With the fast deployment of fiber optics as a transmission medium, a speed mismatch problem has been created between transmission and switching. In recent years, the speed of transmission facilities has been improved by several orders of magnitude. In comparison, the progress made in switching seems to be relatively slow. To meet the challenges posed by the increasing transmission speed, laboratories around the world are working on new photonic and fast electronic switching devices. These recent developments will heavily influence the future direction of switching research. But it is universally recognized that advances in switching devices are not enough; new architectures are needed. The component-by-component approach simply cannot make the best use of the new technology. The purpose of this research is to explore novel architectures and design paradigms for constructing high-speed switching systems which can harness the recent advances in the switching technology. Specifically, a new class of switching networks has been identified which possess the following characteristics: self-routing, fault-tolerance, easy path hunt, easy fault-diagnosis and low crosstalk for the directionalcouplers implementation. They are very promising for high speed switching. The research is expected to lead to development of new multicast-multirate electronic and photonic switching architectures and performance models based on this new class of networks.

Agency
National Science Foundation (NSF)
Institute
Division of Computer and Network Systems (CNS)
Type
Standard Grant (Standard)
Application #
9110075
Program Officer
Dwight D. Fisher
Project Start
Project End
Budget Start
1991-08-15
Budget End
1995-07-31
Support Year
Fiscal Year
1991
Total Cost
$69,986
Indirect Cost
Name
Georgia Tech Research Corporation
Department
Type
DUNS #
City
Atlanta
State
GA
Country
United States
Zip Code
30332