Tips for my Graduate Students

 

Below is a list of tips that I have compiled over the years that you will find useful in your graduate program here at OSU. Please let me know if you have any questions and feel free to stop by my office at 416 Dreese Labs, or send me email to set up an appointment any time.

Prof. Passino

Web page: http://www.ece.osu.edu/~passino/

Return to home.

When You First Arrive:

Welcome to OSU! Below is a list of things that you should be sure to do soon after your arrival:

1. If you have not already done so, get a compter account on the ECE computers. To do this, simply ask me and I can send an email to site.
2. Ask me about getting keys and an office, and I then request an office and send an email to Mr. Bill Thalgott to approve your getting the keys. You then go see Mr. Thalgott to get the keys (he is in Rm. 015 CL).
3. Get registered for classes and buy books (ask graduate students where the best place is for a good deal and perhaps consider purchasing them off the web).

Get to Know the Group and Department:

Get to know the group of graduate students. Also, see the following web sites:

• Control Research Laboratory (study the faculty web pages to gain an understanding of the research activities of the group).
• Department of Electrical and Computer Engineering (e.g., study the descriptions of the courses to know what is available to you)

But, do not be narrow in your thinking. Get to know the myriad other groups/departments at OSU.

Your Academic Program: Requirements, What Classes to Take:

Early on you should come see me to discuss what classes to take. Note the following:

• It is your responsibility to understand the Graduate Student Handbook for the ECE Dept. and ensure that you meet all the requirements for your degree.
• Questions on registration throughout your academic program, including questions about BUCKEYELINK, should be directed to other graduate students.
• I will help you with scheduling, but you should read all the requirements, figure out what courses you are interested in taking (talk to other graduate students about these also), make sure there are no time conflicts, and generally make up a plan for your academic program (of course one that results in you meeting the requirements of your academic program) before coming to see me about this.

Matlab and C (for people planning on going to graduate school):

• Learn Matlab as soon as possible if you do not already know it (one way to do this is to start solving many of your homework problems with it). Start with the help documents and demonstrations. If you generate plots in Matlab you can use an option under printing the figure (when it is open) to save it as a .eps file, and then the .eps file can be easily incorporated into Latex (via, e.g., the useful \scalefig macro in the shell.tex file given below). The same holds for .pdf files. Also, since we often use superscripts, subscripts, and Greek characters, it is useful to label your plots this way in Matlab. To see how to do this say "help texlabel" in Matlab (Matlab implements some Tex commands for mathematical formulas). For more sophisticated labeling of Matlab-generated .eps files it is recommended that you use Adobe Illustrator which allows you to directly open and edit .eps files; however see the discussion below on IPE.
• For more sophisticated simulations you may want to consider using C since for the same program it is sometimes 10 times faster than Matlab. For the programs written in C it is easy to output the data to a file and plot it in Matlab. For a brief explanation of how to output data from programs and plot it in Matlab, click here.
• For code implementing a wide variety of methods for intelligent control and optimization see my web page.

Word Processing: Latex

While Microsoft Word may be useful for writing simple letters, it is quite substandard when it comes to writing nice looking technical reports that have lots of mathematical equations. The preferred package is Latex. It is therefore important that you learn Latex as soon as possible, and one good approach to do this is to use it to type in a few homework problem solutions just to get started. It is quite helpul when getting started with Latex to keep a few things in mind:

• Ask someone who already knows Latex to sit with you the first time to "hold your hand." This will save you a significant amount of time.
• Buy the book: Leslie Lamport, Latex: User's Guide and Reference Manual, Second Edition, Addison-Wesley, 1994. Well, this is still good advice, I think - but really, there are many good resources on the web (some of which are listed below) or at bookstores.
• Accept that in the beginning there will be a "learning curve." It is not WYSIWYG.
• It is easiest to start from existing files ("shells"). For one that can be used for homework problem solutions, project reports, and journal paper submissions see: shell.tex (but for this you will need mycommands.tex, but this file should be viewed as something that you will build on as you see fit). One very important macro in the mycommands.tex file is \scalefig. Learn it and use it as it will make the placement of figures in your documents much easier (especially considering its robustness to changes in format such as when you go to a two-column format for some publications). Important for talks - make them up directly from your Latex papers: For a shell for a white background but with other nice color capabilities use: seminar_whitebackgrnd.tex (this is the one that I prefer all my students to use for all talks, here at OSU and at conferences). For a shell for a black-and-white seminar talk (for slides, with nice borders) see: seminar_bw_shell.tex and for one that also includes a blue background (not preferred in many cases) see: seminar_color_shell.tex For these you will need the OSU logo: osu-logo.eps (osu-logo-color-rw.eps) and you will need some macros in mycommands.tex (that was mentioned above); the other files that are used by the seminar style file can be found on the standard package at the local site or the CTAN site. Of course, it is probably better to move all the commands in those shells into another file and just \input that for your slides. BUT this use of the "seminar" approach has become dated; hence, if you are just getting started read on...
• The "prosper" package for Latex presentations has nice features not in the seminar style file so it may be of interest. For some examples of its use see: http://wiki.lyx.org/pmwiki.php/Examples/Prosper. For more information see: http://amath.colorado.edu/documentation/LaTeX/prosper/. For some examples of its use see: http://wiki.lyx.org/pmwiki.php/Examples/Prosper. BUT it seems that the most modern/best approach to use if you are just getting started is to use "beamer" (you search on it).
• You will often find it useful to produce .ps and .pdf files from Latex. Note that when you generate your .ps file you ALWAYS need to "include the fonts" for the .pdf file to come out with good resolution (without too large of a file). To do this under Linux:
  1) Use the following command to generate the .ps file.
  
  dvips -tletter -Pamz -Pcmz -o *.ps *.dvi
  
  Note that this can be only used in Linux because the fonts in HP-UX cannot be located.
  
  2) Then use Acrobat distiller 4.0 to change *.ps to a clear .pdf file.
• If you work in Miktex (see below) on the PC, then click here for information on how to include the fonts in that package.
• Putting other file types into Latex: Adobe illustrator or Photoshop is useful for converting many file types (e.g., .gif, .jpg) to .eps files so that they can be easily included in Latex (e.g., via the \scalefig command in "mycommands.tex" above). What about movies? The only way I know to do this is to Latex the presentation, print it to a file (creates the .ps file, and do include the fonts), convert it to a .pdf file using Adobe Distiller, then open the file with Adobe Acrobat. Now, Acrobat allows you to edit the .pdf file in several ways, and one of them is to put in movies (e.g., you leave a blank space in the presentation and insert the movie using Acrobat).
• Use of Latex in labeling figures in Matlab. See this example.

The Latex software is free (or it is shareware, but there are also commercial versions) and there are many helpful web-based resources for it. Some of these are (but I am sure that you can find others):

• For OSU/EE relevant information (e.g., style files for theses and dissertations so that they fit the required Graduate School format): http://www.ece.osu.edu/computing/latex2e/
• For help for Latex beginners, download the manual latex_manual.ps, or see: http://fangorn.ci.tuwien.ac.at/docs/services/latex2e.html, or see the following for help and other resources: http://www.loria.fr/services/tex/english/index.html
• To get Latex for the PC, and for other information on Latex and Tex (on which Latex builds), and for a link to the useful "CTAN" site, see: http://www.tug.org/ One package that most PC people recommend can be obtained at: http://www.miktex.org (many recommend using WinEdt with this package, http://www.winedt.com/).
• For Mac people see (I used to use Oztex for the Mac, and the editor Alpha, but note that at the bottom of that page there is a pointer to access to other Mac goodies - particularly "TexShop" that many people, including me, are now using on the Mac): http://www.esm.psu.edu/mac-tex/
• It is likely that you will be typing lots of mathematical equations, and while you can probably get by on Latex, there is a package called AMS-Latex that can be used to extend Latex's mathematical capabilities with some nice features. To use it, you have to see the above sites to get the package (i.e., you need to get it if you are working on your own PC, but it is already on the EE system). then you use the command \usepackage{amsmath} in the header of your file. To see the user's manual for the AMS-Latex package, to get an idea of what it has to offer, see: amsldoc.ps

Next, note that as you build a bibliographic data base you should use Bibtex for automating the referencing process. To get started, the following is a .bib file of some books in the control systems area: refs.bib To use this file you set a bibliographic style, and use the \cite command in the Latex document. You should share your .bib files with other graduate students to save time typing the bibliographic information into the proper format. Try to use a standard method for the labeling of the references (note that one that is commonly used can be seen in the file refs.bib above). Also, note that there are ways of downloading bibliographic information from search sites (e.g., IEEEXplore or Compandex) and converting that to a .bib format for use with Bibtex then Latex: See http://www.scripps.edu/~cdputnam/software/bibutils.html

Latex, Equations, and Powerpoint:

While the above "beamer" approach coupled with the ability to embed movies in .pdf files (see above) provides an ability to present very nice presentations, there are additional features and reasons (sometimes ridiculous ones) to use MS Powerpoint. A problem, however that typically arises is that you have to use mathematical formulas, ones that you already have in Latex. One solution to this problem is "TexPoint":

http://raw.cs.berkeley.edu/texpoint/index.htm

where the approach is to allow for the using Latex-style syntax for equations directly in Powerpoint.

If you are a Mac person the solution is easy: Use "Latex Equation Editor" (search and download from the web); it converts Latex equations to, for instance, .pdf that can be directly included in Keynote or PowerPoint (and thankfully with fully included fonts so they are not large bit-mapped files as they are for TexPoint above). BUT see below in the discussion on figures for information on IPE. Lately, I have been using LaTeXIT since it can be used as a "service", does color matching, and works very nicely with Keynote.

Drawing Figures/Equations for Reports/Publications:

• A program that is often used is "X-figure" on the HPs; however, now many are switching to IPE below.
• A better (more general) package, however, is Adobe Illustrator (it is on the PC in the 343 lab). It has many more capabilities than X-fig, including the ability to edit Matlab-generated .eps files (e.g., so that you can put all types of mathematical symbols on the plots, arrows to highlight portions of the plots, etc.). Moreover, it produces much higher quality .ps files and hence looks better in print. Keep in mind, however, that in Matlab several plot-labeling capabilities are included (e.g., "texlabel" and other Matlab commands).
• I have learned, however, that on the PC it is better to use something called IPE: http://ipe.compgeom.org/ This appears to be a great solution to the figure problem - it is a simple program that seems to give what we typically need in engineering (e.g., for line drawings), it does not have alot of distracting features that are in Adobe Illustrator, and it has the nice feature that it is simple to put Latex equations into the figure (difficult in Illustrator). MOREOVER, it is possible to use IPE to create .eps files of equations directly that can then be put into PowerPoint. All that said, however, I have not used it so I am not sure about this.

Style/Conventions/Recommendations for Technical Writing:

You should regularly use Strunk and White's "The Elements of Style," 4th edition, Allyn and Bacon, Boston, 2000 (I recommend that you purchase it since it is only about $5). There are certain stylistic standards and conventions for writing technical reports and papers that you should follow at all times; those are given in Strunk and White's book, but a few, especially ones specific to our field, are the following:

• Before you write, construct a detailed outline (often it is best to do it down to the paragraph level). Often, getting the logical organization established is half the battle of writing a clear document.
• Think very carefully about how to produce a concise (and short) document. It is more difficult to write a short document, but a short document that gets to the point is much more valuable (we have a serious problem with the information explosion-do not contribute to it).
• It is often good to print out your first draft, get away from it for a little while, then return to the paper copy and edit it with a pencil, and put the resulting changes in. Practically speaking it is probably not good to use only electronic editing.
• Use a spell-checker regularly (there are spell-checkers for Latex documents that ignore all the codes used).
• Captions of figures go on the bottom of the figure, and captions for tables go on the top. Latex can easily be made to do this. Captions start with the first word with an initial capital letter, and the remainder of the words are not capitalized (unless, of course there is, e.g., an acroymn) and the end of the caption should have a period (this is IEEE style).
• Figures and tables should be placed near the point they are first referenced. You should use an automatic referencing scheme so that if you later add more figures, it autmatically numbers them, and refers to them properly. Latex does this easily, especially via the macro \scalefig that is discussed above.
• You should also use auto-numbering for equations, and refer to them via that number also. This is easy to do in Latex.
• You should use a bibliographic database and auto-referencing (once you have a database in place future references are easy). This is easy to do in Latex as discussed above.
• Do not use contractions in technical writing (e.g., don't).
• Try not to use "first person" (never say "I", not good to say "we" although many people do).
• When using quotes, a period or comma goes within (not after) the second quote.

There are many other issues involved in writing a good technical paper and it is recommended that you pay careful attention to the process of writing and try to improve your writing (afterall, we are judged by what we communicate, in written form, or orally, so your abilities to write and speak clearly are very important). Some ideas on writing technical papers are given in Bernstein's paper listed below under research methodology.

Compose and Post a Web Page for Yourself:

• I recommend that you use a wiki (you figure it out).
• Send me a link after you get it set up so I can link to you.

Join Linkedin

• Sign up, set your profile
• Send me an invite to connect to you.

Research Methodology / Working Effectively:

It is useful for you to think about research methodology before and during your research. To assist you along these lines I would recommend that you see some nice essays on the web page of M. Loui at the Univ. of Illinois and read: D. S. Bernstein, "A Student's Guide to Research," IEEE Contr. Sys. Mag., Vol. 19, pp. 102-108, February 1999. There are lots of self-help books out there, some that focus on how to be a good manager, team-player, etc. A book that focuses on how to help make you work effectively, no matter what job you get, that you should consider reading is: Stephen Covey, The 7 Habits of Highly Effective People, Simon and Schuster, NY 1989

Library Research:

Get to know the Science and Engineering library early, but try to use electronic searches via the web to help with your research. Some useful web sites that have search facilities are (some may ask for username and password):

• The OSU library: http://www.lib.ohio-state.edu/ (e.g., see e-Journals, books, research data bases, and the ability to get books sent via campus mail to your office/mailbox)
• IEEE Xplore, accessible via http://ieeexplore.ieee.org/Xplore/dynhome.jsp is very useful. "Ohiolink" a very useful information source: http://www.ohiolink.edu/ (see links to Inspec, Compendex, PapersFirst, as well as Science Citation Index)
• One particularly useful part of Ohiolink is the ISI Web of Knowledge. This database provides, for examplem, the list of papers which cites a paper. This list is often a good way to start a literature overview.

Reviewing Papers:

During your program you may receive some papers to review that are under consideration for publication at a conference or journal. I will discuss with you how to review papers, but it is also useful if you read: D. S. Bernstein, "Peer Review," IEEE Contr. Sys. Mag., Vol. 20, pp. 8-11, June 2000.

Technical Presentations:

The technical presentations that you give (e.g., seminars, conference talks, thesis defense) are very important (in some ways they define how others view you). It is recommended that you:

• Get a copy of the chapter "Engineering your presentations" that is from D. Beer and D. McMurrey, "A Guide to Writing as an Engineer," John Wiley, NY, 1997. Read it carefully and follow its recommendations.
• Often it will be convenient to convert your latex files into presentations and the seminar/prosper/beamer is very useful for that. You should consider using color, especially for interviews and conference talks, and in the Latex section above it is explained how to do to that. Of course then you will either need to get access to a color printer for making slides, take a .ps or .pdf file to Kinko's (or CopEz) and let them make the color slides, or get a laptop so you can make an electronic presentation.

Get to Know Your Professional Community:

It is very important that you get to know your professional community. You should certainly be a member of IEEE and IEEE Control Systems Society. The "E-LETTER on Systems, Control, and Signal Processing" is a useful electronic newsletter. Some useful sites are:

• The IEEE: http://www.ieee.org/
• The IEEE Control Systems Society (has several key journals, and runs the IEEE Conf. on Decision and Control, so there are pointers here to all the CDCs): http://www.ieeecss.org/
• American Automatic Control Council (runs the American Control Conference, so there are pointers here to all the ACCs): http://www.ece.nwu.edu/~ahaddad/aacc/
• International Federation on Automatic Control (has the journal "Automatica" and runs the IFAC World Congress every three years): http://www.ifac-control.org/
• Systems and control archive at Dallas (has many useful pointers for the control systems researcher): http://scad.utdallas.edu/

Teaching as a Career:

If you are considering being a professor a good book to read is: P. Wankat and F. Oreovicz, "Teaching Engineering", McGraw-Hill, NY, 1993. You might be interested in: "How People Learn". This highly acclaimed work was published by the National Research Council in 2000. (It is available online at http://books.nap.edu/books/0309070368/html/index.html if you want to check it out.) You might find the article: "Landing and academic job: the process and the pitfalls" by J.A. Dantzig useful (however, I do not agree with quite all of his recommendations).

Information for Graduate Students Nearing Completion of their Degree:

Well, I hope that it was fun! After you leave do not forget us here at OSU. Please stay in touch and come back and visit. Below is a list of things that you should be sure to do before you leave OSU.

1. Make sure that you meet all the requirements for your degree and that the graduate school is satisfied with the form of your thesis (if you completed one). Be sure to consult the Graduate Student Handbook to check that you meet all requirements and have filled out all the necessary paperwork.
2. Organize your computer account into subdirectories containing your thesis, and perhaps publications. Make the names for the directories clear and put a "readme.yourlastname" in your root directory explaining what is in each directory and what is in each file. Also, see me about getting your files backed up - this is important!
3. Make sure that you get me a master copy of your thesis (if you completed one) in .pdf format (I do not want paper).
4. Make sure that you return any books that you have borrowed from me.
5. Be sure to return all of your keys to Mr. Bill Thalgott (he is in Rm. 015 CL).

Please let me know if you have any questions about any of these items. Good luck!!!

By: K. Passino, with the help of several past and current graduate students and colleagues, on either the above links and writing, or on the files that are linked to this page (in particular, A. Kwong, J. Spooner, V. Gazi, M. Maggiore, Y. Diao, Y. Liu, H. Nounou, M. Baum, J. Finke, A. Serrani, Ted Pavlic). Last updated: 3/16/13

Of course, this page is continually under construction - if you have ideas please email me! Keep in mind, however, I do not want lots of links. I want a few of the best ones, and brief to-the-point discussion.

Return to home.