Project Types

1. Analysis of traces (I can point you to how to obtain different traces)
2. Implementation and simple experiments
3. Simulations/comparisons using ns-2, SSFNet, or other network simulators
4. Measurements on the PlanetLab wide-area testbed and analysis of the results (I can help you obtain a slice)
5. Experiments on the Emulab or DETER testbed and analysis of the results (I can help you obtain an account)
6. Mathematical modeling and analysis

Project Topics

• Traffic measurements and characterization
• Internet topology generation and topology visualization tools
• Instrumentation and data visualization tools for network testbed experiments
• Quantifying Internet topology and protocol policies impact on performance and security
• Topology and performance inference techniques and their applications for routing and congestion control
• IP telephony: implementation and measurements
• Application-level multicast or other overlays: comparisons and scalability
• TCP-friendly congestion control: implementation and measurements
• Quantifying buffer requirements
• Wireless networks: measurements and adaptation techniques
• ............... Any other topic we agree upon!

Students can choose to work alone or in teams of 2-3 (if you would like to form a team of more than 3 members, please discuss it with me first). I will consider the number of members of the team when grading project proposals and reports.

The project proposal is due Sunday, March 11th, 2007 by 9:00 PM. If you are working in a group, only one proposal is to be submitted for the entire group.

The final project report (in the form of a paper including references) is due on Tuesday, May 1st, 2007 at 11:00 AM.

Please e-mail your proposal and then your report as PDF attachments to fahmy@cs. (No MS Word documents, please.)

The project proposal consists of just two parts:

I. Problem, Goals, and Methodology (two to five pages)
What is the problem you will address, and what are your goals from the study?
Why do you believe your work will be interesting or unique?
What will be the main contributions of your work?
How will you know if you partially/fully succeeded?

Describe the methodology you will follow: Are you designing a new system/algorithm? are you analyzing the performance of a previously developed system? are you simulating one or more algorithms? are you comparing the performance of several techniques/algorithms? are you implementing a previously described system? In all these cases, what are the metrics you plan to measure to evaluate the system/algorithm, and what parameters do you plan to vary in your experiments? What is the hypothesis you are trying to validate with your experiments?

II. Annotated Bibliography (two to five pages)
What are the primary papers/RFCs/Internet drafts you are planning to use as a basis of your project? Include a one-two sentence description of what each reference contains, and why it is relevant to your project.

See "Useful Links" for links to some simulation, implementation and measurements tools, and useful manuals and papers that can help with your project.

Please let me know as soon as you can if you will need an account on PlanetLab, Emulab, etc, or if you need any special equipment or software.

Also see Collected advice on research and writing for advice on writing and speaking, and some very funny articles. Also see Writing Technical Articles for more helpful hints on writing your paper/report.