Employment/Research Opportunities
There are several projects I am currently working on, for which students
(undergraduate or graduate) are
welcome to join. The immediate outcomes for the student might include
publications, travelling to conferences, senior projects,
masters theses, a Turing award(!), or a simple feeling of satisfaction-:)
The work involves meetings, reading
relevant articles, creative thinking and code implementations.
Experience with any one of Java, Java3D, C, C++, OpenGL is a plus.
Below you can find short summaries of some of
these projects. Please contact me to find out more.
Sensor Networks and Localization (Funded by TUBITAK-106E071)
The wide use of sensor networks in quite different application areas
necessitates research on finding efficient solutions to problems of these
networks such as underlying hardware architectures, network distribution, and
management. Localization problem is that of finding out the physical location of
each sensor node in the network by using limited information coming from the
sensors while respecting the various structural constraints of sensor networks.
A precondition of making meaningful observations of physical phenomena using a
sensor network is to achieve low error localization of the network.
The goal of this project is to find low-error and efficient
localizations of sensor networks, expected to have increased usage in
the near future, while respecting the specific properties and requirements of such
networks. Along these lines, we will propose
new models and algorithms that satisfy these constraints. The models and the algorithms
developed will be implemented as
software and the performances of all different parts of the algorithms will be tested.
one by one and also as a whole.
Robotics Map Building
The problem is that of extracting a spatial model of an environment using
mobile robots. A knowledge of the environment is only useful if
the physical location is known. On the other hand the location information
can be gathered from observations of the landmarks in the environment.
The problem therefore is usully referred to as the SLAM (Simultaneous Localization
and Map Building) problem. Much research is devoted to the problem in literature.
However various interesting versions of the problem are still open.
Visualizing Telecommunication Networks
The problem is that of sensibly
visualizing large-scale information (telecommunication networks
in this case) using limited resources (time, memory, more importantly limited screen).
Many aspects of this general problem still remain unresolved and
efficient solutions are highly desirable.
Visualizing Colorful Graphs
Graphs are useful models to represent relational information. Since they are
commonly used in a wide range of applications, it is important to come up
with algorithms/systems that automatically draw an input graph on the screen.
Many different methods have been suggested in literature depending on the
types of graphs to be visualized. For this project,
given a graph where the vertices are clustered into k disjoint sets V_1,
V_2, ..., V_k (the disjoint sets can be thought of as colors assigned to vertices)
we would like to come up with methods to visualize the graph such that
the color partitioning is nicely visible.