Robust Determination of Wi-Fi Throughput Tests Being Indicative of Broadband Bottlenecks

The measurement of network speed, specifically broadband speed/throughput as measured by tools like iPerf, has long been used as a key performance indicator for home broadband. Unfortunately, home users rarely have the capability to conduct reliable wired tests, instead being only able to measure using Wi-Fi. Hence, home wireless is often viewed as an unreliable indicator of network speed, leaving home users with little recourse to challenge the quality of broadband speed being delivered. To that end, we seek to answer the extent to which such tests are unreliable and, more importantly, to understand if one can accurately determine if the result was indicative of broadband as a bottleneck or if the measurement was limited by Wi-Fi. In our paper, we demonstrate that such a determination is eminently possible and, moreover, such a determination can be done drawing only on features and groups of features already reported by iPerf. We show through extensive experiments that one can capture the goodness (test was indicative of broadband speeds) or badness (test was not indicative of broadband speeds) with a 92. 5% accuracy, drawing only on the median throughput and interquartile range with second-by-second windowing reported by iPerf. Finally, we show that there is a negative correlation between the ratio of Wi-Fi throughput to Ethernet (broadband) throughput and the interquartile range normalized with its corresponding Wi-Fi throughput.