Pith. sign in

REVIEW 3 major objections 4 minor 34 references

SIDMA turns multi-user semantic collisions into denoisable noise by interleaving features and allocating power by importance, supporting up to 100 concurrent users with higher reconstruction fidelity than prior schemes.

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-4.5

2026-07-13 07:47 UTC pith:DQMW3KKP

load-bearing objection Clean IDMA-style interleaving transplanted to semantic features, with solid moderate-K gains and careful asymptotics; the 100-user near-ideal claim needs the extra power-compensation step under pure AWGN. the 3 major comments →

arxiv 2607.08777 v1 pith:DQMW3KKP submitted 2026-05-27 cs.IT math.IT

SIDMA: Semantic Interleave Division Multiple Access Communication System

Yunlu Wang , Chen Dong , Sen Wang , Lei Teng , Yaping Sun , Xiaodong Xu , Ping Zhang This is my paper
classification cs.IT math.IT
keywords Semantic CommunicationInterleave Division Multiple AccessAdaptive Power AllocationStructural WhiteningMulti-user InterferencePeak Sidelobe LevelSwin Transformer
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.

Semantic communication tries to send meaning, not bits, but when many users share the same channel their structured semantic features crash into one another and the reconstruction collapses. This paper argues that the fix is not stricter isolation of models or symbols, but deliberate scrambling: each user’s semantic map is randomly permuted so that any other user’s interference becomes unstructured noise that a hierarchical decoder can simply denoise. An importance-aware power allocator then protects the few features that matter most for image quality. Theory shows the residual cross-correlation vanishes as feature dimension grows, and simulations on DIV2K images claim better PSNR/SSIM than OMDMA, DeepMA, Shared Embedding, and classical JPEG-style multiple access, while scaling to 100 simultaneous users—the largest concurrency the authors report for semantic multiple access.

Core claim

By applying independent random permutations to importance-weighted semantic feature maps, multi-user interference is converted from structured semantic collisions into asymptotically white noise that a Swin-based decoder can suppress, and importance-aware power allocation further protects the critical elements, yielding higher reconstruction fidelity and the ability to support far denser concurrent users than existing semantic multiple-access schemes.

What carries the argument

Semantic structural whitening via a user-specific permutation operator (extended from classical IDMA interleaving) that drives Peak Sidelobe Level and multi-user cosine similarity to zero in high dimension, paired with the ImpPA neural module that maps saliency and SNR into per-element power weights under a total-power constraint.

Load-bearing premise

Semantic feature values stay bounded by a fixed constant no matter how large the map grows, and the hierarchical decoder will treat the scrambled interference purely as denoisable unstructured noise rather than residual structured artifacts.

What would settle it

Measure PSNR/SSIM for K=100 concurrent users on DIV2K at moderate SNR: if SIDMA falls below DeepMA or SE, or if the reconstructed images show structured hallucinations instead of mere noise, the whitening-plus-denoising claim fails.

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

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 / 4 minor

Summary. The paper proposes SIDMA, an end-to-end multi-user semantic communication architecture that applies user-specific random permutations (structural whitening) to Swin-Transformer feature maps, combined with an Importance-aware Power Allocation (ImpPA) module that maps attention-derived saliency and instantaneous SNR into a power matrix under a total-power constraint. After superposition over AWGN, the receiver de-interleaves and decodes. Theoretical analysis shows that the Peak Sidelobe Level of a randomly permuted, zero-mean unit-variance feature vector converges to zero (asymptotic whitening) and that independent interleavers drive multi-user cosine similarity to zero in probability. Simulations on DIV2K claim superior PSNR/SSIM versus OMDMA, DeepMA, SE and classical JPEG/JPEG2000 for small-to-moderate K, plus near-ideal reconstruction up to K=100 after an additional power-compensation step.

Significance. If the claims hold under realistic conditions, SIDMA would be a useful addition to the emerging semantic multiple-access literature by moving interference management into the interleaving domain and by coupling it with content-aware power control. The PSL and cosine-similarity derivations are carefully executed under the stated normalizations, the ablation isolating ImpPA is informative, and the comparison set includes the most relevant recent AI-native baselines. The 100-user scaling result, if robust, would be the first reported demonstration of that density for semantic MA.

major comments (3)
  1. [§V-D / Fig. 7] Section V-D and Fig. 7: the headline claim of near-ideal reconstruction for K up to 100 rests on an additional “power compensation” step that is never defined by equation in §III-C (ImpPA eqs. (14)–(17)) or elsewhere. The ablation text itself states that pure structural whitening without this step is worse than Direct Superposition; consequently the massive-connectivity result is not produced by the theoretically analyzed interleaving + ImpPA alone. The precise mapping, whether it is a simple SNR-dependent gain, a modified ImpPA, or an extra free parameter, must be stated and the corresponding curves re-generated.
  2. [§V (all experiments)] All numerical results (Figs. 4–7) are generated exclusively under perfect-CSI AWGN (eq. (6)) at fixed target SINR. The abstract and conclusion assert “robustness in resource-constrained environments” and “scalability o 100 concurrent users.” Residual finite-N correlations after interleaving can interact with multipath or fading; without at least Rayleigh or frequency-selective experiments the strong scalability claim remains untested.
  3. [§IV-A] Section IV-A, eqs. (38)–(39): the PSL o0 proof invokes a uniform bound max |z_p|≤M independent of dimension N so that fourth moments remain O(N). Neural feature maps produced by a Swin Transformer need not satisfy this a priori; an empirical histogram of feature amplitudes (or a moment-based relaxation) is required to confirm that the asymptotic argument applies to the actual encoder outputs used in the simulations.
minor comments (4)
  1. [Title / Abstract] Title uses “Interleaved” while the abstract and many body occurrences use “Interleave”; standardize the acronym expansion.
  2. [Fig. 3] Fig. 3 heatmaps lack color-bar units and absolute PSL values; quantitative before/after numbers would strengthen the seed-selection claim.
  3. [§III-C] Eq. (17) writes the interference term with a composite permutation index that is never formally defined; a short clarifying sentence after eq. (10) would help.
  4. [§V-B] The OMDMA MSE differentiation metric is useful but its precise computation (which encoder–decoder pairs, how many images) is only sketched; a one-sentence protocol would aid reproducibility.

Circularity Check

0 steps flagged

No load-bearing circularity: asymptotic whitening/orthogonality is derived from independent random permutations + standard combinatorial/CLT/EVT arguments under stated boundedness assumptions; simulations compare against external baselines on public data.

full rationale

The core derivation chain (structural whitening via permutation operators, PSL o0, multi-user cosine-similarity variance 1/(N-1) and collision probability o0) starts from explicit assumptions (zero-mean/unit-variance normalization of features, independent uniform random permutations without replacement, uniform boundedness max|z_p|≤M independent of N) and applies standard combinatorial expansions, Combinatorial Central Limit Theorem, Extreme Value Theory and Chebyshev; none of these steps define the claimed output in terms of itself or fit a free parameter to the target quantity and then re-label it a prediction. ImpPA is a trained feed-forward map from attention-derived importance + SNR to power weights under a total-power constraint; its gains are measured by ablation against uniform allocation, not by construction. Seed selection minimizes empirical PSL on encoder features but is a design choice, not a circular prediction of reconstruction metrics. Self-citations exist (e.g., OMDMA by overlapping authors appears as a baseline), yet they are not invoked as uniqueness theorems or load-bearing premises for the whitening claim; the paper evaluates against them (and against DeepMA, SE, JPEG/JPEG2000) on the public DIV2K set. The 100-user near-ideal curves rely on an additional underspecified “power compensation” step under idealized AWGN, but that is an incompleteness/correctness issue, not a reduction of a claimed first-principles result to its own inputs. Hence only a minimal residual score for non-load-bearing self-citation of related semantic-MA work.

Axiom & Free-Parameter Ledger

4 free parameters · 4 axioms · 3 invented entities

The central claims rest on standard wireless and deep-learning assumptions plus two paper-specific modeling choices (bounded feature amplitudes and decoder noise-like treatment of scrambled interference). Free parameters are the usual neural-network weights and the discrete seed set; no new physical constants are introduced.

free parameters (4)
  • ImpPA network weights f_ϕ
    Learned mapping from importance map + SNR to power coefficients; fitted end-to-end on DIV2K training images.
  • Swin Transformer encoder/decoder parameters θ, ϕ
    Standard semantic transcoder weights trained jointly with ImpPA.
  • Interleaving seed set S_opt
    Discrete seeds chosen by Algorithm 1 to minimize empirical PSL on a feature set; finite candidate pool.
  • Total power P_total and SNR operating points
    Simulation hyper-parameters that set the absolute scale of the power allocation and noise variance.
axioms (4)
  • ad hoc to paper Semantic feature vectors after normalization have exact zero mean and unit variance, and their absolute values are uniformly bounded by a finite M independent of dimension N.
    Invoked in Section IV-A to bound fourth-order moments and prove PSL → 0; not derived from first principles of the Swin encoder.
  • domain assumption Hierarchical neural decoders treat unstructured high-frequency interference as ordinary additive noise that can be suppressed by the reconstruction network.
    Stated in III-B as the mechanism that converts whitened collisions into manageable noise; standard in JSCC literature but not proven for the residual multi-user case.
  • domain assumption Wireless channel is AWGN (or flat fading with known gains) and users share identical time-frequency resources with perfect synchronization.
    System model (6) and all simulations; classical wireless assumption.
  • standard math Combinatorial Central Limit Theorem and Extreme Value Theory apply to the sidelobe statistics of random permutations of bounded sequences.
    Used to obtain the O(√(N ln N)) extreme-sidelobe bound in IV-A.
invented entities (3)
  • Semantic Interleave Division Multiple Access (SIDMA) architecture no independent evidence
    purpose: End-to-end multi-user semantic transceiver that couples feature-domain permutation whitening with importance-aware power allocation.
    The paper’s central proposed system; no independent experimental existence outside this work.
  • Importance-aware Power Allocation (ImpPA) module no independent evidence
    purpose: Neural network that maps semantic saliency and instantaneous SNR to per-element power weights under a total-power constraint.
    New learned component introduced to protect high-value features; trained only on the paper’s DIV2K setup.
  • Composite scrambling operator Φ_{j→k} = Π_k^{-1} ∘ Π_j independent evidence
    purpose: Mathematical object that describes the effective channel seen by interfering users after mismatched de-interleaving.
    Defined in III-B to analyze whitening; purely notational, no new physics.

pith-pipeline@v1.1.0-grok45 · 20732 in / 3274 out tokens · 34069 ms · 2026-07-13T07:47:11.851064+00:00 · methodology

0 comments
read the original abstract

Multiple Access (MA) technology has consistently served as the core driving force behind the evolution of mobile communications. As a promising paradigm for next-generation communications, Semantic Communication explores entirely new semantic spatial resources by mining the deep meaning of information. However, the inherent spatial correlation and importance heterogeneity of semantic features often cause semantic collisions and semantic collapse in multi-user concurrent transmission scenarios. To address these challenges, this paper proposes a Semantic Interleaved Division Multiple Access (SIDMA) technique. By utilizing a permutation operator to perform structural whitening on semantic features and combining it with an Importance-aware Power Allocation (ImpPA) module for differentiated protection, SIDMA scatters core features across the interleaving domain and adaptively optimizes power levels based on real-time channel conditions. Simulation results demonstrate that, compared with traditional MA techniques and advanced semantic multiple access schemes including Orthogonal-Model Division Multiple Access (OMDMA), Deep Multiple Access (DeepMA), and Shared Embedding (SE), the proposed SIDMA exhibits superior reconstruction fidelity and scalability in multi-user concurrent transmissions, effectively enhancing the communication quality and robustness in resource-constrained environments.

Figures

Figures reproduced from arXiv: 2607.08777 by Chen Dong, Lei Teng, Ping Zhang, Sen Wang, Xiaodong Xu, Yaping Sun, Yunlu Wang.

Figure 1
Figure 1. Figure 1: The overall framework of the proposed Semantic Interleaving Division Multiple Access (SIDMA) system for multi-user downlink transmission, [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: The detailed network architecture of SIDMA, featuring a semantic Encoder, a neural network-based adaptive power allocation module, and a semantic [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Comparison of PSL matrices (autocorrelation heatmaps) for semantic features. (a) Under a random interleaving seed, visible sidelobes indicate residual [PITH_FULL_IMAGE:figures/full_fig_p009_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Performance comparison of SIDMA, DeepMA, SE, OMDMA, and traditional MA schemes in terms of average PSNR and SSIM versus SNR on [PITH_FULL_IMAGE:figures/full_fig_p010_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Performance comparison of multiple access schemes under varying user densities and SNR levels. The results reveal a clear operational trade-off: [PITH_FULL_IMAGE:figures/full_fig_p011_5.png] view at source ↗
Figure 7
Figure 7. Figure 7: Ablation study evaluating the synergy between structural whitening [PITH_FULL_IMAGE:figures/full_fig_p012_7.png] view at source ↗
Figure 6
Figure 6. Figure 6: Ablation study comparing ImpPA-based adaptive power allocation [PITH_FULL_IMAGE:figures/full_fig_p012_6.png] 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

34 extracted references

  1. [1]

    Semantic Communication System Based on Semantic Slice Models Propagation,

    C. Dong, H. Liang, X. Xu, S. Han, B. Wang, and P. Zhang, “Semantic Communication System Based on Semantic Slice Models Propagation,” IEEE Journal on Selected Areas in Communications, vol. 41, no. 1, pp. 202–213, 2023

    2023

  2. [2]

    A wearable obstacle avoidance device for visually impaired individuals with cross-modal learning,

    Y . Gao, D. Wu, J. Song, X. Zhang, B. Hou, H. Liu, J. Liao, and L. Zhou, “A wearable obstacle avoidance device for visually impaired individuals with cross-modal learning,”Nature Communications, vol. 16, no. 1, p. 2857, 2025

    2025

  3. [3]

    Toward Generic Cross-Modal Transmission Strategy,

    X. Wei, J. Liao, L. Zhou, H. Sari, and W. Zhuang, “Toward Generic Cross-Modal Transmission Strategy,”IEEE Transactions on Communi- cations, vol. 72, no. 10, pp. 6059–6072, 2024

    2024

  4. [4]

    sDMCM—A Semantic Digital Modulation Constellation Mapping Scheme for Semantic Communica- tion,

    L. Teng, W. An, C. Dong, and X. Xu, “sDMCM—A Semantic Digital Modulation Constellation Mapping Scheme for Semantic Communica- tion,”IEEE Internet of Things Journal, vol. 12, no. 12, pp. 20885–20901, 2025

    2025

  5. [5]

    Multiple Access in Cognitive Radio Networks: From Orthogonal and Non- Orthogonal to Rate-Splitting,

    S. Gamal, M. Rihan, S. Hussin, A. Zaghloul, and A. A. Salem, “Multiple Access in Cognitive Radio Networks: From Orthogonal and Non- Orthogonal to Rate-Splitting,”IEEE Access, vol. 9, pp. 95569–95584, 2021

    2021

  6. [6]

    Digital communications by satellite,

    J. J. Spilker Jr, “Digital communications by satellite,”Englewood Cliffs, 1977

    1977

  7. [7]

    A Satellite Time-Division Multiple-Access Experiment,

    T. Sekimoto and J. Puente, “A Satellite Time-Division Multiple-Access Experiment,”IEEE Transactions on Communication Technology, vol. 16, no. 4, pp. 581–588, 1968

    1968

  8. [8]

    On the capacity of a cellular CDMA system,

    K. Gilhousen, I. Jacobs, R. Padovani, A. Viterbi, L. Weaver, and C. Wheatley, “On the capacity of a cellular CDMA system,”IEEE Transactions on V ehicular Technology, vol. 40, no. 2, pp. 303–312, 1991

    1991

  9. [9]

    Data Transmission by Frequency-Division Multiplexing Using the Discrete Fourier Transform,

    S. Weinstein and P. Ebert, “Data Transmission by Frequency-Division Multiplexing Using the Discrete Fourier Transform,”IEEE Transactions on Communication Technology, vol. 19, no. 5, pp. 628–634, 1971

    1971

  10. [10]

    The Road to Next- Generation Multiple Access: A 50-Year Tutorial Review,

    Y . Liu, C. Ouyang, Z. Ding, and R. Schober, “The Road to Next- Generation Multiple Access: A 50-Year Tutorial Review,”Proceedings of the IEEE, vol. 112, no. 9, pp. 1100–1148, 2024

    2024

  11. [11]

    Non-orthogonal multiple access (NOMA) for future radio access,

    R. Razavi, M. Dianati, and M. A. Imran, “Non-orthogonal multiple access (NOMA) for future radio access,” in5G Mobile Communications, pp. 135–163, Springer, 2016

    2016

  12. [12]

    Evolution of NOMA Toward Next Generation Multiple Access (NGMA) for 6G,

    Y . Liu, S. Zhang, X. Mu, Z. Ding, R. Schober, N. Al-Dhahir, E. Hossain, and X. Shen, “Evolution of NOMA Toward Next Generation Multiple Access (NGMA) for 6G,”IEEE Journal on Selected Areas in Commu- nications, vol. 40, no. 4, pp. 1037–1071, 2022

    2022

  13. [13]

    A rate-splitting approach to the Gaussian multiple-access channel,

    B. Rimoldi and R. Urbanke, “A rate-splitting approach to the Gaussian multiple-access channel,”IEEE Transactions on Information Theory, vol. 42, no. 2, pp. 364–375, 1996

    1996

  14. [14]

    A Primer on Rate-Splitting Multiple Access: Tutorial, Myths, and Frequently Asked Questions,

    B. Clerckx, Y . Mao, E. A. Jorswieck, J. Yuan, D. J. Love, E. Erkip, and D. Niyato, “A Primer on Rate-Splitting Multiple Access: Tutorial, Myths, and Frequently Asked Questions,”IEEE Journal on Selected Areas in Communications, vol. 41, no. 5, pp. 1265–1308, 2023

    2023

  15. [15]

    A Tutorial on Wideband XL-MIMO: Challenges, Opportunities, and Future Trends,

    M. Parvini, B. Banerjee, M. Q. Khan, T. Mewes, A. Nimr, and G. Fet- tweis, “A Tutorial on Wideband XL-MIMO: Challenges, Opportunities, and Future Trends,”IEEE Open Journal of the Communications Society, vol. 6, pp. 5509–5534, 2025

    2025

  16. [16]

    Orthogonal Model Division Multiple Access,

    H. Liang, K. Liu, X. Liu, H. Jiang, C. Dong, X. Xu, K. Niu, and P. Zhang, “Orthogonal Model Division Multiple Access,”IEEE Trans- actions on Wireless Communications, vol. 23, no. 9, pp. 11693–11707, 2024

    2024

  17. [17]

    DeepMA: End-to-End Deep Multiple Access for Wireless Image Transmission in Semantic Communication,

    W. Zhang, K. Bai, S. Zeadally, H. Zhang, H. Shao, H. Ma, and V . C. M. Leung, “DeepMA: End-to-End Deep Multiple Access for Wireless Image Transmission in Semantic Communication,”IEEE Transactions on Cognitive Communications and Networking, vol. 10, no. 2, pp. 387– 402, 2024

    2024

  18. [18]

    NOMA Enhanced Semantic Communication with Swin Transformer in STINs,

    D. Li, K. Wang, X. Wang, and J. Liu, “NOMA Enhanced Semantic Communication with Swin Transformer in STINs,” inIEEE INFOCOM 2025 - IEEE Conference on Computer Communications Workshops (INFOCOM WKSHPS), pp. 1–6, 2025

    2025

  19. [19]

    Rate-Splitting Multiple Access for Coexistence of Semantic and Bit Communications,

    Y . Liu and B. Clerckx, “Rate-Splitting Multiple Access for Coexistence of Semantic and Bit Communications,”IEEE Transactions on Commu- nications, vol. 74, pp. 4367–4383, 2026

    2026

  20. [20]

    Semantic Communication Empowered 6G Networks: Techniques, Applications, and Challenges,

    Y . Wang, H. Han, Y . Feng, J. Zheng, and B. Zhang, “Semantic Communication Empowered 6G Networks: Techniques, Applications, and Challenges,”IEEE Access, vol. 13, pp. 28293–28314, 2025

    2025

  21. [21]

    Transformer-Based Shared Embedding for Multiple Access in Semantic Communications,

    K.-H. Lee, H.-H. Choi, and J.-R. Lee, “Transformer-Based Shared Embedding for Multiple Access in Semantic Communications,”IEEE Journal on Selected Areas in Communications, vol. 44, pp. 2622–2637, 2026

    2026

  22. [22]

    Interleave division multiple- access,

    L. Ping, L. Liu, K. Wu, and W. Leung, “Interleave division multiple- access,”IEEE Transactions on Wireless Communications, vol. 5, no. 4, pp. 938–947, 2006

    2006

  23. [23]

    Knowledge Distillation-Driven Semantic NOMA for Image Transmission With Dif- fusion Model,

    Q. Wang, Z. Gao, S. Sun, Z. Qin, X. Xu, and M. Tao, “Knowledge Distillation-Driven Semantic NOMA for Image Transmission With Dif- fusion Model,”IEEE Transactions on Wireless Communications, vol. 25, pp. 11783–11798, 2026

    2026

  24. [24]

    Semantics- Empowered Non-Orthogonal Multiple Access for Downlink Transmis- sion of Correlated Information Sources,

    W. Li, Y . Liu, C. Dong, X. Xu, P. Zhang, and L. Li, “Semantics- Empowered Non-Orthogonal Multiple Access for Downlink Transmis- sion of Correlated Information Sources,”IEEE Transactions on Wireless Communications, vol. 24, no. 9, pp. 7874–7891, 2025

    2025

  25. [25]

    Semantic Communication over Sparse Code Multiple Access Systems,

    Q. Wang, T. Qin, X. Wang, J. Liang, K. Han, and J. Hu, “Semantic Communication over Sparse Code Multiple Access Systems,” in2025 IEEE International Conference on Communications Workshops (ICC Workshops), pp. 1699–1704, 2025

    2025

  26. [26]

    Rate Splitting Multiple Access-Enabled Adaptive Panoramic Video Semantic Transmission,

    H. Gao, M. Sun, X. Xu, S. Han, B. Wang, J. Zhang, and P. Zhang, “Rate Splitting Multiple Access-Enabled Adaptive Panoramic Video Semantic Transmission,”IEEE Transactions on Wireless Communications, vol. 24, no. 11, pp. 9050–9068, 2025

    2025

  27. [27]

    Compression Ratio Allocation for Probabilistic Semantic Communication With RSMA,

    Z. Zhao, Z. Yang, Y . Hu, C. Zhu, M. Shikh-Bahaei, W. Xu, Z. Zhang, and K. Huang, “Compression Ratio Allocation for Probabilistic Semantic Communication With RSMA,”IEEE Transactions on Communications, vol. 73, no. 9, pp. 7304–7318, 2025

    2025

  28. [28]

    Rate-Splitting Multiple Access Enabled Green Probabilistic Semantic Communication Over Wireless Networks,

    R. Xu, Z. Yang, Y . Mao, C. Huang, Q. Yang, L. Xu, W. Xu, and Z. Zhang, “Rate-Splitting Multiple Access Enabled Green Probabilistic Semantic Communication Over Wireless Networks,”IEEE Transactions on Green Communications and Networking, vol. 9, no. 4, pp. 1472– 1486, 2025

    2025

  29. [29]

    Semantic Feature Division Multiple Access for Digital Se- mantic Broadcast Channels,

    S. Ma, Z. Sun, B. Shen, Y . Wu, H. Li, G. Shi, S. Li, and N. Al- Dhahir, “Semantic Feature Division Multiple Access for Digital Se- mantic Broadcast Channels,”IEEE Internet of Things Journal, vol. 12, no. 11, pp. 17123–17136, 2025

    2025

  30. [30]

    Semantic Feature Division Multiple Access for Digital Semantic Multi- ple Access Channels,

    B. Shen, S. Ma, R. Chen, Y . Wu, H. Li, G. Shi, S. Li, and N. Al-Dhahir, “Semantic Feature Division Multiple Access for Digital Semantic Multi- ple Access Channels,”IEEE Transactions on Cognitive Communications and Networking, vol. 12, pp. 283–297, 2026

    2026

  31. [31]

    Token-Domain Multiple Access: Exploiting Semantic Orthogonality for Collision Mit- igation,

    L. Qiao, M. B. Mashhadi, Z. Gao, and D. G ¨und¨uz, “Token-Domain Multiple Access: Exploiting Semantic Orthogonality for Collision Mit- igation,” inIEEE INFOCOM 2025 - IEEE Conference on Computer Communications Workshops (INFOCOM WKSHPS), pp. 1–6, 2025

    2025

  32. [32]

    Generative AI Empowered Semantic Feature Multiple Access (SFMA) Over Wireless Networks,

    J. Wang, Y . Yang, Z. Yang, C. Huang, M. Chen, Z. Zhang, and M. Shikh- Bahaei, “Generative AI Empowered Semantic Feature Multiple Access (SFMA) Over Wireless Networks,”IEEE Transactions on Cognitive Communications and Networking, vol. 11, no. 2, pp. 791–804, 2025

    2025

  33. [33]

    Semantic Feature Multiple Access (SFMA) Over Wireless Networks,

    J. Wang, Z. Yang, C. Huang, Z. Zhang, M. Shikh-Bahaei, and M. Chen, “Semantic Feature Multiple Access (SFMA) Over Wireless Networks,” inIEEE INFOCOM 2025 - IEEE Conference on Computer Communi- cations Workshops (INFOCOM WKSHPS), pp. 1–6, 2025

    2025

  34. [34]

    New Algorithms for Designing Unimodular Sequences With Good Correlation Properties,

    P. Stoica, H. He, and J. Li, “New Algorithms for Designing Unimodular Sequences With Good Correlation Properties,”IEEE Transactions on Signal Processing, vol. 57, no. 4, pp. 1415–1425, 2009

    2009