作者: ZhiGuo Hu , HongRen Yan , Tao Yan , HaiJun Geng , GuoQing Liu
DOI: 10.1016/J.NEUCOM.2019.12.072
关键词:
摘要: Abstract The quality of experience (QoE) the end-users is a critical criterion measurement in VoIP (Voice over Internet Protocol) systems for technical and commercial purposes. We investigate how service (QoS) influences QoE assesses communication. Our contributions are three-fold. First, impacts QoS on comprehensively analyzed by experimental means an association test method, instead independently studying each parameter. Second, algorithm proposed to integrate effects parameters with spatial or temporal characteristics QoE. Third, we apply machine learning regression algorithms impairments, noise echo impairments nonintrusive voice prediction different network environments. results from numerous experiments show that fairly accurate can be obtained these models. work will achieve more evaluation using parameters, clarify influence IP environments, reliability traffic, provide parameter requirements application runs at desired level.