Slashdot Mirror
Cornell University FPGA Class Projects for 2008
Matt writes "The new crop of Cornell University ECE 5760 projects are now online. Some really cool projects, as well as the previous two years' worth of projects." Since it's mid-December, many other schools, too, have either just let out or are about to; can you point to any other online collections of cool technical projects?
They're class projects, they're not supposed to be as awesome as something you'd do in your own time. My project for a similar class at my uni was to build a simple PIC-based LEGO controller, it was nothing special but lots of fun to play with of course, which is the most important thing :)
I took the microcontroller equivalent of this class at Cornell when I was there. (I couldn't fit 576 into my schedule, unfortunately.) I have to say, that despite the weeks of all nighters we put into the projects, Bruce Land's class was the best I've ever had, and it did more to keep my interest in ECE and computers than all of the other CS and ECE classes I took. I literally got sick from working on the project too much, but it was so fun that it was worth it. If you ever want to try your hand at microcontrollers or FPGA's, and don't have much of a background in them, I recommend trying this out. The equipment you need is fairly cheap, the labs are fun, and the knowledge is priceless. There's a lot of toil in the workplace, but remembering this class (and working on similar things on the side) keeps my interest in electronics and programming ticking.
Or the FPGA implementation of a bouncing breasts simulator.
Not all of them are trivial Tristan Rocheleau's lock-in amplifier project is not something I'd expect to see from an undergrad.
Comment removed based on user account deletion
The websites for Georgia Tech's senior design projects can be found here:
http://www.ece.gatech.edu/academic/courses/ece4007/web/index.html
i read about it in a blog once
Looks like they should had made a "web accelerator."
For those that don't know, trepanation is a medical procedure (of dubious value) where a hole is drilled in the skull to relieve pressure, although in this case, someone may have defecated in the opening instead. It's not clear if this is an insult (implying "shit for brains", so to speak), if the poster is concerned about illegal, unlicensed, and unsanitary medical procedures taking place, or if the poster is seeking someone to perform this procedure.
Do you even lift?
These aren't the 'roids you're looking for.
I used the same board during my senior project at the University of Utah. It is a great FPGA with tons of options. Our group was sponsored by Micron and built a testing platform for NAND Flash memory that got us a spot presenting at the 2008 FLASH Memory Summit.
http://courses.ece.uiuc.edu/ece445/?g=Home&p=Projects&c=Featured%20Projects
Includes some crazy stuff like a photographing UAV, a PC-based oscilloscope, and a combination lock brute-forcer.
I'm in this class. I worked on the Speaker Recognition project. It was very hard. Some comments and responses to other posts:
If you have any questions about the class, I'd be happy to answer them.
To answer the poster's question, the opencores.org site provides a wealth of free FPGA hardware designs.
You can find a full list of their projects here.
A disclaimer: i hold a phd degree in physics and am working in research. When i studied, libraries were still the most common way to acquire knowledge, so i am biased.
However, i observe the following thing: AFAIR Wikipedia itself says it is not meant to be a "first source". Wikipedia can give you hints where to search in detail, and for sure that *is* great. However, a citation in a paper or your report serves two purposes:
a) make your work understandable for the reader (being nice to the reader)
b) give credit to the original author (being nice to the original author)
c) make clear what you have done/not done (being nice to yourself by specifically avoiding to be accused of scientific misconduct)
The traditional approach is that general text books should seldom be cited, and if so, very specifically. To me, if a student cites a specific wikipedia page the latter condition is fulfilled. So if a reasearch group somewhere on the world used FPGAS in a certain way, it is fair to cite their works and not an wikipedia article which was written from an enthousiast about an article which cooked the results of that group down in an popular science journal. However i suggest, if the wikipedia version is well written, to insert a sentence in the introductory part of the report like "Technique x using y is now widely researched and review reports and intodudory materials are commonly available [a,c,b]", which [a,b,c] beeing wikipedia, a textbook or something (not that you may put several references in a single citation). If it helped you, it can be mentioned. Dont however mention textbook knowledge which is expected from you and your peers.
The following things should be kepti in mind:
a) anything referring to a standard should carry the standards official publisher in the reference
Bad example: cite http://en.wikipedia.org/wiki/IEEE_802.11 for the standard instead of the standard itself. *However* iff the article on wikipedia contains additional information like ABOUT the standard and you want to mention this informally in a meta-sentence (e.g. "IEEE 801.11 is seen by the broad public as the only WLAN standard [quote to wikipedia]"), then it is for sure allowed.
b) dont fall for the illusion that wikipedia is faster than the scientic journals. i assure you its not. In the subjet i work, wikipedia is at least 4 years behind the *published* knowledge and understanding.
c) Wikipedia tends to be good for general knowledge and bad for specific in depth-knowledge. The theory behind the subject i am researching in mentioned only on the surface, but even the context with some papers from the beginning of the *last* century is missing (i'll add it when i find time).
So all in all: Saying to a student: "start at wikipedia" might be ok. One should also say "but follow the threads".
I just saw the poster presentations from CS 229, Machine Learning, at Stanford. The current batch of projects aren't on line yet, but the ones from previous years are.
The projects were very impressive. A vision-guided autonomous helicopter. A system for separating out instruments and vocals from existing audio. A CAPTCHA solver. De-blurring of out of of focus images. Flower recognition. Recognition of hostile network traffic. And those were just a few of the projects. Machine learning really works now.
Hear, hear. As a guy with an ECE degree who now writes software for a living, I constantly marvel at how different the two disciplines are, and how most "software engineering" strikes me as... not really engineering.
While it's true that it's possible to mathematically prove many pieces of software to be correct (heh, or to mathematically prove them incorrect, as would be the case with most software out there), it's pretty rarely done. To be fair, it's incredibly difficult with most non-trivial programs, of course.
But there's something incredibly satisfying and elegant about having a hardware design that you can prove is correct.
Now, of course, many other things can horribly break that design (yay for analog effects, process deficiencies and defects, etc.), but that's a far cry from "well, it compiles, so it'll probably work." But that's reality for ya.
Xfce: Lighter than some, heavier than others. Just right.