Seminar Speaker
Date: Monday, September 20th, 2010, 3:30PM, SB 104 AUD
Title: WiFi Network Discovery via Cross-platform Interference Signatures
Speaker: Guoliang Xing -- Michigan State University (CSE)
Abstract
WiFi networks have enjoyed an unprecedented penetration rate in recent years. However, because of the limited coverage, existing WiFi infrastructure only provides intermittent connectivity for mobile users. Once leaving the current network coverage, WiFi interfaces often actively scan radio spectrum for new WiFi hotspots wasting precious energy of mobile devices. Although several solutions have been proposed for more efficient WiFi discovery, they either require significant modifications to existing network infrastructures or rely on historical information that is not available in unknown environments.
Modern mobile devices are increasingly equipped with multiple radios including WiFi, Bluetooth and ZigBee. Working in the same open 2.4 GHz band, these radio platforms often cause significant interference among each other. In this talk, I will describe a new approach to WiFi discovery that leverages on such cross-platform interference. By analyzing the unique interference signatures generated by WiFi signals, a low-power radio such as Bluetooth or ZigBee can detect the existence of WiFi hotspots. As a result, a mobile device can significantly reduce its power consumption by powering off the WiFi interface when leaving the network coverage and wake it up only when a new WiFi hotspot is detected by the low-power radio. Based on this idea, we have developed a system called ZiFi that utilizes ZigBee radio to detect the interference signatures generated by 802.11 beacons, i.e., periodic management frames broadcasted by WiFi access points. ZiFi employs a novel digital signal processing algorithm called Common Multiple Folding to accurately amplify 802.11 beacons among interference signals received by ZigBee radio. We also design a stochastic signal detector to improve the detection performance of ZiFi. Our extensive experiments on a Linux netbook and a Nokia N73 smartphone show that, under typical settings, ZiFi can detect WiFi hotspots with high accuracy.