Joshua Gabet Faculty Advisor: Dr. Carl Baum Case Western Reserve UniversityClemson University

1
A New Class of Mobility Models for Wireless
Mobile Communication Networks
Joshua Gabet Faculty
Advisor Dr. Carl Baum Case Western Reserve
University Clemson University
Introduction An overall area of weakness in
research on wireless ad-hoc (peer-to-peer)
networks is in the area of mobility modeling.
Perhaps the most widely used model, called the
random waypoint model, has nodes picking future
destinations, or waypoints, according to a
two-dimensional uniform distribution over a
rectangular region. Nodes move in a straight line
at a fixed velocity from waypoint to waypoint. A
new random velocity (also uniform) is chosen when
a new destination is chosen.
R1000 m
v0,40 m/s c20 s
v20 m/s
N128, R1000 m
v0,40 m/s c20
v20 m/s

• Velocity Model Comparison
• The random waypoint average velocity falls off as
  the velocity is spread. This occurs because of
  the possibility of a small velocity on a long
  route. An exponential timer is steady, because
  its duration is independent of the distance
  between waypoints.

R1000 m
a1, b600 m, c20 s v0,40 m/s, R1000 m
R and T are independent random variables that
define the location of the next waypoint with
respect to the current location. T generated
with the parameter a equal to 0 has no memory of
its previous direction, while one generated with
a0 does. The parameter b defines how large
radius of movement is.
a0, b/R2
a0, b/R0.2
a1, b/R0.2
Future Research Areas Test the model again many
current simulations to ascertain the impact of
the new model amongst a variety of
applications. Generate a method for initializing
the model in steady state.
R1000 m