Presents:Spatial Stochastic Networks and their Applications in
Telecommunication, Cell Biology and Material Sciences
Guest Speaker: Volker Schmidt, Ulm University Professor, Institute of Stochastics
Date: May 16th, 2008
Time: 11:00am - Noon
Location: Room 2512, EBU1, UCSD
Host: Massimo Franceschetti, ECE and Advanced Network Sc
DESCRIPTION/ABSTRACT:
Spatial networks with an irregular geometric structure can be found in various kinds of macroscopic-geographical maps as well as in microscopic image data. Examples are street systems in large cities, or nationwide road systems. The functional properties of these systems, like the distribution of shortest path lengths, significantly depend on their spatial structure. The same is true e.g. for keratin and actin networks in biological cells, which are part of the cyto-skeleton in epithelial tissues. Their architecture was found to regulate the viscoelastic properties of individual cells and influence cell motility. Similarly, the microstructure of the gas diffusion layer (GDL) in polymer electrolyte membrane (PEM) fuel cells, which can be described by networks of fibres, influences the transport processes of gas and water in the GDL and, therefore, the efficiency of the fuel cell.
Models from stochastic geometry are proposed in order to describe the spatial structure of such networks. In particular, we consider so-called random tessellations, i.e., random subdivisions of the space into irregular convex polytopes. Their edges form random geometric graphs in 2D and 3D. These networks of intersecting lines and segments, and irregular point patterns located on them, can be used to analyze e.g. the distribution of shortest path lengths between "customers" and "equipment stations" in telecommunication networks. Another example of application is related with dynamic modifications of the cytoskeleton network architecture in biological cells, which have been linked to cancer progression and metastasis. It turned out that nested and superimposed tessellations, respectively, are appropriate models in this case. Moreover, a network model is proposed to describe the microstructure of the GDL in PEM fuel cells. It consists of a stack of planar (2D) random line tessellations which are dilated with respect to 3D.
Models from stochastic geometry are proposed in order to describe the spatial structure of such networks. In particular, we consider so-called random tessellations, i.e., random subdivisions of the space into irregular convex polytopes. Their edges form random geometric graphs in 2D and 3D. These networks of intersecting lines and segments, and irregular point patterns located on them, can be used to analyze e.g. the distribution of shortest path lengths between "customers" and "equipment stations" in telecommunication networks. Another example of application is related with dynamic modifications of the cytoskeleton network architecture in biological cells, which have been linked to cancer progression and metastasis. It turned out that nested and superimposed tessellations, respectively, are appropriate models in this case. Moreover, a network model is proposed to describe the microstructure of the GDL in PEM fuel cells. It consists of a stack of planar (2D) random line tessellations which are dilated with respect to 3D.
SPEAKER BIO:
Dr. Schmidt is a professor in the Faculty of Mathematics and Economics, and affiliated with the Institute of Stochastics, at Ulm University (Germany), where he has taught since 1992 as a Professor of Applied Probability and Statistics. He received the equivalent of a Ph.D. from Technical University Freiberg in 1988. Dr. Schmidt is the recipient of the Merckle Research Prize (2005) and Ulm University's Cooperation Prize (2007). His research interests are in probability and statistics, in particular stochastic geometry and spatial statistics.
Click here for more information.
Date:
Friday, May 16, 200811:00AM-12:00AM
Add to Calendar
Location:
9500 Gilman DriveLa Jolla, CA 92093
