Format: Use 1 inch margins (left and right), 1 inch margins (top and bottom), 11 point times font for the main text, and use 10 point courier font for computer code. Use your judgement for other situations (for example indented, italics, and 10 point courier font for quotations). Single space your text. Make the text fully-justified (where the letters are aligned on both the left and right). The text should be in 2-columns, with 3/8 of an inch of space between columns. Your paper should be 4 pages long. Having only 3 pages is fine, as long as the content is good. Anything 6 pages or longer will not be graded.

If you want to use LaTeX, here are two examples with the formatting already done. You can just remove the text that you do not need. You will also need either the ieee.cls class file or the IEEEconf.cls class file. Just download both class files until you decide which one you like.

1. Example final report, journal style .tex file and its output.
2. Example final report, conference style .tex file and its output.

You are allowed to have appendices, as needed. Appendices are mainly for code or mathematical derivations. You do not have to turn in all code used in your experiment; use your best judgement. You may want to include only relevent sections of code. Appendices do not count in the page count. For example, if you have 4 pages of report, you may also turn in an appendix that is as long as you like. The appendix should be separate, with your name(s) on it.

The references must be in the same 2-column format as the rest of your paper.

You can use this PDF example, but follow the instructions below. If you want to use LaTeX, here are directions and an example file you can use as a template.

• Under authors' names, instead of address put the Class name, number, date, and instructor
• Abstract should be no more than 150 words.
  The abstract is a short summary of the paper. If you had to re-state what your paper says in 150 words or less, what would you say?
  
  For a conference paper, most people will read the abstract to see if they find it interesting enough to read the whole paper. This makes a lot of sense if you go to a conference in a topic that interests you, but find that there are 100+ other papers.
  
  By the way, I recommend writing the abstract LAST, since it is easier this way.
• Introduction
• Why your topic is important (convince us!)
• Where is it used? Applications
• What you will talk about/do
• Overview of the rest of your paper (section 2 covers...section 3 presents...)
• Background
• Any relevant and specific info
e.g. hardware statistics, equipment used

• What other people had to say on this topic(s)
(be sure to cite your references, and quote as appropriate)

• What other people did on this topic (or related topics)
• Problems and shortcomings of their work
• How your work is different and better
• Project
• Your approach to the problem
• What you did
• Design
- what you already had (and where it came from)
- what you added/changed
- for parts, include close-up drawings (e.g. Magic screenshots)

• What did/didn't work?
• Include graphs, equations, pictures, etc. as appropriate
• Results
Include relevant observations, measurements, and statistics. For example, for the VLSI Class: Include statistics such as timing information if available by simulation, or if not, your own analysis about critical path, delays, and clock cycles. Be sure to include size information: the total size of the circuit measured (X lambda by Y lambda), and the transistor count.
• Summary
• Try to draw together the intro, background, and project sections.
• How do they all relate together? (They may appear to be disjoint sections to an unfamiliar reader).
• Restate important results
• Conclusions
• What was accomplished / learned
• What you would have done differently
• Future work
• References
• Cite the papers/books that you used
• Anything you found useful
• Include textbooks from class if you want

See the paper summary feedback for useful examples of what to do when writing a technical document.

Things to include in the report:

• Pictures
• Your observations and measurements
• Equations
• Graphs
• Figures
• Simulation, model (E.g. our design took 200 cycles to do task #1. We ran it 10 times, and the time for each run is: 12, 15, 16, 14, 13, 15, .. Therefore, we expect the second task, which is twice as long, to take about 30 sec).

For references, use the following style:

[Strang97] Strang and Nguyen, Wavelets and Filter Banks, Revised Edition, Wellesley-Cambridge Press, 1997.

(journal papers)
[Weeks99] Michael Weeks, Beth Lumetta, Magdy Bayoumi, "The Black Jack Tutor Chip: Dealing From Idea to Silicon," IEEE Potentials, April/May 1999, pages 38-42.

(conference papers)
[Zhang99] Guoqing Zhang, Mike Talley, Wael Badawy, Michael Weeks and Magdy Bayoumi, "A Low Power Prototype for a 3-D Discrete Wavelet Transform Processor," IEEE International Symposium on Circuits and Systems (ISCAS '99), Orlando, Florida, May 30-June 2 1999, pages 80-83.

(webpages)
[Clarke04] Peter Clarke (Silicon Strategies), "Silterra demonstrates 0.13-micron 8-Mbit SRAM", EE Times, http://www.eetimes.com/semi/news/showArticle.jhtml;jsessionid=0JJT0OEQDLM3MQSNDBGCKH0CJUMEKJVN?articleID=54201193, posted November 30, 2004 (5:54 AM EST), accessed November 30, 2004.

ACM style
In the article "Online First: Evolving the ACM Journal Distribution Program," the ACM Publications Board presents the idea that research has moved to a primarily electronic realm, with print publications serving a secondary role [Boisvert07]. This is a big shift in thinking; previously the printed version was the standard, with on-line copies merely a convenient access method. ACM citations will now include a Digital Object Identifier (DOI), a permanent identification code (like an ISBN number) which can be translated to a web address by services like http://doi.acm.org.

Another significant change with the ACM style is to number the pages starting at 1 for all articles. A paper with 10 pages will have page numbers 1-10, regardless of where it appears in the journal. So an article number must be introduced to properly reference the pages of the journal. Also, the date refers to when the article is posted.

Ironically, while reading my print copy of CACM, I tried to find this article on-line, only to realize that I do not have on-line access. Perhaps I do have an account and password buried somewhere. Or maybe I did not pay extra for it, since I do not read articles on-line (at least without printing them first). The print version does not list an article number, nor the DOI, and does not start at page 1. So here is my incomplete "online first" citation in ACM style:

[Boisvert07] Boisvert, R. F., Irwin, M. J., and Rushmeier, H. 2007. Online First: Evolving the ACM Journal Distribution Program. Communications of the ACM, Article unknown (September 2007), 2 pages. (Pages 19-20). DOI = unknown. http://doi.acm.org/unknown.

See the following guide to documenting your sources.