Varshika Srinivasavaradhan

Shivani Kalamadi, Aditya Bej, Sachin Kumar Singh et al.

Despite numerous technological advancements, the digital divide remains a pressing issue affecting millions worldwide. We present a framework for diagnosing internet inequality at the Census Block Group level by pairing approximately 170 million crowdsourced Ookla speed tests (2021--2025) with U.S. Census demographics across six metropolitan regions. After quantifying and correcting for sampling bias, we use Random Forest regression with permutation importance to identify the socio-economic drivers of download speed, upload speed, and latency. Population density dominates all three metrics at the regional level, but this dominance is an artifact of scale: once areas are stratified into density bins, its influence vanishes in medium- and higher-density neighborhoods, revealing that socio-economic conditions are the true differentiators of internet quality in most urban settings. After controlling for density, income and racial composition emerge as the primary drivers, income consistently dictating upload speed and racial composition proving to be a stronger predictor of download speed than either income or education. Our findings demonstrate that internet inequality is locally configured: no single national narrative explains it, and effective policy demands region-specific intervention.

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.