Epsilon: Patching Mobile WiFi Networks

Open Wi-Fi networks offer a chance to have ubiquitous, mobile connectivity by opportunistically leveraging previously deployed resources. Open Wi-Fi access points are densely deployed in many cities, offering high bandwidth at no cost to the mobile node. Unfortunately, Wi-Fi networks are riddled with coverage holes, resulting in poor network performance, even if planned for blanket coverage. To back this claim, we present the results of a measurement study of a small city’s Wi-Fi network—both planned and unplanned— using mobile nodes, verified with data collected from a second city. We find that holes can be broadly classified into two categories: (1) permanent holes due to a lack of Wi-Fi coverage; and (2) transient holes that are due to mobility and channel characteristics. We show that these holes have a severe, adverse effect on the performance of network transport protocols. Unfortunately, fixing these holes by adding WiFi base stations is an expensive and difficult process–there is not always the connectivity, power, and legal authority, to place equipment. Instead, by enhancing the network with a broader area, but still unlicensed, backbone channel we can patch holes in connectivity. This broad area network is low-bandwidth, but as we show in this paper, the backbone radio has a multiplicative effect on bandwidth because it keeps the mobile user’s TCP’s congestion window open and preventing retransmission timeouts on the high-bandwidth Wi-Fi channel. This effect comes with no modifications to the TCP protocol or stack, making it a generally deployable solution. Moreover, the low bandwidth radio has low energy consumption, allowing us to cover holes with solar-powered devices. We evaluate the effectiveness of this system, named Epsilon, for improving the performance of TCP/UDP sessions for a wide range of application workloads. We show that Epsilon produces a 2x to 13x improvement in TCP throughput while providing nearly ubiquitous connectivity at low cost.