10/7: Project proposal due
11/4: Progress report due
11/4: In class progress presentation
11/30 and 12/2: In class final presentations
12/11: Final report due
What is a research project? First, it is not a “book report” that summarizes papers or industry directions. Nor is it repeating simulations from papers or computing numerical examples of theoretical results from papers. Lastly, it is not a design report. Instead, as research, your goal is to advance the state of the art. To do so, it is important to define a problem, summarize the state of the art and why it fails to solve the proposed problem, describe your solution to the problem and why it is novel, and present quantitative results demonstrating the effectiveness and limits of your approach.
A typical course project has a story line such as “medium access algorithms that exploit varying channel conditions to improve throughput were proposed in [1-5]. However, prior ignored mobile users’ objectives of minimizing power consumption. In this project, we design a set of experiments to analyze the power consumption of the algorithm in [1-5]. Moreover, we propose a new power-saving extension and demonstrate its advantages via simulations.” In any case, while the project must be research, as a semester project it must be limited in scope.
Each project will have one to several key references from the research literature: a protocol you are studying, an algorithm you are trying to improve, a model you are validating, etc. Beyond that, there might easily be a dozen other related papers that you will cite in your final report.
The key to finding the topic is finding a paper that interests you: medium access, mobility, energy management, etc. The TAs and I will help you locate key papers in your area of interest. You will also find http://scholar.google.com useful for searching yourself. Rice has on-line access to IEEE and ACM journals and conferences so most papers can be downloaded free from a campus computer or VPN. Many papers are also posted on the web and a google search will usually find them. The main conferences are INFOCOM, MobiCom, and SIGCOMM. Journals typically contain extended versions of conference papers several years later and are less useful for your course project. A useful link for a starting point is http://networks.rice.edu.
Each 537 project will have at least one “base paper” serving as background to the algorithm(s) or theoretical result that you are evaluating, comparing, etc. Once you find that paper, it is easier to narrow the project topic. Moreover, the base paper will also point you to related papers to explore further. You are welcome to propose any topic relevant to the course. I’ll also be happy to discuss your ideas and make suggestions.
Project Proposal and Groups
Each group must consist of 2 students. If there are an odd number of students, one group of 3 or 1 will be permitted.
Each group must have a distinct projects: if two groups propose identical or nearly identical topics, we will meet the week after the proposal is due to find two separate topics.
Each group must submit one single page proposal. This proposal must contain the project title, group members, and at least two relevant references meaningfully cited. The main text of the proposal should contain a statement of the problem definition and motivation in a style similar to the introductions of conference proceedings. The proposal must also describe the expected outcomes and the mechanisms by which these outcomes will be achieved. The proposal, progress reports, and all final reports, must be formatted in an 11pt font and according to the IEEE transactions style.
The progress report must must 2-4 pages long and must be formatted in the same two-column format as the final report. It must contain:
- title and project members
- a precisely formulated description of the problem you are working on
- at least three references cited and placed in the context of what you are doing and why
- a description of the work accomplished to date
- at least one quantitative preliminary result and a preliminary finding. The required format for presenting experiments consists of four steps:
- Issue and hypothesis, i.e., the rationale for doing the experiment. Here, it is useful to comment on what is already known before doing the experiment vs. what specifically the experiment is designed to explore.
- Experimental setup. Describe the experiment design (parameters, factors controlled and varied, etc.) used to address the above issue.
- Results. Describe the data. What are the axes, curves, key data points, etc.
- Findings. What is learned from the experiment? What conclusions are drawn that were not available from prior work.
- a detailed description of the key steps/experiments/… to completion
In class, each group will present for 10 minutes. The presentation must contain:
- title and project members
- 1-2 slides background with tutorial-like information on the topic you are studying.
- 1-2 slides on the problem(s) that you are addressing. What are the issues unaddressed by the papers that you will explore?
- 1-2 slides describing a simulation experiment that you have completed. Not simply a list of parameters, but highlight the key issues. What is the hypothesis? What factors are you varying? Why?
- 1-2 slides describing the results. This should include a graph or table with all axes legible and clearly defined. The slide should also have bullet points summarizing the findings/insights. This slide can be animated if space is tight.
- 1-2 slides on future work. What are your main items for the next month?
Each team will have a 25 minute slot followed by questions and answers from myself and the class.
Your presentation will be graded on your effectiveness in conveying both the problem you are trying to solve (what are you doing and why) as well as your solution (what you discovered in the course of the project).
Each group will turn in one term project report that should be:
- Not more than 10 pages long, including figures and tables, in the IEEE 11 point, 2 column format.
- Well organized and without spelling, grammatical, or syntactic errors, and written in a good tech-report style
- With as much introductory material as necessary to allow me to understand the rest of the report (keep the introduction and background to no more than, say 2 pages)
- With as many linkages to the material covered in class and the main thrusts of the course as reasonable
If code that you have written represents a substantial effort of the project for which you wish to receive credit, refer to a URL in your report from which I can download the source for consideration in your grade.
If you have a really hard time squeezing the material you have into 10 pages, please remember that it is better to cover well only part of it (referring to the rest in the text, or putting it in an appendix, or simply omitting it) than all of it in too compact, confusing, or superficial a way.
Your report’s grade will be based on:
- Technical correctness of content
- Understanding of subject matter conveyed
- References cited (at least 5) and understanding of references conveyed
- Depth and scope of study
- Originality and/or synthesis of different sources
- Quality of presentation, including format and style
- Grammar and spelling
As discussed in class, the key to scoring high marks in categories such as understanding, depth, or originality is to demonstrate that you have synthesized the material and problem, and gone well beyond a summary of the referenced papers and repetition of their results. You must articulate what it is you discovered. Phrases such as “our contribution is…”, “we find that, unlike , …”, “our key technique is to study the problem experimentally from the perspective of…”, “using this methodology, we show…” should be prevalent in the introduction. Emphasis should be placed on the explanations of your experimental results, the why’s behind your data, with relationships back to the theory, its assumptions, your performance measures, and your experimental methodology.
|11/30||John Yan and Chia-Yi Yeh||Detecting MAC Misbehavior in 802.11 using Advanced DOMINO under Asymmetric Topology|
|11/30||Boqiang Fan and Shi Su||Multicast Service in Highly-Directional 60 GHz WLANs and Multicasting Strategies for Different Multicast Groups|
|11/30||David Huang and Jonghun Park||Performance Evaluation of Directional Mobile Networks|
|12/2||Jack Wang and Stephen Xia||Over the Air Study of Wireless Denial of Service Attacks|
|12/2||Marissa Levy and Michael Tsehaie||Exploring VLC Network Topologies to Maximize Throughput and Minimize Shadowing|
|12/2||Aria Hasanzade and Neetu Raveendran||Effect of Spatial Reuse on Indoor VLC Systems|
|12/2||Saadiah Ahmed and Harry Gao||Over the Air Evaluation of MAC Layer Greedy Behavior in WLANs|