On Congestion Control and Fair Bandwidth Allocation in the Internet

PhD Thesis Proposal Defence


Title: "On Congestion Control and Fair Bandwidth Allocation in the Internet"

by

Mr. Shan CHEN


Abstract:

The Internet is definitely one of the most important inventions in 
mankind's history. It has remarkably changed our daily life affecting all 
its aspects. Undoubtedly the Internet will continue to thrive and expand 
to reach its full potential. However, the TCP/IP architecture, upon which 
the Internet has been built, has shown its limits. In the layered design 
within the TCP/IP architecture, the Network Layer only provides 
best-effort packet delivery service, while the efficiency and fairness in 
the network are relinquished to the congestion control mechanism 
implemented at the Transport Layer. As the Internet evolves the original 
TCP/IP design faces more and more heterogeneous environments, which are 
far more complex than the network scenario for which it was designed. In 
particular, some of the changes violate some basic assumptions of the 
design. For example, the conventional TCP cannot perform well in networks 
with high bandwidth-delay product or in wireless networks with 
non-congestion-related packet losses; on the other hand, designed for 
packet routing, without any control the IP network is vulnerable to 
misbehaving end systems. Quality of Service (QoS) enhancements on IP 
networks have been proposed for decades and yet the Internet is still not 
there due to the deployment complexity involved with such mechanisms. With 
the current scale of the Internet, any radical changes to the TCP/IP 
protocol stack are very unlikely. Effective yet simple enhancements 
involving modification at only one part in the Internet, the end system or 
the intermediate routers, seem easy to deploy and more practical. This 
thesis proposal focuses on congestion control and fair bandwidth 
allocation issues in the Internet and discusses how the intermediate 
routers or the end systems can be enhanced to make the architecture more 
robust in heterogeneous Internet environments.

More specifically, we first study how the synchronized loss effect affects 
the fairness principle achieved by many newly proposed TCP variants in 
high-speed networks. We then propose a new low-cost AQM scheme, SiFTM, to 
provide better intra- and inter-protocol fairness to support heterogeneous 
transmission protocols. SiFTM can also prevent end users from acting 
intrusively. Finally we proposed a new congestion control protocol to 
achieve improved performance in presence of IEEE 802.11 wireless networks 
along the connection path.


Date:     		Tuesday, 25 November 2008

Time:                   10:30a.m.-12:30p.m.

Venue:                  Room 3501
 			lifts 25-26

Committee Members:      Dr. Brahim Bensaou (Supervisor)
 			Prof. Mounir Hamdi (Chairperson)
                         Dr. Jogesh Muppala
 			Prof. Danny Tsang (ECE)


**** ALL are Welcome ****