Visualization of Wireless Computer Networks

1
Visualization of Wireless Computer Networks
University of Kansas Geography Department
Dunbar, Matt D., Kansas Applied Remote Sensing
Program and University of Kansas Geography
Department, and Becker, Brett, Information and
Telecommunication Technology Center, University
of Kansas
Project Description Data visualization is one
of the most effective means for conveying
information and concepts. Geographic Information
Systems (GIS) have proven to be a useful tool for
both the display and analysis of geo-spatial
data. While there are many commonly used sources
of GIS data, such as census, topographical, and
landcover information, newly emerging
technologies provide a wide variety of data
sources that can benefit from spatial analysis.
Wireless computer networks (IEEE standard
802.11b) use microwave frequencies to transmit
network traffic, freeing the network user from
traditional wired constraints. At the same time
that these network devices are communicating with
intended network users, data signals are openly
broadcast throughout the physical environment.
By collecting signal strength data sampled from
these networks, we can interpolate the
distribution of the networks signal field in
geographic space. When these network fields are
overlaid upon high-resolution aerial photography,
the resulting visualizations place the extent of
the wireless network signal into the context of
the surrounding environment. These
illustrations highlight a number of important
spatial concepts associated with wireless
networking, including the region of network
coverage and the potential security hazards
associated with wireless transmissions. While
the GIS techniques used for this study are well
established, the results demonstrate a new way of
representing data created by this growing
technology. This research illustrates the power
of visual data representation and the
applicability of GIS to a wide array of
disciplines.
Visualizing Networks with Signal Strength Data
FIGURE 3. A graduated signal-strength field for
the test network, created by an Inverse Distance
Weighted (IDW) interpolation of the point data
shown in FIGURE 2..
FIGURE 2. Point data (green) collected by
walking around an experimental wireless network
transmitter (white cross).
Equipment
Laptop - Sony VAIO The backbone of the scanning
system is a Sony Vaio Laptop. Specs PII 366,
128megs ram, 12.1in. display, 1in. thick,
3.5lbsWireless Network Card - Orinoco
GOLDFunctionally like ethernet card, but signals
are transmitted in the 2.4GHz range instead of
over physical cable. Once a wireless LAN PCMCIA
card is placed into the laptop, we can receive
wireless network signals.Scanning Software -
NetstumblerThe software that used to scan for
wireless networks. As the program identifies
wireless networks it logs each unique signal it
receives, the networks name, received signal
strength and a variety of other information about
the network. Antenna - Orinoco 10dbi
Omni-DirectionalTo boost the wireless signals
detected by Netstumbler, a 10dbi omni antenna was
connected to the wireless network card. An
omni-directional antenna is preferred over a
directional for mapping, as the antenna's
orientation does not effect the received
signal. GPS - Garmin GPS II PlusAfter
connecting a Global Position System (GPS) unit to
the USB port of the laptop, Netstumbler can log
geographic coordinates along with each wireless
network signal it detects. An external antenna
was connected to this unit for more accurate
signal reception.Mobile Setup - Pickup Truck
After loading all of the equipment gear into a
pickup truck, we are able to scan for wireless
networks covering large areas more quickly.


FIGURE 4. Data points collected during a driving
scan of a commercial wireless network. Points
are color coded by signal strength and the
network location is shown with a cross.
FIGURE 5. Network field resulting from the
interpolation of the point data shown in FIGURE 4.
Other Network Visualization Examples
FIGURE 7. This map shows six unique networks
located near downtown Lawrence, Kansas. This
type of image helps to communicate the signal
propagation characteristics of wireless network
devices.
FIGURE 6. Residential network visualization
illustrating this techniques ability to pinpoint
the origin of a network. Note that the strongest
network strength is found directly in front of
the network owners home.
Multiple Access Point Networks
Mapping Wireless Network Point DataPrior to the
visualization of network fields, it is often
useful to map all detected networks directly from
the point file collected by Netstumbler. This
technique has already been used by many others
who work with wireless networks to map network
locations upon a vector road map. Here, we use
aerial photography as a background due to its
visual appeal and easy interpretation.
FIGURE 1. A large collection of wireless
networks detected while driving in Lawrence,
Kansas. Each unique hardware address is indicated
by a different colored point. The left column
shows a subset of these MAC addresses.
FIGURE 8. A single network composed of nine
access points (AP's). Each AP is denoted by an
asterisk. The complete network coverage is shown
by the shaded area while the unique field for
each AP is bounded by its respective color.
FIGURE 9. In this image a single AP is isolated
from the network and a signal strength field is
shown for this device. Security analysis can be
performed on the network by looking at each
device to determine if the signal leak is
acceptable.