Accurately predicting end-to-end network latency is essential for enabling reliable task offloading in real-time edge computing applications. This paper introduces a lightweight latency prediction scheme based on rational modelling that uses features such as frame size, arrival rate, and link utilization, eliminating the need for intrusive active probing. The model achieves state-of-the-art prediction accuracy through extensive experiments and 5-fold cross-validation (MAE = 0.0115, R2 = 0.9847) with competitive inference time, offering a substantial trade-off between precision and efficiency compared to traditional regressors and neural networks.
@article{arxiv.2511.02501,
title = {Lightweight Latency Prediction Scheme for Edge Applications: A Rational Modelling Approach},
author = {Mohan Liyanage and Eldiyar Zhantileuov and Ali Kadhum Idrees and Rolf Schuster},
journal= {arXiv preprint arXiv:2511.02501},
year = {2025}
}
Comments
Presented at the ICCS 2025 - 5th International Conference on Computer Systems, Xian, China