EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Heterogeneity-Aware Dynamic Federated Split Learning with Adaptive Compression (HADFL-AC) Edge–Cloud Inference in IoT Environments

Norah Alrusayni, Asma A. Al-Shargabi

Future Internet · 17.04.2026 · 18 (4) · 213

DOI: 10.3390/fi18040213 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Future Internet, online erschienen am 17.04.2026. Verfasst von Norah Alrusayni und 1 weiteren Autor:innen. Die Arbeit nennt 76 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Qassim University.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

In resource-constrained environments, distributed split learning allows for collaborative training; however, the system suffers from high communication overhead and is sensitive to system heterogeneity. Despite advances in IoT data reduction and distributed learning, existing approaches treat heterogeneity, adaptability, and communication efficiency as separate problems. As a result, the Heterogeneity-Aware Dynamic Federated Split Learning with Adaptive Compression (HADFL-AC) framework is proposed, enabling adaptive adjustment of communication payloads to instantaneous bandwidth conditions during training. This approach distinguishes itself by focusing on feature-representation-level adaptation, offering seamless transitions between linear PCA, nonlinear Tiny Autoencoder (TinyAE), and hybrid PCA–AE compression methods without requiring changes to architecture or retraining. Experiments were conducted using the CIFAR10 and CI=NIC datasets with a lightweight ResNet-18 backbone under Dirichlet-based non-IID data partitioning and fluctuating network scenarios. HADFL-AC achieves significant communication reductions of 80.86% on CIFAR-10 and 77.2% on CINIC-10, as well as significant reductions in training time and energy consumption. In addition, the framework achieved these gains while maintaining competitive performance, reaching 79.58% on CIFAR-10 and exhibiting stable convergence on CINIC-10. Consequently, the results demonstrate that leveraging network heterogeneity as an adaptive signal facilitates efficient and scalable distributed learning while effectively balancing communication efficiency and model accuracy.

Förderung

  • Qassim University

Zitieren

APA

Alrusayni, N., & Al-Shargabi, A. A. (2026). Heterogeneity-Aware Dynamic Federated Split Learning with Adaptive Compression (HADFL-AC) Edge–Cloud Inference in IoT Environments. Future Internet, 18(4), 213. https://doi.org/10.3390/fi18040213

Vancouver

Alrusayni N, Al-Shargabi AA. Heterogeneity-Aware Dynamic Federated Split Learning with Adaptive Compression (HADFL-AC) Edge–Cloud Inference in IoT Environments. Future Internet. 2026;18(4):213. doi:10.3390/fi18040213

BibTeX
@article{Alrusayni2026,
  author = {Alrusayni, Norah and Al-Shargabi, Asma A.},
  title = {Heterogeneity-Aware Dynamic Federated Split Learning with Adaptive Compression (HADFL-AC) Edge–Cloud Inference in IoT Environments},
  journal = {Future Internet},
  year = {2026},
  volume = {18},
  number = {4},
  pages = {213},
  doi = {10.3390/fi18040213}
}
Kostenlos starten →