PNN-based QoE measuring model for video applications over LTE system

Abstract

Users' quality of experience (QoE) is a key factor in the success of video applications over the Long Term Evolution (LTE) networks. Thus, evaluating the QoE of video applications is of tremendous importance in design and optimization of wireless video processing and transmission systems. In this paper, we propose a QoE measuring model for the quality of video applications by using probabilistic neural network (PNN). We conduct a subjective test, in which OPNET modeler is employed to build a system level simulation platform for the wireless network and distortion is added into original video sequences when transmitting on the platform. A subject pool is utilized to evaluate the distorted videos. Based on the subjective test, we create a distorted-video database. PNN is used to train the mapping function between the corresponding parameters and QoE. The results demonstrate the effectiveness of the proposed model and show that it can provide a high correlation rate with human perception.