A User-Centric Method for Wi-Fi QoE Visualization and Prediction Using Smartphone Measurements

Wireless Quality of Experience (QoE) monitoring is essential for identifying and resolving connectivity issues. Traditional approaches rely on dedicated sensors, which provide accurate data but are expensive and difficult to scale. In this study, we propose a user-centric method that leverages smartphones to collect Received Signal Strength Indicator (RSSI) and throughput data. These measurements are spatially interpolated to estimate QoE in unmeasured areas. By applying Inverse Distance Weighting (IDW) and Gaussian Process Regression (GPR), we predict network performance and visualize it as heatmaps. Experiments conducted in a university library demonstrate that GPR outperforms IDW in terms of prediction accuracy. Our approach offers a practical and scalable solution for QoE monitoring in resource-constrained environments.