EUVIMEDEUVIMED entdecken
Crossref

Long-Term Production-Scale Empirical Characterization of SRT Transport Metrics for Reinforcement Learning-Based Adaptive Streaming Control

Ahmed F. K. Koysha, Ali Güneş, Selçuk Şener, Ferdi Sönmez

Journal of Communications · 2026-04 · 21 (2) · 309-321

DOI: 10.12720/jcm.21.2.309-321 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Communications, 2026-04. Verfasst von Ahmed F. K. Koysha und 3 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

—This study presents the first long-term, production-scale empirical characterization of Secure Reliable Transport (SRT) protocol metrics for Reinforcement Learning (RL)-based adaptive streaming control in low-latency live video applications. Production data spanning 722 hours (approximately 30 days) were collected from an IEEE 802.11ac wireless network, yielding 64,340 aligned multivariate samples. These data were analyzed using statistical methods and machine learning-based anomaly detection. Key findings include: (1) strong positive correlation between average Round-Trip Time (RTT) and jitter ( r = 0.912, 95% CI: 0.909–0.915), indicating coupled latency dynamics suitable for state-space reduction; (2) Isolation Forest detected 4.8× more anomalies than the Interquartile Range (IQR) method (6,425 vs. 1,339), demonstrating superior multivariate pattern recognition; (3) three distinct anomaly regimes were identified: congestion-driven, timing-driven, and bandwidth-limited; (4) a data-driven three-level operational monitoring framework (Green/Yellow/Red) was derived for five SRT metrics using convergent statistical evidence; and (5) comprehensive RL validation demonstrated effective reward signal translation: rule-based (+82%) and Q-learning (+54%) controllers outperformed static baselines at hourly resolution; at 5-minute resolution, Q-learning achieved +138.1% improvement with 98% oscillation reduction; in closed-loop emulation, Q-learning achieved +81.3% improvement with minimal red-zone violations (0.23%). These findings establish empirical foundations for future RL-based SRT controllers, bridging network measurement research and adaptive streaming development. Limitations regarding single-network generalizability and recalibration for heterogeneous environments are discussed.

Zitieren

APA

Koysha, A. F. K., Güneş, A., Şener, S., & Sönmez, F. (2026). Long-Term Production-Scale Empirical Characterization of SRT Transport Metrics for Reinforcement Learning-Based Adaptive Streaming Control. Journal of Communications, 21(2), 309-321. https://doi.org/10.12720/jcm.21.2.309-321

Vancouver

Koysha AFK, Güneş A, Şener S, Sönmez F. Long-Term Production-Scale Empirical Characterization of SRT Transport Metrics for Reinforcement Learning-Based Adaptive Streaming Control. Journal of Communications. 2026;21(2):309-321. doi:10.12720/jcm.21.2.309-321

BibTeX
@article{Koysha2026,
  author = {Koysha, Ahmed F. K. and Güneş, Ali and Şener, Selçuk and Sönmez, Ferdi},
  title = {Long-Term Production-Scale Empirical Characterization of SRT Transport Metrics for Reinforcement Learning-Based Adaptive Streaming Control},
  journal = {Journal of Communications},
  year = {2026},
  volume = {21},
  number = {2},
  pages = {309--321},
  doi = {10.12720/jcm.21.2.309-321}
}
Kostenlos starten →