QoE Inference for Video Services in Home Wi-Fi Networks

The operators strive to enhance home networks by focusing on customer experience. To this end, a notion of the customers quality-of-experience (QoE) is vital which was most of the time neglected in favor of quality-of-service (QoS). In this paper, we aim to close a significant gap between QoS and QoE in home networks by proposing a framework for inferring the customer experience from remotely collected network QoS metrics. We focus on video services (e.g. YouTube application) as the main contributor and generator of the indoor network traffic. A case study is performed where experimentally obtained data set comprising of network and application QoS parameters is obtained under varying conditions (i.e., poor coverage, network overload, contention and interference). Regression analysis is then used to build a predictor for different QoE classes given the network QoS metrics remotely accessible from APs using industry adopted standards (i.e., TR-181). This enables operators to infer specific QoE metric using remotely collected passive network measurement with no knowledge of application specific parameters. We show that the proposed framework achieves accuracy in the range from 85 to 95 percent depending on the QoE class, demonstrating the effectiveness and potential of our approach.