Ellen Zegura

Hanyang Jiang, Henry Shaowu Yuchi, Elizabeth Belding et al.

Modeling and estimation for spatial data are ubiquitous in real life, frequently appearing in weather forecasting, pollution detection, and agriculture. Spatial data analysis often involves processing datasets of enormous scale. In this work, we focus on large-scale internet-quality open datasets from Ookla. We look into estimating mobile (cellular) internet quality at the scale of a state in the United States. In particular, we aim to conduct estimation based on highly {\it imbalanced} data: Most of the samples are concentrated in limited areas, while very few are available in the rest, posing significant challenges to modeling efforts. We propose a new adaptive kernel regression approach that employs self-tuning kernels to alleviate the adverse effects of data imbalance in this problem. Through comparative experimentation on two distinct mobile network measurement datasets, we demonstrate that the proposed self-tuning kernel regression method produces more accurate predictions, with the potential to be applied in other applications.

Varshika Srinivasavaradhan, Morgan Vigil-Hayes, Ellen Zegura et al.

Characterizing cellular network performance is complex. Current representations of cellular coverage, such as service provider and FCC coverage maps, focus only on the minimal level of available bandwidth (e.g., 35/3Mbps download/upload speed for 5G) and omit critical dimensions of quality: network usability and stability over space and time. Because cellular performance can vary substantially along both dimensions, a more fine-grained characterization is necessary. We introduce Quality of Coverage (QoC), a novel multi-dimensional set of key performance indicators (KPIs) that capture measured temporal and spatial performance quality, usability and stability. To evaluate QoC, we first analyze whether the QoC KPIs accurately reflect expected network behavior at individual locations and across spatially-aggregated regions. Then, we apply QoC to more than 15 million measurements from a production network to evaluate its ability to characterize real-world network behavior. Together, our results demonstrate the need for KPIs that capture the full spectrum of cellular performance and show how QoC enables rigorous evaluation of coverage quality across multiple geographic scales.