B.J. Bennington and C.R. Bartel. Wireless Andrew: Experience building a high speed, campus-wide wireless data network. In Proceedings of MOBICOM, September 1997.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....to understand the characteristics of the wireless traffic and wireless medium itself. Such wireless networks use a wired connection also, which is typically used to connect the access points of the 802.11 network to the Internet, for example. A number of measurement studies [1] 6] 7] [3], 5] have examined traffic characteristics in wireless networks. In these studies, the measurements have been conducted on the wired portion of the network. Measurements with a wired connection can provide accurate traffic statistics as seen in that portion of the network. However they are ....

B.J. Bennington and C.R. Bartel. Wireless Andrew: Experience building a high speed, campus-wid e wireless data network. In Proceedings of MOBICOM, September 1997.

....subnet. The 161 covered buildings span 81 subnets, so in many cases a wireless client roaming from one building to another will be forced to obtain a new IP address. Dartmouth chose not to construct a separate campus wide subnet for the wireless network, unlike the Wireless Andrew project [BB97] Dartmouth College has about 5,500 students and 1,215 full time professors. Most of the approximately 4,200 undergraduate students live on campus. Each is required to own a computer. Each year, approximately 1000 undergraduate students enter Dartmouth College, and most purchase a computer ....

....and Figure 40: tcpdump] Total traffic, by TCP or UDP protocol, normalized. 0 10 20 30 40 50 60 70 80 90 100 http unidentified dantz blitzmail ftp netbios ssn ssh snmp afpovertcp instsrv Percent of traffic Outbound their papers discuss the design and deployment of that network [BB97, Hil99, HJ96] Although they hint of plans for a usage study [BB97] there are as yet no published results. Another recent study used data from Bell Mobility s Personal Communications Services (PCS) cellular network to study the characteristics of customers using WAP web browsers on their cell ....

[Article contains additional citation context not shown here]

Bernard J. Bennington and Charles R. Bartel. Wireless Andrew: Experience building a high speed, campus-wide wireless data network. In Proceedings of the Third Annual International Conference on Mobile Computing and Networking, pages 55--65. ACM Press, September 1997.

....18 of sessions were roaming. Total Intra subnet Extra subnet 5 10 15 20 25 Roams per roaming session (truncated) How does traffic vary across APs, and which have most traf fic There were 476 APs installed by the end of the study. The data in this section are based on the 430 APs in the syslog trace and the 451 responding to our SNMP polls. A detailed identification of the busiest APs is perhaps only of internal interest at Dartmouth College, and in any case we examine the related question about the busiest buildings in the next subsec tion. The APs with the most ....

....traffic were from residences. Figure 15 shows the variation in the number of APs active each day. Clearly visible are the weekly cycle, the Thanksgiving holiday, and a general trend to use more APs, as the number of cards increased and as people used the network more. Each day saw between 171 and 352 access points in use, with a median of 292. Otherwise, we found (not shown) that the temporal patterns of active APs follows a pattern similar to the number of active cards shown in Figures 10 and 11. Over the life of the trace, the APs varied widely in the amount of traffic they handled ....

[Article contains additional citation context not shown here]

B. J. Bennington and C. R. Bartel. Wireless Andrew: Experience building a high speed, campus-wide wireless data network. In Mobicom '97, pages 55-65. ACM Press, September 1997.

....subnet. The 161 covered buildings span 81 subnets, so in many cases a wireless client roaming from one building to another will be forced to obtain a new IP address. Dartmouth chose not to construct a separate campus wide subnet for the wireless network, unlike the Wireless Andrew project [3]. Dartmouth College has about 5,500 students and 1,215 full time professors. During Fall 2001 approximately 3,330 undergraduate students lived on campus. Each is required to own a computer. Each year, approximately 1000 undergraduate students enter Dartmouth College, and most purchase a computer ....

....Of their TCP traffic, 46 was http (ours was 53 ) and 18 ssh (ours was 0.8 ) reflecting their computerscientist audience. The Wireless Andrew project at Carnegie Mellon University created the first large WaveLAN installation, and their papers discuss the design and deployment of that network [3, 6, 7]. Although they hint of plans for a usage study [3] there are as yet no published results. Kunz et al. studied customers using WAP web browsers on their cell phones [9] For seven months they used tcpdump to capture packets at the WAP gateway. Unfortunately, they were unable to identify unique ....

[Article contains additional citation context not shown here]

B. J. Bennington and C. R. Bartel. Wireless Andrew: Experience building a high speed, campus-wide wireless data network. In Mobicom '97, pages 55--65. ACM Press, September 1997.

....However, since public wireless LANs have only recently become widely deployed, such workload characterizations are scarce. Initial studies of wireless networks have explored lowlevel error models and RF signal characteristics [5] installation and maintenance issues of a campus wireless network [3], user mobility in a low bandwidth metropolitan area network [18] and user behavior and traffic characteristics in a university department network [19] and, very recently, a college campus [11] In this paper, we extend previous studies by presenting and analyzing user behavior and network ....

....and less on user behavior. For example, researchers at CMU [5] examined their campus wide WaveLAN installation. The focus of their study was on the error model and signal characteristics of the RF environment in the presence of obstacles. Another study of the same campus wireless network [3] described the issues involved in installing and maintaining a large scale wireless LAN and compared its performance to a wired LAN. A joint research effort between CMU and Berkeley [15] proposed a novel method for network measurement and evaluation applicable to wireless networks. The technique, ....

B. J. Bennington and C. R. Bartel. Wireless Andrew: Experience building a high speed, Campus-Wide Wireless Data Network. In Proceedings of ACM MobiCom'97, pages 55--65, August 1997.

....determined that measurements of the network differed significantly from the simulation and model results. Both of these studies focus more on the radio s properties rather than on the behavior of the users of the radios. Other researchers also studied the campus wide WaveLAN installation at CMU [7]. However, this study focuses on installing and managing a wireless network rather than on user behavior. Another related set of papers is joint work from Berkeley and CMU [45] The main paper in this set outlines a method for mobile system measurement and evaluation, based on trace modulation ....

Bennington, B.J. and Bartel, C.R. Wireless Andrew: Experience Building a High Speed, Campus-Wide Wireless Data Network. Proceedings of the Third Annual ACM/IEEE International Conference on Mobile Computing and Networking. August, 1997. p. 55-65.

....[6] This study focuses on characterizing how the WaveLAN radio itself behaves, in terms of the error model and signal characteristics given various physical obstacles, rather than on analyzing user behavior in the network. Other researchers also studied the campus wide WaveLAN installation at CMU [2]. However, this study focuses on installing and managing a wireless network rather than on user behavior. Another related effort is joint work from Berkeley and CMU [11] The researchers outline a method for mobile system measurement and evaluation, based on trace modulation rather than network ....

Bennington, B.J. and Bartel, C.R. Wireless Andrew: Experience Building a High Speed, Campus-Wide Wireless Data Network. Proceedings of the Third Annual ACM/IEEE International Conference on Mobile Computing and Networking. August, 1997. p. 55-65.

No context found.

B.J. Bennington and C.R. Bartel. Wireless Andrew: Experience building a high speed, campus-wide wireless data network. In Proceedings of MOBICOM, September 1997.

No context found.

Bennington, B.J. and C.R. Bartel. "Wireless Andrew: Experience Building a High Speed, Campus-Wide Wireless Data Network," In Proceedings of the 3rd ACM/IEEE International Conference on Mobile Computing and Networking (MobiCom 97), Budapest, Hungary, September 1997. www.cmu.edu/computing/wireless/articles.html

No context found.

B. J. Bennington and C. R. Bartel, "Wireless Andrew: Experience Building a High-Speed Campus-wide Wireless Data Network," Proceedings of MobiCom'97, Budapest, Sept. 26, 1997.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC