pith. sign in

arxiv: 2605.00897 · v1 · submitted 2026-04-28 · 📡 eess.SP · cs.CV· eess.IV

SPAT: A Semantic Port-Aware Adaptive-Rate Transmission Protocol for Semantic Communication

Yunhao Wang , Shuai Ma , Bin Shen , Shouhan Shi , Youlong Wu , Guangming Shi , Xiang Cheng This is my paper

Pith reviewed 2026-05-09 21:09 UTC · model grok-4.3

classification 📡 eess.SP cs.CVeess.IV
keywords semantic communicationport-aware transmissionadaptive rate controlsemantic embeddingreconstruction qualitylow latency6G networksimage reconstruction
0
0 comments X

The pith

SPAT embeds port information into semantic representations to enable robust transmission without relying on explicit headers.

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

The paper introduces a new transmission protocol called SPAT for semantic communication systems. It embeds source and destination port details directly into the semantic features rather than using separate headers that can be corrupted. This is paired with separate handling for uplink and downlink traffic and a controller that changes how many semantic channels are sent based on conditions. Tests on image datasets show better reconstruction of meaning compared to standard protocols like TCP and UDP, even when signals are noisy, while keeping transmission quick. If true, this would make semantic approaches more practical for future wireless networks focused on delivering intent instead of every bit.

Core claim

The SPAT protocol jointly embeds source and destination port information into semantic representations to reduce dependence on explicit port headers. It uses a differentiated semantic processing mechanism for uplink service recognition via port identification and downlink selective decoding via destination-aware conditional gating. An adaptive-rate controller dynamically adjusts the number of transmitted semantic channels according to channel conditions and feature importance, resulting in improved robustness and efficiency.

What carries the argument

Semantic port embedding, which folds port details into the semantic data stream itself to support port-aware operations without separate headers, combined with the adaptive-rate controller for dynamic channel selection.

If this is right

  • SPAT achieves higher reconstruction quality than TCP, UDP, and SITP across various signal-to-noise ratios.
  • The protocol maintains low-latency transmission despite the semantic adaptations.
  • Differentiated uplink and downlink mechanisms allow service recognition and selective decoding.
  • Adaptive adjustment of semantic channels improves efficiency under changing conditions.

Where Pith is reading between the lines

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

  • Similar embedding techniques might apply to other types of metadata in semantic systems to reduce overhead.
  • The approach could be tested in multi-user scenarios where port conflicts are more common.
  • Real-world deployment would benefit from integration with existing semantic encoders for end-to-end optimization.
  • If the embedding proves robust, it might reduce the need for error-correcting codes on headers in semantic links.

Load-bearing premise

That the embedded port information stays decodable and useful even when the channel introduces errors that corrupt traditional headers.

What would settle it

Running the protocol over a real channel with high noise levels and finding that port identification fails or reconstruction quality falls below that of TCP.

Figures

Figures reproduced from arXiv: 2605.00897 by Bin Shen, Guangming Shi, Shouhan Shi, Shuai Ma, Xiang Cheng, Youlong Wu, Yunhao Wang.

Figure 1
Figure 1. Figure 1: Comparison between conventional explicit port-header trans [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: The overall architecture of the SPAT-enabled SemCom framework. The proposed SPAT framework embeds port information into [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: The overall architecture of the proposed STAP-based digital semantic communication framework, illustrating the asymmetric processing [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: The network architectures of the proposed Channel ModNet, Joint Port-Payload Encoder, and Port Identification modules. [PITH_FULL_IMAGE:figures/full_fig_p006_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: The network architecture of the proposed Conditional Gating module for destination-aware selective semantic decoding. [PITH_FULL_IMAGE:figures/full_fig_p007_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: The network architecture of the Adaptive-Rate Controller. [PITH_FULL_IMAGE:figures/full_fig_p008_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Adaptive-rate behavior of the proposed SPAT framework under [PITH_FULL_IMAGE:figures/full_fig_p009_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Comparison of latency performance among TCP, UDP, and SPAT with a packet loss rate of 0.25. (a) illustrates the latency PDFs, [PITH_FULL_IMAGE:figures/full_fig_p010_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Performance comparison among SPAT, SITP, TCP, and UDP over AWGN channels on the AFHQ dataset in both uplink and downlink [PITH_FULL_IMAGE:figures/full_fig_p011_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Performance comparison among SPAT, SITP, TCP, and UDP over AWGN channels on the ImageNet10 dataset in both uplink and [PITH_FULL_IMAGE:figures/full_fig_p012_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: The real-world testbed for the proposed SPAT system. [PITH_FULL_IMAGE:figures/full_fig_p012_11.png] view at source ↗
Figure 12
Figure 12. Figure 12: Real-world PSNR comparison of SPAT, SITP, and UDP in [PITH_FULL_IMAGE:figures/full_fig_p013_12.png] view at source ↗
read the original abstract

With the evolution of 6G, semantic communication has emerged as a promising paradigm by prioritizing the delivery of task-relevant meaning over strict bit-level correctness. However, existing transport mechanisms still rely on explicit port headers and bit-level validation, making them vulnerable to header corruption and the resulting packet loss. To address this issue, this paper proposes a Semantic Port-Aware Adaptive-Rate Transmission Protocol (SPAT) for semantic communication. The proposed framework jointly embeds source and destination port information into semantic representations, thereby reducing dependence on explicit port headers while enabling robust port-aware transmission. Furthermore, a differentiated semantic processing mechanism is developed for uplink and downlink scenarios, where port identification is introduced for uplink service recognition and destination-aware conditional gating is designed for downlink selective decoding. In addition, an adaptive-rate controller is incorporated to dynamically adjust the number of transmitted semantic channels according to channel conditions and feature importance, thereby improving both robustness and transmission efficiency. Experimental results on the AFHQ and ImageNet-10 datasets, together with real-world experimental measurements, demonstrate that SPAT consistently outperforms TCP, UDP, and SITP in reconstruction quality across different SNRs while maintaining low-latency transmission.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

2 major / 2 minor

Summary. The paper proposes SPAT, a Semantic Port-Aware Adaptive-Rate Transmission Protocol for semantic communication. It jointly embeds source and destination port information into semantic representations to reduce dependence on explicit headers, introduces differentiated uplink/downlink mechanisms (port identification for uplink, destination-aware conditional gating for downlink), and incorporates an adaptive-rate controller that adjusts the number of transmitted semantic channels based on channel conditions and feature importance. Experiments on AFHQ and ImageNet-10 datasets plus real-world measurements claim that SPAT outperforms TCP, UDP, and SITP in reconstruction quality across SNRs while maintaining low latency.

Significance. If the port-embedding robustness holds and the performance gains are reproducible with proper controls, SPAT could meaningfully advance transport-layer design for semantic communications by mitigating header corruption risks and enabling efficient adaptive transmission. The differentiated uplink/downlink processing and feature-importance-driven rate control represent practical engineering contributions that address real 6G challenges. However, the significance is limited by the absence of verification for the load-bearing assumption that embedded port information survives channel impairments without explicit headers.

major comments (2)
  1. [Experimental Results] Abstract and Experimental Results section: The central claim of consistent outperformance in reconstruction quality relies on the assumption that jointly embedded port information remains reliably decodable after transmission. No ablation study, metric (e.g., port-recovery accuracy vs. SNR), or analysis of port misidentification rates is provided, despite the paper criticizing explicit headers in TCP/UDP as fragile. If semantic-feature corruption causes port errors at moderate SNRs, the protocol reverts to the vulnerabilities it claims to solve and the reported gains cannot be attributed to the port-aware design.
  2. [Proposed Framework] Proposed Framework section: The adaptive-rate controller is described as dynamically adjusting the number of transmitted semantic channels according to channel conditions and feature importance, yet no equations, thresholds, or training details for feature importance are given. It is unclear whether this controller introduces additional free parameters beyond the 'number of transmitted semantic channels' or how it interacts with the port embedding without degrading semantic fidelity.
minor comments (2)
  1. [Experimental Results] The abstract mentions 'real-world experimental measurements' but provides no details on the testbed, hardware, or channel models used; this should be expanded in the Experiments section for reproducibility.
  2. [Proposed Framework] Notation for uplink/downlink mechanisms (e.g., 'destination-aware conditional gating') is introduced without a clear diagram or pseudocode; a figure illustrating the data flow would improve clarity.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the thorough and constructive review. The comments highlight important aspects of our claims regarding port embedding robustness and the adaptive-rate controller. We address each major comment below and will incorporate the suggested clarifications and additional analyses in the revised manuscript.

read point-by-point responses

  1. Referee: [Experimental Results] Abstract and Experimental Results section: The central claim of consistent outperformance in reconstruction quality relies on the assumption that jointly embedded port information remains reliably decodable after transmission. No ablation study, metric (e.g., port-recovery accuracy vs. SNR), or analysis of port misidentification rates is provided, despite the paper criticizing explicit headers in TCP/UDP as fragile. If semantic-feature corruption causes port errors at moderate SNRs, the protocol reverts to the vulnerabilities it claims to solve and the reported gains cannot be attributed to the port-aware design.

    Authors: We agree that explicit verification of port recovery is necessary to substantiate the core advantage of SPAT over header-based protocols. In the revised version, we will add a dedicated subsection in the Experimental Results with an ablation study reporting port identification accuracy and misidentification rates as functions of SNR on both AFHQ and ImageNet-10. This will include comparisons under the same channel conditions used for the reconstruction-quality experiments, allowing direct assessment of when the embedded-port mechanism remains effective. revision: yes

  2. Referee: [Proposed Framework] Proposed Framework section: The adaptive-rate controller is described as dynamically adjusting the number of transmitted semantic channels according to channel conditions and feature importance, yet no equations, thresholds, or training details for feature importance are given. It is unclear whether this controller introduces additional free parameters beyond the 'number of transmitted semantic channels' or how it interacts with the port embedding without degrading semantic fidelity.

    Authors: We acknowledge the need for greater technical detail. The revised Proposed Framework section will include the explicit formulation of the adaptive-rate controller, specifying the feature-importance scoring function, the SNR-based thresholds, and the training procedure used to learn importance weights. We will also clarify that the controller operates on the already-embedded semantic features (including port information) and does not introduce new trainable parameters beyond those already present in the semantic encoder; the interaction is designed to preserve semantic fidelity by prioritizing channels with high task relevance. revision: yes

Circularity Check

0 steps flagged

No circularity: protocol design and empirical claims are independent of self-referential inputs

full rationale

The paper describes an engineering protocol (joint port embedding into semantic features, uplink/downlink differentiation, adaptive-rate controller) whose performance claims rest on experimental measurements over AFHQ/ImageNet-10 and real-world channels. No equations, derivations, or fitted parameters are shown in the provided text that reduce the claimed reconstruction-quality gains to tautological re-statements of the design choices themselves. The outperformance versus TCP/UDP/SITP is presented as an external empirical result rather than a prediction forced by the model's own definitions or prior self-citations. The load-bearing assumption about port-embedding robustness is an unverified engineering hypothesis, not a circular reduction. This is the normal non-circular case for a protocol paper whose central content is design plus measurement.

Axiom & Free-Parameter Ledger

1 free parameters · 2 axioms · 0 invented entities

The protocol rests on standard assumptions from semantic communication and wireless channel modeling; no new physical entities are postulated. Free parameters such as the exact mapping from feature importance to channel count and the gating thresholds are introduced by design but not quantified in the abstract.

free parameters (1)
  • number of transmitted semantic channels
    Dynamically chosen by the adaptive-rate controller according to channel conditions and feature importance; exact mapping function and thresholds are design choices.
axioms (2)
  • domain assumption Semantic representations can reliably carry port identification information without explicit headers under typical wireless impairments.
    Invoked when the paper states that joint embedding reduces dependence on explicit port headers.
  • domain assumption Differentiated uplink and downlink processing plus conditional gating will improve robustness and efficiency.
    Central to the differentiated semantic processing mechanism described.

pith-pipeline@v0.9.0 · 5524 in / 1615 out tokens · 27728 ms · 2026-05-09T21:09:53.410550+00:00 · methodology

discussion (0)

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

Reference graph

Works this paper leans on

34 extracted references · 34 canonical work pages

  1. [1]

    The role of digital twin in advancing industrial internet of things: Insights, applications, and future directions,

    J. Chen, Z. Wang, H. Gao, Q. He, J. Mou, and W. Wang, “The role of digital twin in advancing industrial internet of things: Insights, applications, and future directions,”IEEE Internet Things J., 2026

    work page 2026

  2. [2]

    A flexible digital compute-in-memory chip for edge intelligence,

    A. Yan, J. Yan, P. Shen, Y . Fu, E. Zhang, J. Song, Q. Zhang, Z. He, X. Li, Z. Panet al., “A flexible digital compute-in-memory chip for edge intelligence,”Nature, vol. 649, no. 8099, pp. 1165–1171, Jan. 2026

    work page 2026

  3. [3]

    Generative ai empowered network digital twins: Architecture, technologies, and applications,

    T. Li, Q. Long, H. Chai, S. Zhang, F. Jiang, H. Liu, W. Huang, D. Jin, and Y . Li, “Generative ai empowered network digital twins: Architecture, technologies, and applications,”ACM Comput. Surv., vol. 57, no. 6, pp. 1–43, 2025

    work page 2025

  4. [4]

    Itu-r framework for imt-2030: Review and future direction,

    N. Alliance and G. D ¨usseldorf, “Itu-r framework for imt-2030: Review and future direction,” 2024

    work page 2030

  5. [5]

    6g hyper reliable and low-latency communication–requirement analysis and proof of concept,

    T. Tao, Y . Wang, D. Li, Y . Wan, P. Baracca, and A. Wang, “6g hyper reliable and low-latency communication–requirement analysis and proof of concept,” inProc. IEEE 98th Veh. Technol. Conf. (VTC-Fall). IEEE, 2023, pp. 1–5

    work page 2023

  6. [6]

    A survey on semantic communications: Technologies, solutions, applications and challenges,

    Y . Liu, X. Wang, Z. Ning, M. Zhou, L. Guo, and B. Jedari, “A survey on semantic communications: Technologies, solutions, applications and challenges,”Digit. Commun. Netw., vol. 10, no. 3, pp. 528–545, Jun. 2024

    work page 2024

  7. [7]

    Semantic-enabled 6g communication: A task-oriented and privacy-preserving perspective,

    S. Guo, A. Zhang, Y . Wang, C. Feng, and T. Q. Quek, “Semantic-enabled 6g communication: A task-oriented and privacy-preserving perspective,” IEEE Network, 2025

    work page 2025

  8. [8]

    Semantic communication: A survey on research landscape, challenges, and future directions,

    T. M. Getu, G. Kaddoum, and M. Bennis, “Semantic communication: A survey on research landscape, challenges, and future directions,” Proceedings of the IEEE, vol. 112, no. 11, pp. 1649–1685, 2025

    work page 2025

  9. [9]

    A mathematical theory of communication,

    C. E. Shannon, “A mathematical theory of communication,”Bell Syst. Tech. J., vol. 27, no. 3, pp. 379–423, 1948

    work page 1948

  10. [10]

    Swin kansformer-based seman- tic communication systems for wireless image transmission,

    Y . Wang, S. Ma, D. Gao, and G. Shi, “Swin kansformer-based seman- tic communication systems for wireless image transmission,” inProc. IEEE/CIC Int. Conf. Commun. China (ICCC Workshops). IEEE, 2024, pp. 265–270

    work page 2024

  11. [11]

    Noc4sc: A bandwidth-efficient multi-user semantic communication framework for interference-resilient transmission,

    Y . Wang, S. Ma, P. He, D. Gao, G. Shi, and X. Cheng, “Noc4sc: A bandwidth-efficient multi-user semantic communication framework for interference-resilient transmission,”arXiv preprint arXiv:2512.09356, 2025

    work page arXiv 2025

  12. [12]

    Task-oriented explainable semantic communications,

    S. Ma, W. Qiao, Y . Wu, H. Li, G. Shi, D. Gao, Y . Shi, S. Li, and N. Al- Dhahir, “Task-oriented explainable semantic communications,”IEEE Trans. Wireless Commun., vol. 22, no. 12, pp. 9248–9262, Dec. 2023

    work page 2023

  13. [13]

    Semantic communications for image recovery and classification via deep joint source and channel coding,

    Z. Lyu, G. Zhu, J. Xu, B. Ai, and S. Cui, “Semantic communications for image recovery and classification via deep joint source and channel coding,”IEEE Trans. Wireless Commun., vol. 23, no. 8, pp. 8388–8404, Aug. 2024

    work page 2024

  14. [14]

    Deepjscc-1++: Robust and bandwidth-adaptive wireless image transmission,

    C. Bian, Y . Shao, and D. G ¨und¨uz, “Deepjscc-1++: Robust and bandwidth-adaptive wireless image transmission,” inProc. IEEE Global Commun. Conf. (GLOBECOM). IEEE, 2023, pp. 3148–3154

    work page 2023

  15. [15]

    Stochastic geometry-based semantic performance analysis for text semantic communication,

    H. Zhu, K. Zhu, Y . Zhang, and D. Niyato, “Stochastic geometry-based semantic performance analysis for text semantic communication,”IEEE Trans. Commun., vol. 73, no. 9, pp. 8032–8044, 2025

    work page 2025

  16. [16]

    Goal-oriented semantic communication for wireless video transmission via generative ai,

    N. Li, Y . Deng, and D. Niyato, “Goal-oriented semantic communication for wireless video transmission via generative ai,”IEEE Trans. Wireless Commun., 2026

    work page 2026

  17. [17]

    Channel- adaptive cross-modal generative semantic communication for point cloud transmission,

    W. Yang, Z. Xiong, Q. Yang, P. Zhang, R. Tafazolliet al., “Channel- adaptive cross-modal generative semantic communication for point cloud transmission,”IEEE Trans. Cogn. Commun. Netw., 2026

    work page 2026

  18. [18]

    Cross-modal generative semantic communications powered by semantic knowledge base,

    Z. Fang, M. Sun, S. Wang, X. Xu, H. Gao, J. Huang, S. Han, and P. Zhang, “Cross-modal generative semantic communications powered by semantic knowledge base,”IEEE Trans. Netw. Sci. Eng., vol. 13, pp. 5568–5585, 2026

    work page 2026

  19. [19]

    Conquering high packet-loss erasure: Moe swin transformer-based video semantic communication,

    L. Teng, S. Fan, C. Dong, H. Liang, Z. Bao, X. Xu, R. Meng, and P. Zhang, “Conquering high packet-loss erasure: Moe swin transformer- based video semantic communication,”arXiv preprint arXiv:2508.01205, 2025

    work page arXiv 2025

  20. [20]

    Sitp: A high-reliability semantic information transport protocol without retrans- mission for semantic communication,

    Y . Wang, S. Ma, Y . Wu, G. Shi, X. Cheng, Y . Liu, and P. He, “Sitp: A high-reliability semantic information transport protocol without retrans- mission for semantic communication,”arXiv preprint arXiv:2512.09291, 2025

    work page arXiv 2025

  21. [21]

    Tcp in wireless environments: problems and solutions,

    Y . Tian, K. Xu, and N. Ansari, “Tcp in wireless environments: problems and solutions,”IEEE Commun. Mag., vol. 43, no. 3, pp. S27–S32, Mar. 2005

    work page 2005

  22. [22]

    Udt: Udp-based data transfer for high-speed wide area networks,

    Y . Gu and R. L. Grossman, “Udt: Udp-based data transfer for high-speed wide area networks,”Comput. Netw., vol. 51, no. 7, pp. 1777–1799, May 2007

    work page 2007

  23. [23]

    Short-packet communications: Recent advances and research challenges,

    T.-H. Vu, M. Zeng, S. Kim, H. V . Poor, and Q.-V . Pham, “Short-packet communications: Recent advances and research challenges,”IEEE Netw., 2025

    work page 2025

  24. [24]

    Reducing communication overhead in the iot–edge–cloud continuum: A survey on protocols and data reduction strategies,

    D. Krekovi ´c, P. Krivi ´c, I. P. ˇZarko, M. Ku ˇsek, and D. Le-Phuoc, “Reducing communication overhead in the iot–edge–cloud continuum: A survey on protocols and data reduction strategies,”Internet Things, vol. 31, Art. no. 101553, May 2025

    work page 2025

  25. [25]

    Fdlora: Tackling downlink-uplink asymmetry with full-duplex lora gateways,

    S. Yu, X. Xia, Z. Zhang, N. Hou, and Y . Zheng, “Fdlora: Tackling downlink-uplink asymmetry with full-duplex lora gateways,” inProc. ACM Conf. Embedded Netw. Sensor Syst. (SenSys), Nov. 2024, pp. 281– 294

    work page 2024

  26. [26]

    Comparative analysis of tcp and udp protocol performance in sending text messages using chatting applications,

    W. Haryono, “Comparative analysis of tcp and udp protocol performance in sending text messages using chatting applications,”J. Inotera, vol. 10, no. 1, pp. 41-48, Jan. 2025

    work page 2025

  27. [27]

    A taxonomy and survey of sctp research,

    L. Budzisz, J. Garcia, A. Brunstrom, and R. Ferr ´us, “A taxonomy and survey of sctp research,”ACM Comput. Surv., vol. 44, no. 4, Art. no. 18, pp. 1–36, Aug. 2012

    work page 2012

  28. [28]

    Applying pr-sctp to transport sip traffic,

    X. L. Wang and V . C. Leung, “Applying pr-sctp to transport sip traffic,” inProc. IEEE Global Telecommun. Conf. (GLOBECOM), vol. 2. IEEE, Nov. 2005

    work page 2005

  29. [29]

    Larzon, M

    L.-A. Larzon, M. Degermark, and S. Pink,UDP lite for real time multimedia applications. Hewlett-Packard Laboratories, Apr. 1999

    work page 1999

  30. [30]

    A robust image semantic communication system with multi-scale vision transformer,

    X. Peng, Z. Qin, X. Tao, J. Lu, and K. B. Letaief, “A robust image semantic communication system with multi-scale vision transformer,” IEEE J. Sel. Areas Commun., vol. 43, no. 4, pp. 1278–1291, 2025

    work page 2025

  31. [31]

    Task-oriented semantic communication with adaptive semantic reconstruction network,

    Z. Jin, T. Song, W.-K. Jia, W. Zou, and X. Song, “Task-oriented semantic communication with adaptive semantic reconstruction network,”IEEE Internet Things J., 2025

    work page 2025

  32. [32]

    Transmit what you need: task-adaptive semantic communications for visual information,

    J. Park and S. W. Yoon, “Transmit what you need: task-adaptive semantic communications for visual information,”IEEE J. Sel. Areas Commun., 2025

    work page 2025

  33. [33]

    Swinjscc: Taming swin transformer for deep joint source-channel coding,

    K. Yang, S. Wang, J. Dai, X. Qin, K. Niu, and P. Zhang, “Swinjscc: Taming swin transformer for deep joint source-channel coding,”IEEE Trans. Cogn. Commun. Netw., vol. 11, no. 1, pp. 90–104, Feb. 2025

    work page 2025

  34. [34]

    Digital semantic communications: An alternating multi-phase training strategy with mask attack,

    M. Gong, S. Wang, S. Bi, Y . Wu, and L. Qian, “Digital semantic communications: An alternating multi-phase training strategy with mask attack,”IEEE Trans. Wireless Commun., 2025

    work page 2025