Open Access

ARTICLE

Mean Opinion Score Estimation for Mobile Broadband Networks Using Bayesian Networks

by Ayman A. El-Saleh¹, Abdulraqeb Alhammadi^2,*, Ibraheem Shayea³, Azizul Azizan⁴, Wan Haslina Hassan²

1 College of Engineering, A'Sharqiyah University (ASU), Ibra, 400, Oman
2 Communication Systems and Networks Research Lab, Malaysia-Japan International Institute of Technology, Universiti Teknologi Malaysia, Kuala Lumpur, 54100, Malaysia
3 Department of Electronics and Communication Engineering, Faculty of Electrical and Electronics Engineering, Istanbul Technical University (ITU), Istanbul, 34467, Turkey
4 Advanced Informatics Department, Razak Faculty of Technology and Informatics, Universiti Teknologi Malaysia, Kuala Lumpur, 54100, Malaysia

* Corresponding Author: Abdulraqeb Alhammadi. Email:

Computers, Materials & Continua 2022, 72(3), 4571-4587. https://doi.org/10.32604/cmc.2022.024642

Received 25 October 2021; Accepted 25 January 2022; Issue published 21 April 2022

Abstract

Mobile broadband (MBB) networks are expanding rapidly to deliver higher data speeds. The fifth-generation cellular network promises enhanced-MBB with high-speed data rates, low power connectivity, and ultra-low latency video streaming. However, existing cellular networks are unable to perform well due to high latency and low bandwidth, which degrades the performance of various applications. As a result, monitoring and evaluation of the performance of these network-supported services is critical. Mobile network providers optimize and monitor their network performance to ensure the highest quality of service to their end-users. This paper proposes a Bayesian model to estimate the minimum opinion score (MOS) of video streaming services for any particular cellular network. The MOS is the most commonly used metric to assess the quality of experience. The proposed Bayesian model consists of several input data, namely, round-trip time, stalling load, and bite rates. It was examined and evaluated using several test data sizes with various performance metrics. Simulation results show the proposed Bayesian network achieved higher accuracy overall test data sizes than a neural network. The proposed Bayesian network obtained a remarkable overall accuracy of 90.36% and outperformed the neural network.

---

Keywords

---