Project Ideas Need help in finding a suitable project.

I’ve got a 6 month project which commences in March this year. I’m a Telecommunications Engineering Student in the middle of my MSc course.

We have certain areas which we can focus on, one of them is parallel computing, I want to undertake a project which would 1. be complex enough to get me a decent mark and ultimately a pass and 2. have more research and reading involved than actual practical work.

I know that what I’ve stated is quite vague so I can give a bit more info if needed. It is still a little while away but I’d still like to start reading into it more and getting on with it. I have a lot of knowledge on electrical systems and control systems (which I did for my undergrad) but not a lot in parallel computing. Also the reason why I’ve opted this field is because the other fields like fibre optics do not interest me.

a Viterbi decoding algorithm on the GPU?

a realtime 2x2 MIMO system running entirely in the ACOUSTIC spectrum? (set up 2 speakers and 2 microphones in a room and transmit and receive 2 distinct data streams through that system). Any stereo capable sound card should do. As an alternative to data transmission, consider simultaneously transmitting analog audio of Bach and Metallica from two speakers and then try to separate the two audio signals at the receiver as cleanly as possible.

a Turbo or LDPC channel coder running on the GPU?


Firstly thank you for your response, although I have some knowledge in Electronics and in Networking, often there are so many branches that some terms and ideas are way above me. I have heard of the Viterbi decoder but I have no idea of the theory behind it, so I’ll have a read up on it.

The same applies to the LD Parity check, I have heard of it and know it’s been implemented in Satellite TV tramission? (correct me if I’m wrong), again it would need me to do a lot of reading on this parity check.

The idea which seems most appealing is the MIMO one, it’s used by WANs and LANs so it would be of real world use as well, the downside to this would be the practical side, I don’t want to get myself into too much practical stuff.

Once again thanks a lot for the ideas, I’ll have a word with my professor to see what he makes of them.

My suggestions draw heavily from my background in communications engineering, in particular mobile radio communications.