On June 19, 2007, Professor Parhami's UCSB ECE website moved to a new location. For an up-to-date version of this page, visit it at the new address: http://www.ece.ucsb.edu/~parhami/rsrch_paper_gdlns.htm
The following is a suggested format for your course-related research or term paper. The general guidelines are followed by sample title pages.
Title page should contain all of the following (see the sample title
Title of paper: max 20 words (make it short, yet descriptive); avoid
Name of author
Telephone number(s) and e-mail address for quick contact in case of
"Research Paper for ECE x", where x is the course
number (252B, 254B, etc.)
Quarter and year (e.g., "Winter 2020")
Body of abstract (summary of contributions and/or results; approx. 1 line per
"Keywords:" followed by 5 to 10 keywords and key phrases describing the content
Give background on the topic (provide context and include references on prior work), justify your interest in the topic, prepare the readers for what they will find in later sections, and summarize (in a few sentences) your main findings and/or contributions. This section must be kept short. If it exceeds, say, 3 pages, you may wish to break it up by including an additional section that covers the necessary details for one or more of the above aspects. Use a descriptive title such as "2. REVIEW OF PRIOR WORK" or "2. NOTABLE APPLICATIONS OF ... " for this section.
Body of the paper (10-20 pp. is considered reasonable) should consist of
sections dealing with various aspects of the investigation as appropriate; e.g.,
theory, applications, design issues, tradeoffs, evaluation, experiments,
comparisons with other methods or approaches. Don't be afraid to compare,
criticize, and generally leave your personal mark on the paper. There is no
general rule, except that subdivisions must be coherent and of reasonable
length. Avoid the extremes of single-paragraph and 10-page sections. For very
long sections, consider dividing up or moving some details to an appendix. If
you present performance or speed-up comparisons for your ECE 254B research
paper, then make sure to read:
Crowl, L.A., "How to Measure, Present, and Compare Parallel Performance", IEEE Parallel & Distributed Technology (now IEEE Concurrency), Vol. 2, No. 1, pp. 9-25, Spring 1994.
Give a brief summary (in a few sentences) of what has been presented and/or accomplished. Emphasize the advantages and disadvantages of the proposed approach, technique, or design. Discuss possible extensions of the work and any interesting/open problem that you can envisage. Like the INTRODUCTION, this section must be fairly short.
Provide complete bibliographic information for each reference (see any paper in IEEE Trans. Computers for examples). As a rule of thumb, citing 5-20 references is reasonable; review or survey-type papers tend to have much more extensive bibliographies and original contributions breaking new ground may have fewer references. However, don't take this rule too seriously.
Figures/Tables If possible, include each figure or table close to where it is first referenced in the text. Figures and tables must be numbered and have descriptive captions. Elements of figures (boxes, curves, axes) and tables (columns and/or rows) must be clearly labeled, with units shown where appropriate. Do not copy/paste figures or tables from books, journals, or conference papers.
I will keep all submitted papers and will return to you my comments on a
separate sheet or in an e-mail message. Please make a copy for yourself before
submitting. Plagiarism, in any form (outright copying or failure to properly
credit ideas), will not be tolerated and will result in a failing grade.
There are many good references to help you with research methodology and technical writing. The following book, which covers writing techniques, research methods, refereeing, and oral presentations is particularly recommended.
Zobel, J., Writing for Computer Science, Springer, 2nd ed., 2004. [T11.Z62 2004]
Return to: Top of this page
Sample title page #1 for a research paper:
A Procedure for Writing A-Level Research Papers
in Computer Arithmetic
(987) 654-3210 Office
(123) 456-7890 Home
Research Paper for ECE 252B
The importance of earning good grades in graduate and upper-level undergraduate courses is well-recognized in academic circles. Although many instructors assign course grades in a seemingly random fashion, numerous studies have shown that the quality of a student’s work does in fact affect the final course grade. In this paper, I present a detailed procedure for submitting successful (A-level) papers in the field of computer arithmetic. The procedure starts with the important step of selecting the paper topic and continues with finding relevant references, taking notes, doing the actual research, organizing the material, writing the paper, and going through the appeals process if the final grade turns out to be unsatisfactory. Experimental results were obtained by asking 80 students to apply the procedure in producing their term papers for ECE 252B over a period of four years. Some 75% of these students earned grades of A+ or A, 15% ended up with A–, and the remaining 10% failed the course because the instructor discovered that they had copied papers published many years ago in certain obscure conferences.Keywords: Cheating, Course grades, Grade-point average, Graduate courses, How-to guide, Research paper, Student performance evaluation, Term paper.
Return to: Top of this page
Sample title page #2 for a research paper:
VLSI Approach to Computational Complexity*
(987) 654-3210 Office
(123) 456-7890 Home
Research Paper for ECE 254B
Advanced Computer Architecture: Parallel Processing
The rapid advances of VLSI and the trend toward the decrease of the geometrical feature size, through the submicron and the subnano to the subpico, and beyond, have dramatically reduced the cost of VLSI circuitry. As a result, many traditionally unsolvable problems can now (or will in the near future) be easily implemented using VLSI technology. In recent years, the cost of VLSI components has decreased exponentially. Hence, the application of an exponential number of processors does not cause any cost increase, and the application of only a polynomial number of processors leads to substantial cost savings. A system having an exponential number of processors will be able to solve many NP-complete problems in polynomial time. Furthermore, plotting processor costs as a function of time reveals that the cost of VLSI processors will soon become negative. Undoubtedly, this trend will create a competition for solving the same problem with more processors. We conclude that with the rapid advances of VLSI technology anything is possible and that the worth of a research paper is directly proportional to the number of processors in the system being proposed.
Keywords: Computational complexity, Concurrency, Massive parallelism, Negative-cost elements, NP-complete problems, Parallel processing, VLSI.
* Copied, with some
modifications, from a piece of the same title by Danny Cohen,
in VLSI Systems and Computations, Computer Science Press, 1981, pp. 124-125.