부산대학교 도서관

The “Measuring Broadband America” (MBA) program created by the Federal Communications Commission (FCC) recently paused a large-scale thirteen-year study of access network performance in the United States. Throughout the program, continuous active measurement across a diversity of service providers, network technologies, and testing topologies rendered a rich set of observations for examining the evolution of fixed broadband access speed and latency performance. This paper examines the MBA latency survey methodology followed by a rigorous longitudinal analysis of the multi-year data set. Our evaluation compares the corpus of historical MBA network observations to evolving broadband latency performance expectations and measurement interpretations. Following our longitudinal analysis, we discuss opportunities to clarify consumer perception of latency measurement methodology. We conclude the study by discussing how different latency measurement methodologies and descriptive interpretations of these observations may impact consumer understanding of the FCC's proposed broadband label. This study's data products and software artifacts are made available to the research community at https://github.com/UCBoulder/bclear/tree/main.