Key result
Transformer-based deep learning outperforms traditional models in forecasting short-term throughput and traffic extremes.
Why the study?
Real-time throughput prediction is needed for network management in RTC applications, but contemporary solutions struggle to predict extreme traffic throughput conditions such as peaks, valleys, and abrupt changes.
A Transformer-based deep learning model using packet-level information improves short-term throughput prediction in real-time communications, particularly for extreme traffic conditions.
May enhance real-time network traffic management; leaves open prospective validation before adoption.
Amidst the thriving advancement of networks, further catalyzed by the COVID-19 pandemic, we have witnessed a marked escalation in the worldwide adoption of Real-Time Communications (RTC) applications. In this context, there is a compelling necessity to cultivate intelligent and robust network infrastructures and technologies. Real-time throughput prediction emerges as a promising candidate for this purpose to foster network observability and provide preemptive functions, supporting advanced system management, e.g., bandwidth allocation and adaptive streaming. Nonetheless, contemporary solutions grapple with predicting extreme conditions in traffic throughput, notably peaks, valleys, and abrupt changes. To address the challenges, we propose a Transformer-based Deep Learning (DL) Neural Network (NN), leveraging solely packet-level information and adopting a multi-task learning paradigm, to predict short-term throughput, with an emphasis on critical values. In particular, our work is grounded in voluminous traffic traces procured from real video-teleconferencing sessions, and we formulate a time-series regression problem, comparing numerous technologies, from an adaptive filter to Machine Learning (ML) and DL approaches. Conclusively, our methodology exhibits superior efficacy, especially in forecasting traffic extremities.
No takes yet. Share an insight, caveat, or question.
Song et al. (2024) studied Real-Time Communications (RTC) traffic. Transformer-based Deep Learning Neural Network vs. Adaptive filter, Machine Learning, and other Deep Learning approaches was evaluated on Short-term throughput prediction, emphasizing critical values (peaks, valleys, abrupt changes). A Transformer-based Deep Learning Neural Network exhibited superior efficacy in forecasting short-term throughput and traffic extremities compared to traditional and other ML/DL approaches.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: