Advanced Computer and Communications Networks
CS577/EE537
Spring 1997
COURSE PROJECT

NOTE- This description of the project process presumes the ``old'' form of submitting information via a hard copy report. I will discuss in class my thoughts on moving much of this effort to a Web-based effort This course includes a class project. Because of the diversity of the students, the intent is for students to select from a wide spectrum of possible projects. I strongly encourage students to work in groups of at least two people.

Students should select projects that they find interesting and that they are qualified to undertake. The following is a list of possible project types:

1. An in-depth research paper/term paper on some aspect of computer networks not covered in the course. This is the least technical choice in that no programming or building of hardware is required. The paper could be of a survey nature with an extensive literature search or it can an in-depth probe into a specific issue in computer networks.
2. A mathematical treatment of a specific problem in computer networks. This could be a continuation of an algorithm or an analysis of a network performance problem discussed in the literature. For this type of project to be acceptable the final report must demonstrate that students have a very thorough understanding of a specific research problem and have not simply extracted the result from a paper.
3. To build and exercise a computer program which simulates some aspect of computer networks. Prior knowledge of simulation techniques is required for this type of project.
4. To utilize an available simulation tool to do an extensive analysis of some aspect of computer networks. Particularly, I have an undergraduate project working with an ATM simulator which some students may want to employ.
5. The design, implementation and testing of a computer network benchmark program. This is the project that I would like to encourage students to do. If three or four groups complete interesting benchmarks, the goal would be to put them together as a benchmark suite at the end of the course.
6. Some type of network performance study or comparison. This is similiar to the previous idea except that in this case less time will be spent building a tool and more time will be spent on the comparison and analysis.

Other types of network project proposals may be submitted but all proposals must be approved. NOTE: If you have not come up with a good idea for a class project or need help in getting a group partner, please discuss this with me by February 4th.

The project has three due dates:

1. Proposal (Due: February 11, 1997)

   Each project group must submit a typed project proposal. The proposal includes: an explanation of the project including expected outcomes, a description of the work to be carried out, resources needed to do the project, and a discussion of the value of the project related to the objectives of this course.

2. Progress/Design Report (Due: April 1, 1997)

   This report should clearly state the current status of the project. If the project involves building something (e.g. software, hardware or conducting experimental data collection), this report should be a complete design report. If the project involves algorithm analysis or an in-depth investigation of some aspect of computer networks, this report must include a clear discussion of the problem and include the current state of your analysis or investigation. Reports should include complete bibliographies, must be typed and less than 15 pages (not including pages with figures).

   This report will receive a letter grade based on all the standard criteria of a professional technical report (i.e., grammar, writing style, typos/misspellings and content will ALL be considered. For design reports pseudo-code is unacceptable.

   Note: This report can easily be the basis for the final report. The key is to demonstrate that a sufficient amount of work has been done at this point.

3. Final Project and Report (Due: May 6, 1997)

   The final report should be a well-presented technical report discussing your project. If your project is primarily a programming effort, you should explain how the program works, give specific sample runs and analyze the results. You must turn a hard copy of your programming which must conform to standard commenting expectations.

   The final report may include parts of your progress report. The written report should be ten to twenty pages in length. Please turn in your graded progress report with your final report.

To minimize the energy you put into this course, a good strategy is to select a research paper to present which is connected to your course project in some way. For example, a group may choose to do a research on the current state of network management and present a paper which features SNMP.

• About this document ...