With the rapid improvement in the wireless communication technology and the support of leading communication companies, wireless local area networks (WLANs) have found extremely large application areas. Consequently, efficient management and monitoring of WLANs becomes an important aspect of the performance of these networks. In this paper, we characterize the user behavior and network performance in the Georgia Tech Local Area Walkup & Wireless Network (LAWN). More specifically, we analyze the network performance in terms of average user activity, user distribution, mobility, system load, and network evolution over the last 4 years. To the best of our knowledge, this paper is the first research effort focusing on the analysis of the campus-wide wireless networkevolution and the relationship between network performance metrics such as session duration, user bandwidth, and user distribution, and physical properties of the wireless signals such as received signal strength at the access points (APs). Moreover, the vendor distribution of the wireless cards used in Georgia Tech LAWN is presented. This analysis shows that passive analysis of WLANs can reveal important factors effecting the performance of these networks.

Keywords

Wireless Local Area Networks (WLAN), WLAN monitoring, WLAN analysis, Georgia Tech LAWN, User-based analysis, Flow-based analysis.