Resources, Uday Khankhoje
Technical Notes
A set of slides on uncertainty quantification in two parts: first half is a tutorial overview of stochastic collocation and generalized polynomial chaos. The second half deals with the application of these methods to the rough surface scattering problem.
A tutorial on a vector/edge element-based formulation of the finite element method in two dimensions.
A brief note on deriving a scalar electromagnetic field integral equation, with connections to forward and inverse problems.
A brief tutorial on the concepts involved in computing band structures of periodic materials.
Research: Getting Started
Interested in research but don't know where to begin?
Start by reading articles at the popular science level. For example, try these: IEEE Spectrum, Physics Today, Physics World, Scientific American, Wired science blogs, or just some curious questions on reddit. Read until you find something interesting.
After having found a few interesting directions, go over to specific research journals that deal with these directions. Here, start by reading the table of contents, and the abstracts of articles that catch your fancy. For example, let's say you found something interesting related to antennas; next, go over to the website of IEEE Transactions on Antennas and Propagation (or any other relevant journal) and browse through the most popular articles. This will give you a good idea of: some open research problems, and the mathematical skills required to solve some of these problems. Now, try to find a relevant faculty member to go discuss your ideas with.
Remember that a research problem is very different from a homework/assignment problem. Research is much more open ended, and there is no solution key!
Finally, to get a flavour of what research is, how it's done, etc., read this popular essay titled You and your research by Richard Hamming. Or, read this delightful blog, titled “The Birth of an Idea”, where scientists talking about the genesis of their ideas.
Research: Writing
LaTeX is the preferred tool for technical writing, i.e., for reports, papers, etc. Two options to get you started:
First install a LaTeX distribution, for example MiKTeX on Windows, or the package texlive on Ubuntu.
Next, a LaTeX aware text editor, such as the cross-platform tool, TeXmaker. If you can't install LaTeX locally, you can use online tools such as Overleaf. These also allow collaborative editing of documents, ideal for group projects and such.
Some documentation to get you started.
If you find yourself writing a lot (technical or non-technical), then you must read the classic “Elements of Style” by Strunk and White. The IEEE Editorial Style Manual also lists essential aspects of technical writing.
Finally, if this is your first experience with scientific writing, do the world a favour by understanding the meaning of plagiarism and avoiding it like the plague.
Research: Presenting
A few tips on making presentations:
Length: A rough guideline is # of slides = presentation time in minutes x 3/4. So, budget approximately 15 slides for a 20 minute talk.
Identify: Make sure the title slide contains your topic, name, date and optionally: roll number, course number. Slide numbers must be visible on all slides.
Visuals: Choose a colour scheme that is easy on the eyes (for example, no blue font on black background), and font types (avoid comic sans kind of fonts, stick to the normal Helvetica/Arial/Calibri) and sizes that are easily readable.
Title: Every slide should have a meaningful title that conveys what that slide is about. Avoid generic titles like “conclusions”, or “observations”. Try to summarize the key idea of the slide in the title itself.
Brevity: Avoid full sentences and too much text in any slide. Convey ideas graphically, and use bullet points / key words. Don't bombard your audience with too much information, whether it is by means of text, graphics, or animations.
Graphics: Wherever possible, use diagrams/figures to explain ideas. Figures must be captioned, and all axes must have readable labels. If you copy a figure from a paper to a presentation slide make sure that the axes labels don't look tiny.
References: Include references to quoted material on the same slide, possibly in smaller font. This includes putting references for any images taken from the web. In short, references must be put for anything that is not your own material.
Delivery: Never read verbatim from your presentation (or a sheet of paper in your hand), and always make eye contact with your audience.
Attention: The first few slides are very important – they should convey the main question/idea/problem that you want to talk about in very clear terms. The beginning of the talk is when you have maximum audience attention. If you loose them now, you've lost them for the rest of the talk.
White board: Do not assume that you will have a white/black board to explain yourself during the talk. Simulate some basic questions before hand and have that material in backup slides. Usually the need for white board arises because point #6 has been ignored.
Review: Once you have made your slides, go over them to refine them. For e.g., can a whole slide of text be replaced by a explanatory figure? Do you have just an equation on a slide, and it's explanation on another? If so, combine the slides. In giving a mock talk to yourself or to a friend, do you find yourself flipping constantly between slides to explain some concept? If so, rearrange the material so that each slide is more or less self-contained. And lastly, take feedback from the audience after your talk is over about what could be improved.
More resources: How to give a talk (that doesn't put the audience to sleep) by Ramesh Raskar, tips by Krishna Jagannathan (EE@IITM), URL1 URL2, URL3, URL4.
Miscellaneous
|