@article{4780,
  author = {Hajar Ait Lamkademe},
  title = {Network Traffic Characterization and QoS Analysis of Cloud-Rendered Augmented Reality Streams: Insights from Packet-Level Statistics, Latency, Jitter, and Loss Metrics},
  journal = {International Journal of Web Applications},
  year = {2026},
  volume = {18},
  number = {3},
  doi = {https://doi.org/10.6025/ijwa/2026/18/3/145-164},
  url = {https://www.dline.info/ijwa/fulltext/v18n3/ijwav18n3_2.pdf},
  abstract = {Cloud-rendered Augmented Reality (AR) and Extended Reality (XR) offload intensive computational tasks
from lightweight head mounted displays to the cloud, enabling high fidelity immersive experiences. However,
this paradigm imposes stringent Quality of Service (QoS) requirements, including ultra low latency, high
throughput, and minimal jitter, to prevent cybersickness and preserve the sense of presence. While high level
network requirements for XR are well documented, there is a distinct lack of empirical, packet level traffic
characterization necessary for accurate network modeling, simulation, and resource allocation in 5G/6G
environments. This paper presents a comprehensive empirical characterization of cloud rendered AR network
traffic. Utilising real world packet level traces from the IEEE DataPort dataset alongside synthetic validation,
we analyse critical traffic features, including packet size distributions, latency, jitter, and packet loss rates.
Our analysis reveals a multimodal, heavy tailed packet size distribution driven by low latency H.264 encoding
and UDP packetization. Furthermore, we demonstrate a near normal latency distribution contrasted with a
right-skewed jitter profile, exhibiting a moderate positive correlation that highlights shared underlying
network congestion causes. Packet loss remains exceptionally low under stable conditions but exhibits
sporadic bursts that threaten video decoding. These statistical insights provide a foundational baseline for
developing realistic AR traffic models, optimizing network slicing, and designing QoS-aware scheduling
policies to sustain the stringent Quality of Experience (QoE) demands of next generation immersive applications.},
}