CS
Seminar on ITR and Bioinformatics
Fall
2003 Schedule
Wednesday,
Note:
refreshment will be provided, time and room for Bioinformatics seminar may be
changed
Sept. 24 ITR
Super-Resolution Image Reconstruction, a technical Review
by
Oct. 1 Bioinformatics
8 ITR
Position
Estimation in Wireless Sensor Networks
by Dr. Yi Shang, CS Dept.
15 Bioinformatics
22 ITR
I-DEFORM:
Interactive, Dynamic, Efficient Flow for Object Rendering and Modeling
29 Bioinformatics
Nov. 5 ITR
Sensor Network Localization as Parameter Estimation by Dr. Hongchi Shi, CS Dept. (Cancelled, will be rescheduled for Spring 2004)
12 Bioinformatics
19 ITR
Advanced Networking and Internet2 at MU,
by Dr. Gordon Springer,
CS Dept.
26 Thanksgiving,
no meeting
Dec.
3 Bioinformatics
10
ITR TBA (Interpolation or MPEG4-IPMP or ??)
Abstract:
Large-scale networks with
hundreds and even thousands of very small, battery-powered and wirelessly
connected sensor and actuator nodes are becoming a reality. For example, future
sensor networks will involve a very large number of sensor nodes densely
deployed over physical space. In particular, the nodes are typically highly resource-constrained
(processor, memory, and power), have limited communication range, are prone to
failure, and are put together in ad-hoc networks.
It is often useful to know
the geographic positions of nodes in a communications network, but adding GPS receivers
or other sophisticated sensors to every node can be expensive. In this talk,
after a brief introduction to wireless sensor networks, I present MDS-MAP, a
newly proposed localization method based on multidimentional
scaling (MDS). It uses connectivity information---who is within communications
range of whom---to derive the locations of the nodes in the network, and can
take advantage of additional data, such as estimated distances between
neighbors or known positions for certain anchor nodes, if they are available.
Using extensive simulations, we show that the method not only achieves good
performance on relatively uniform layouts, but also performs much better than
previous methods on irregularly-shaped networks.
Abstract:
Deformable models are geometric object models whose
dynamic behaviors are governed by variational
principles and/or partial differential equations (PDEs).
They have proven to be extremely valuable in an increasing number of
applications spanning physical and computational sciences, computer-integrated
engineering, visual computing, data visualization, finite element simulation,
and medical imaging analysis. In this talk, I will present the new PDE-driven,
flow-based, subdivision models which are capable of recovering arbitrary,
complicated shape geometry as well as its unknown topology simultaneously. Our
new deformable models are based on adaptive subdivision geometry, which offers
a potent multi-resolution representation and hierarchical structure, while
their topological flexibility results from the PDEs
associated with the popular level-set approach. After motivating the talk, I
will formulate the mathematics of our new models and demonstrate their
usefulness in many applications (such as surface reconstruction, medical-image
segmentation, iso-surface extraction, and interactive
shape design) for graphics, geometric design, vision, visualization, and
bio-medical imaging through slide shows and video clips.
Abstract:
Sensor networks of hundreds and even
thousands of small, battery-powered, and wirelessly connected sensor and
actuator nodes are becoming a reality. They can be deployed for a wide
variety of applications. They also present many research challenges due
to constraints on their limited computing and communication resources.
One of the research problems in wireless sensor networks is to find the geographic positions of nodes in a sensor network. Adding GPS receivers or other sophisticated sensors to every node can be expensive. A more feasible approach is to use pair-wise measurements such as signal strength and signal arrival time between nodes in the network to localize nodes.
In this talk, I will look at the sensor network
localization problem as a statistical parameter estimation problem and
present the Cramer-Rao lower bounds of
sensor network localization for different types of measurements. I will
then talk about the effect of factors such as anchor placement and anchor
density
on the accuracy of sensor network localization.
Abstract:
This talk will
discuss the evolution of advanced networking activities at MU. Internet2 is an inherent part of these
activities and a history of MU's involvement in the
Internet2 community will be outlined. Current activities and future directions
for the MU Research Network (Rnet) will be provided.
Last Updated :