Help a college student out, please!

Hi, I am a senior Computer Engineering student at Drexel University and I could use the help of my fellow forum members.

Me and 2 other classmates in my team are required to design and build a unique and working "device", software or hardware, that solves a computer/electrical related problem. If it has been done before, then we must improve upon the existing design.

What I could use are some computer related problems that haven't been solved or suggestions for something that woud make life easier. Keep in mind we actually have to build the thing and we are given ~4 months to do so. Also, your ideas don't haven't to be marketable, just interesting and doable.

An example, although already invented and sold, would be a tracking device that uses existing GPS technology to monitor where your loved ones are, here's a link ...would be nice item to rent while vacationing at DisneyLand.

So far my team has 1 idea, feedback is appreciated:

1) A weekly alarm clock that can be programmed to wake you up according to your schedule and not just the same time every morning. Many popular computer software programs allow you to do this, but our idea involves an actual alarm clock (no PC needed)

Thanks greatly for any help you could give my team,

George

P.S: Mods, if this is in the wrong forum then please move it.

You could build a web workstation that doesn't have a hard drive. Hence the OS is unbreakable. Add a usb pen drive to it for saving files. Take an existing Linux OS, like Knoppix and remaster the OS to your own specifications.

It could be a minimal hardware PC and fairly inexpensive.

Arkaine23 said:

You could build a web workstation that doesn't have a hard drive. Hence the OS is unbreakable. Add a usb pen drive to it for saving files. Take an existing Linux OS, like Knoppix and remaster the OS to your own specifications.

It could be a minimal hardware PC and fairly inexpensive.

Click to expand...

If I understand correctly, what you described is a thin client (terminal computer). Such a machine is still vulnerable to network attacks and so is the terminal server. Nice idea, but already developed though

LOL, do a network attack on an alarm clock

My boss has done something really cool. We are avid Wardrivers ( http://www.wardriving.com/about.php ), so he connected his GPS to his laptop (he made an adapter) and he records wardriving GPS locations and uses Microsoft's mapping program to map the points out on a map. It didn't take him too long.

NeoGeo said:

If I understand correctly, what you described is a thin client (terminal computer). Such a machine is still vulnerable to network attacks and so is the terminal server. Nice idea, but already developed though

Click to expand...

Booted from an OS on a CD, most of the filesystem is not writeable hence there's no real way to damage the system. At worst you would reboot and that would fix everything back to the original state.


The users/ home directory exists in ram, but a copy of it can be synced onto a usb pen drive. So in effect the user can create documents and download and save files permanently, and also create customized desktop and application settings that they keep on the pen drive and that are loaded when the system boots.

I know absolutely that this kind of thing has been done, because I've done it. But in putting the work into tweaking the OS and installing your own browsers and software, you really are redsigning an existing product and tayloring in to a more specialized use.

For instance, you could install a nice browser like mozilla firebird. Install the plugin to make it a kiosk (auto-run and maximized all the time). Install flash, jave, and mplayer plugins for it (knoppix doesn't include any of those in its browsers). Alter the desktop and default boot parameters to say use Icewm which has a windows-esque interface and requires less ram than the default KDE.... Well there's a lot of things you could do, those are just a few examples.


It's just an idea. probably would be hard to get done if your team has no understanding of Linux, though not impossible as there are nice guides and forum posts that tell you how to do it.

This is probably a more involved project than you would want to consider, but maybe food for though...

When I did my senior project in computer engineering we worked on "smart" traffic signal (i.e. roadsigns that send data to you). One portion we couldn't implement was a very meaningful user-interface. I wanted to do a HUD (heads-up-display), but the technology wasn't very accessible.

Fast forward two years, and you have products like this appearing on the market: http://www6.tomshardware.com/business/20030730/siggraph_2003-10.html

It might be an interesting project to design something similar that lets you project a multi-color image on a windshield without interfering with the driver's basic view of the road. Obviously you wouldn't play movies on this thing, but you could do skeleton overlays of roadmaps, or get some cheap IR sensors that show you how close objects are to every side of you, things like that.

What if you made a program that could relay the IP address when the MAC address was called upon. Sort of like a Lo-Jack for computers. Im thinking more along the lines of a laptop. If you get the IP you can track down atleast where the NIC is and, hopefully the computer is still attached to it. Let me know what you think.

NeoGeo said:

So far my team has 1 idea, feedback is appreciated:

1) A weekly alarm clock that can be programmed to wake you up according to your schedule and not just the same time every morning. Many popular computer software programs allow you to do this, but our idea involves an actual alarm clock (no PC needed)

Click to expand...

I actually did something which was similar for my final year electronics project at school, which was a cat feeder which could maintain three different feeding times per day (although it fed at the same time of day, but it could be re-worked to act as a variable alarm clock).

That said, it used just a PIC, not a full computer-interface, so something like that might not qualify. But certainly, such a thing would be relatively easy to do and take a lot less than 4 months.

How about software you can put on a PDA to remote start your car? That could also, lower/raise windows, lock/unlock doors, etc. That would be sweet.

Or you can have the car alert the PDA like a security alarm.

There are all kinds of things.

That's my suggestion for something that woud make life easier.... and cooler.

thanks

Great, thanks for all these ideas Our advisor wasn't really into the alarm clock idea, so we're looking into other avenues right now. Keep the thoughts coming.

NookieN: Kind of like the simplistic HUD on the upscale Cadillac/Corvette. I'm thinking maybe such a HUD could tie into a GPS receiver and display roadmaps/directions...interesting.

fishy: going to have to research that a bit

KfistoRok: I'm feelin that, an ultra cool James Bond-esque PDA! I know they have car alarms that call your cell phone these days.

Hey Arkaine, do you work at the Inquirer?

A day or two after you posted your idea, I saw this article on the Inq. It's a cool idea, but it's too much Software Engineering and not enough hardware.

NeoGeo said:

I saw this article on the Inq.

Click to expand...

Written by This Senior

How about a voice actuated watch that can start your car or turn on your lights, turn on your coffeemaker or something along those lines, the technology is there.....

Heh, voice activation, my university ex-roomies probably have tales to tell of this wierd guy who shouted strange random words at his computer all the time...

Something I always wanted to build, a backup GPS that can triangulate on FM radio signals and locate you that way. I think all FM radio stations have a digital ID signal, and there is probably a way to get the locations of all transmitters, so you device would have to seek the 3 strongest stations then using known data, possibly requiring some calculations with known output power, and by IDing the stations, it could triangulate your position. Now GPS is probably far more convenient and probably has higher accuracy, but whenever there is some major conflict the military screws up the GPS signals, so such a system as this would have some benefit. Some interesting problems to solve there including how to program the device such that it can still work if a transmitter that would normally be close is off the air, or several transmitters are off the air in case of a local power outage.

anyhoo, good luck.

Road Warrior