Qoe Assessment Model For Addis Ababa Lte Video Streaming Service Using Machine Learning Techniques

In today’s connected world, the availability of fast Internet access and penetration ofrnsmart-phones has created an opportunity for the emerging of new telecom services.rnSimilarly, in Ethiopia, the improvement of technology brought a change from therntraditional services to more advanced communication like video streaming service. Tornensure whether the customers have satisfied for a given service or not, capturing userrnQuality of Experience (QoE) is important. Traditionally, Internet Service Providersrn(ISP)s monitor the network performance by collecting network key performance indicatorsrnwithout involving users’ perception. However, user-perceived QoE estimationrnis multidimensional, which is affected by different influencing factors. So, estimatingrnuser-centric QoE based on Network-level QoS (NQoS) remains challenging tasks forrnISPs. Yet, QoE assessment model for video streaming services that map Quality ofrnService (QoS) to QoE concerning users’ perception has not been performed in Ethiopia.rnThis thesis proposes video streaming QoE assessment models using machine learningrntechniques to estimate user-perceived experience in the Long-Term Evolution (LTE)rnnetwork. The model predicts perceived QoE in a Mean Opinion Score (MOS), byrnevaluating NQoS, Application-level QoS (AQoS) and contextually formulated surveyrnquestionnaire. The models take NQoS metrics such as upload bit rate and download bitrnrate in Megabits Per Second (Mb/s), latency and jitter in milliseconds (ms), and packetrnloss in percentage. Content-type and resolution also considered from the applicationrnlevel. Contrary to existing models for QoE prediction, the proposed model gives a goodrnestimation of the perceived quality with a minimum Mean Squared Error (MSE) ofrn7.74%; and Pearson and Spearman correlations of 97.94% and 97.43%, correspondingrnto the measured QoE. The result obtained from the model shows that the averagernMOS value is 2.79, which is below the recommended one. Accordingly, the proposedrnmodel allows ISP to monitor the perceived QoE level accurately.

Get Full Work

Report copyright infringement or plagiarism