EUVIMEDEUVIMED entdecken
Crossref

Edge Intelligence: Integrating AI and Machine Learning in Edge Computing Networks

Shallu Yadav

International Journal of Advanced Research in Science Communication and Technology · 31.05.2026 · 1144

DOI: 10.48175/ijarsct-36098 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Advanced Research in Science Communication and Technology, online erschienen am 31.05.2026. Verfasst von Shallu Yadav. Die Arbeit nennt 40 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The exponential proliferation of Internet of Things (IoT) devices, autonomous systems, and latency-sensitive applications has exposed the structural limitations of the conventional cloud-centric computing paradigm, in which raw data is transmitted to distant data centres for processing. Round-trip communication delays, network congestion, escalating bandwidth costs, and growing concerns over data privacy have rendered pure cloud offloading inadequate for many real-time intelligent services. Edge Intelligence (EI), the convergence of artificial intelligence (AI) and machine learning (ML) with edge computing, addresses these challenges by relocating model training and inference from centralised clouds to resource-constrained nodes positioned close to the data source. This paper presents a comprehensive framework for integrating AI and ML within edge computing networks, organised around a three-tier device–edge–cloud architecture. The proposed approach combines lightweight model optimisation—through structured pruning, post-training quantisation, and knowledge distillation—with an adaptive federated learning scheme that enables collaborative model refinement without exposing raw user data. A simulation-based evaluation was conducted across five representative workloads, including image classification, object detection, anomaly detection, speech recognition, and predictive maintenance. Experimental results demonstrate that the proposed edge intelligence framework reduces end-to-end inference latency by up to 74.6% relative to a cloud-only baseline and by 47.9% relative to a conventional edge-only deployment, while lowering energy consumption per inference by 63% and relative network bandwidth utilisation by 78%. Crucially, these efficiency gains are achieved while preserving classification accuracy within 2.3 percentage points of the uncompressed centralised model even at a 75% compression ratio. The study confirms that carefully engineered edge intelligence offers a scalable, privacy-preserving, and energy-efficient pathway for deploying deep learning in next-generation distributed networks.

Zitieren

APA

Yadav, S. (2026). Edge Intelligence: Integrating AI and Machine Learning in Edge Computing Networks. International Journal of Advanced Research in Science Communication and Technology, 1144. https://doi.org/10.48175/ijarsct-36098

Vancouver

Yadav S. Edge Intelligence: Integrating AI and Machine Learning in Edge Computing Networks. International Journal of Advanced Research in Science Communication and Technology. 2026:1144. doi:10.48175/ijarsct-36098

BibTeX
@article{Yadav2026,
  author = {Yadav, Shallu},
  title = {Edge Intelligence: Integrating AI and Machine Learning in Edge Computing Networks},
  journal = {International Journal of Advanced Research in Science Communication and Technology},
  year = {2026},
  pages = {1144},
  doi = {10.48175/ijarsct-36098}
}
Kostenlos starten →