liu, and N. Al-Dhahir, “Pinching- antenna systems (PASS)-enabled secure wireless communications,” arXiv preprint arXiv:2504.13670, 2025

arXiv 2025

[3] [1]

Fluid antenna systems,

K.-K. Wong, A. Shojaeifard, K.-F. Tong, and Y . Zhang, “Fluid antenna systems,”IEEE Trans. Wireless Commun., vol. 20, no. 3, pp. 1950–1962, Mar. 2021

1950

[4] [2]

Modeling and performance analysis for movable antenna enabled wireless communications,

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

2024

[5] [3]

Flexible-antenna systems: A pinching-antenna perspective,

Z. Ding, R. Schober, and H. Vincent Poor, “Flexible-antenna systems: A pinching-antenna perspective,”IEEE Trans. Commun., early access, doi:10.1109/TCOMM.2025.3555866

arXiv 2025

[6] [4]

Pinching- antenna systems (PASS): Architecture designs, opportunities, and out- look,

Y . Liu, Z. Wang, X. Mu, C. Ouyang, X. Xu, and Z. Ding, “Pinching- antenna systems (PASS): Architecture designs, opportunities, and out- look,”arXiv preprint arXiv:2501.18409, 2025

arXiv 2025

[7] [5]

Antenna activation for NOMA as- sisted pinching-antenna systems,

K. Wang, Z. Ding, and R. Schober, “Antenna activation for NOMA as- sisted pinching-antenna systems,”IEEE Wireless Commun. Lett., vol. 14, no. 5, pp. 1526–1530, May 2025

2025

[8] [6]

Minimum data rate maximization for uplink pinching-antenna systems,

S. A. Tegos, P. D. Diamantoulakis, Z. Ding, and G. K. Karagiannidis, “Minimum data rate maximization for uplink pinching-antenna systems,” IEEE Wireless Commun. Lett., vol. 14, no. 5, pp. 1516–1520, May 2025

2025