Multi-Device Task-Oriented Communication via Maximal Coding Rate Reduction
IEEE Transactions on Wireless Communications · 2024-12 · 23 (12) · 18096-18110
DOI: 10.1109/twc.2024.3461336 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Wireless Communications, 2024-12. Verfasst von Chang Cai und 2 weiteren Autor:innen. Die Arbeit wurde bisher 29-mal zitiert (Crossref) und nennt 40 Literaturangaben. Gefördert von: General Research Fund from the Research Grants Council (RGC) of Hong Kong, Area of Excellence Scheme Grant from the RGC of Hong Kong, NSFC/RGC Collaborative Research Scheme from the RGC of Hong Kong.
Abstract (Originalsprache)
In task-oriented communications, most existing work designed the physical-layer communication modules and learning based codecs with distinct objectives: learning is targeted at accurate execution of specific tasks, while communication aims at optimizing conventional communication metrics, such as throughput maximization, delay minimization, or bit error rate minimization. The inconsistency between the design objectives may hinder the exploitation of the full benefits of task-oriented communications. In this paper, we consider a task-oriented multi-device edge inference system over a multiple-input multiple-output (MIMO) multiple-access channel, where the learning (i.e., feature encoding and classification) and communication (i.e., precoding) modules are designed with the same goal of inference accuracy maximization. Instead of end-to-end learning which involves both the task dataset and wireless channel during training, we advocate a separate design of learning and communication to achieve the consistent goal. Specifically, we leverage the maximal coding rate reduction (MCR2) objective as a surrogate to represent the inference accuracy, which allows us to explicitly formulate the precoding optimization problem. We cast valuable insights into this formulation and develop a block coordinate ascent (BCA) algorithm for efficient problem-solving. Moreover, the MCR2 objective serves the loss function for feature encoding and guides the classification design. Simulation results on the synthetic features explain the mechanism of MCR2 precoding at different SNRs. We also validate on the CIFAR-10 and ModelNet10 datasets that the proposed design achieves a better latency-accuracy tradeoff compared to various baselines with inconsistent learning-communication objectives. As such, our work paves the way for further exploration into the synergistic alignment of learning and communication objectives in task-oriented communication systems.
Förderung
- General Research Fund from the Research Grants Council (RGC) of Hong Kong
- Area of Excellence Scheme Grant from the RGC of Hong Kong
- NSFC/RGC Collaborative Research Scheme from the RGC of Hong Kong
- National Natural Science Foundation of China
Zitieren
APA
Cai, C., Yuan, X., & Zhang, Y. A. (2024). Multi-Device Task-Oriented Communication via Maximal Coding Rate Reduction. IEEE Transactions on Wireless Communications, 23(12), 18096-18110. https://doi.org/10.1109/twc.2024.3461336
Vancouver
Cai C, Yuan X, Zhang YA. Multi-Device Task-Oriented Communication via Maximal Coding Rate Reduction. IEEE Transactions on Wireless Communications. 2024;23(12):18096-18110. doi:10.1109/twc.2024.3461336
BibTeX
@article{Cai2024,
author = {Cai, Chang and Yuan, Xiaojun and Zhang, Ying-Jun Angela},
title = {Multi-Device Task-Oriented Communication via Maximal Coding Rate Reduction},
journal = {IEEE Transactions on Wireless Communications},
year = {2024},
volume = {23},
number = {12},
pages = {18096--18110},
doi = {10.1109/twc.2024.3461336}
}
EUVIMED