QOE model in cellular networks based on QOS measurements using Neural Network approach
Patikorn Anchuen, Peerapong Uthansakul, Monthippa Uthansakul · 2016
In the past, the Quality Of Service (QOS) was the key performance of every network because it could be really measured and helped engineers to improve the network services. However, QOS parameters reflect only the network performances which do not directly indicate the satisfaction of users. Some users experience a bad service due to crowd accessibility even though the signal strength is still good. Hence, this paper proposes the estimation model of user satisfaction in terms of Quality Of Experience (QOE) by using neural network approach. The input of this model is obtained from the measured QOS parameters. Firstly, the authors determine the correlation of QOE due to the QOS. The algorithm used for analysis is Artificial Neural Networks (ANN) toolbox in Matlab software. Data Collection QOS parameter and Application Mean Opinion Score (AMOS) in each user of different networks were used to learn in neural networks to get the right weights in QOE correlation model. The role of this research is to demonstrate the use of neural network approach to create QOE model in the assessment of user satisfaction resulted from the QOS parameters instead of directly measuring the Mean Opinion Score (MOS) from users.