Final year project-suggestions

Go To Last Post
27 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hi, I am a student in Telecommunications, and after 1 year i am graduating. At the graduation, we need to make something for the diploma."Something", can be a device, or to write a book, or to make the economical part of opening a company ( in domain of telecommunications).

But I want to do something based on "brains", especially on 8 bit ones. And I need some ideas.

I was thinking initially to make something like a smart home, with some sort of communication with a cell phone, or a computer.
But i stay opened for suggestions, just the project need's to have practically application.

I will really appreciate any of your suggestions.

And one more thing, i don't have another source of information other than internet ( I mean, no help from teachers, they know only about logical gates, and that's it), so probably i will ask a lot of questions ...

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:

no help from teachers, they know only about logical gates, and that's it

Yet you are completing a course of study in telecommunications. So then all of your teaching consisted of traditional and solid-state relays, and switching stations, and switchboards. Nothing of fibre lines, multiplexing, etc.

You can put lipstick on a pig, but it is still a pig.

I've never met a pig I didn't like, as long as you have some salt and pepper.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I'd be interested in hearing more about which area of telecomms you are talking about. In this day and age something that "networks" - either wireless (including but not limited to wi-fi) or something that Ethernets (which I guess means TCP/IP in the most part) would seem like obvious, "on topic" choices.

If you do want to pursue home automation than explore "X10" and as part of the research watch that episode of The Big Bang Theory where they turn the table lamp on ;-)

EDIT: Oh great! It's on YouTube: http://www.youtube.com/watch?v=P...

(Penny sums up your average person's reaction to X10!)

PS I guess it's inevitable that any technology "launch" should include "Thus Sprach Zarathustra" ;-)

Last Edited: Sun. Feb 26, 2012 - 03:17 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

How about a project that involves nterfacing with a video decoder chip using the avr as an accessory controller :)

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:

How about a project that involves interfacing with a video decoder chip using the avr as an accessory controller

Here's one I prepared earlier:

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

@theusch Of course we studied about everything, but in manier of blocks, I mean "this is a telephone central, this black box is a combined time-space multiplexer, and here you insert the telephone cable."
I mean, even teachers they never sow what is inside that big box, and at circuit level, they can't explain how is made.
I studied a lot of analogue electronics, like amplifiers, op-amplifiers, radio circuits;a little of digital electronics; a little of networks, but we never but all of this together. Anyway, all what we studied, is base elementary, everything of that i can find on wikipedia.

@clawson Actually, we should be able to do everything, from computer networks, to all sorts of television, to gsm communication, so we are study a little of everything (but as I mention, nothing deeper than wikipedia can provide) even, like 30% of the courses are psihology, philosophy, ethics, economics.

@MarioRivas thanks, sound's promising, i will do a little research in this area, because i know only that there are integrated videoprocesors ( like TDA_something), and they are doing all the job for decoding video signal.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:

@MarioRivas thanks, sound's promising, i will do a little research in this area, because i know only that there are integrated videoprocesors ( like TDA_something), and they are doing all the job for decoding video signal.


In the picture above there is ahuge lump of silicon doing MPEG2/MPEG4(pt10)(/aka H.264 AVC/AAC) with an ARM running the show on board (chips from Conexant/Broadcomm etc). Then there's a lowly 8 bit micro handling the front panel buttons and light and the interface to the IR controller.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

clawson, just for the interess, how many peoples, and how much time it took, for developing such a product?
I understand that this is a complete solution for a commercial product, certainly more than I can achieve in such time (1 year).
What i feel that i will be able to do, is more like only the IR part of that product.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

clawson wrote:
Quote:

@MarioRivas thanks, sound's promising, i will do a little research in this area, because i know only that there are integrated videoprocesors ( like TDA_something), and they are doing all the job for decoding video signal.


In the picture above there is ahuge lump of silicon doing MPEG2/MPEG4(pt10)(/aka H.264 AVC/AAC) with an ARM running the show on board (chips from Conexant/Broadcomm etc). Then there's a lowly 8 bit micro handling the front panel buttons and light and the interface to the IR controller.

Do you have any video decoder recommendations for the hobbiest that are easy to interface?

I was looking at an SAM9M10 for video decoding, but the BFGA package with +300 pins is something difficult to work with..

Not many useful hobbiest projects can be found online involving video decoding.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

What about using a smartphone to interface with an IOIO based system that could be the frontend to your home automation system? http://www.sparkfun.com/products...

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

andrianu wrote:
clawson, just for the interess, how many peoples, and how much time it took, for developing such a product?
I understand that this is a complete solution for a commercial product, certainly more than I can achieve in such time (1 year).
What i feel that i will be able to do, is more like only the IR part of that product.
It took about 50 people two years so call it one hundred man years. This was using engineers with an average 10-15 years experience. It involves 57000 .c source files, more than 100 concurrent tasks with about 50 queues and over 500 semaphores.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

andrianu wrote:
... ( in domain of telecommunications). ... just the project need's to have practically application.
Then solve, or at least an initial possible solution to, someone's need. Smart homes may be a need but smart water or food (production) is definitely a need. An example is Ali from Iran's bakery customer interface and such. Farmers need to know moisture content while caretaking soil and crops.
For the tele part, the Internet of things is up and coming (IPv6, REST, embedded web servers, Ajax, mega128RFA1, ...). Some software engineers use their test servers to send messages (e-mail, SMS, push to web client, etc.) to their phones (iOS, Android); would of saved me a long overnight at work in the late 1980s.
There's likely someone at your university who needs what you can give (win-win and very local at that).
andrianu wrote:
... write a book ...
Likely better off aiding some open project(s) that need documentation, but the book would look good on a CV especially if definitely striving for a PhD.
andrianu wrote:
... i don't have another source of information other than internet ...
You really need a mentor especially for a project. Look and ye shall find (contemplate, reflect, meditate, pray). Professional societies are one possible source; a local engineer who is searching for a passing on of knowledge is another.
andrianu wrote:
I studied a lot of analogue electronics, like amplifiers, op-amplifiers, radio circuits;a little of digital electronics; a little of networks, but we never but all of this together.
Well done! The project will get the together for you.
andrianu wrote:
... 30% of the courses are psihology, philosophy, ethics, economics.
Then you are a more rounded student than I was. It's essential to have one's intelligence be broad and have a start on being deep.

"Dare to be naïve." - Buckminster Fuller

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

MarioRivas wrote:
I was looking at an SAM9M10 for video decoding, but the BFGA package with +300 pins is something difficult to work with.
TI has a Cortex-A in a 4-layer PCB friendly BGA package (via channels). Freescale i.MX233 (ARM9) has a QFP; an older Chumby (or variant) is an off-the-shelf example (sometimes inexpensive at surplus).
OpenCV with BeagleBoard and BeagleBone from video cameras (USB too); PandaBoard may be best for this (Cortex-A9 IIRC). Some video with Freescale/Genesi i.MX5 series (Cortex-A8 ).
The One Laptop Per Child, latest version, is an inexpensive ARMv6 SoC with a camera.
Some of these ARM SoCs come in PoP (stacked BGA-like, SoC+DRAM) so it's usually easier to add to the off-the-shelf main board your custom interface board.

Project - PV-powered wireless web videocam for bird nest watching (that'll push multiple AVRs ;-). Think New York City redtail hawks :shock: ; btw, that's a great documentary video I paid to watch.

"Dare to be naïve." - Buckminster Fuller

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

adrianu wrote:
What i feel that i will be able to do, is more like only the IR part of that product.
Do not start yourself off short. You don't need to do a "finished" user-ready product; think advanced prototype to possibly first beta but more than proof-of-concept. An Agile development process is what one or a team can complete in about 2 to 4 weeks; repeat until no more time, funds, or to do. Key is to keep moving which you obviously know how to do.

"Dare to be naïve." - Buckminster Fuller

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

clawson wrote:
It took about 50 people two years so call it one hundred man years. This was using engineers with an average 10-15 years experience. It involves 57000 .c source files, more than 100 concurrent tasks with about 50 queues and over 500 semaphores.
It doesn't need a robust response Cliff, but....I pretty much hate my Sky HD box. It's mighty clever, and the HD picture quality is amazing, but when it takes 4 seconds to change a channel or respond to remote key presses (yes, pulled the plug/waited the ridiculous 30 seconds/plugged it back in again - accompanied by the zztt as it makes contact [why hasn't it got a proper power switch?], rebuilt the software blah blah blah), doesn't record things you set it to record and breaks completely every 2 years or so and needs to be replaced, then I do wonder why I've paid Mr Murdoch about £8000 in the last 10 years (yes, not all HD years of course)

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

gchapman wrote:
MarioRivas wrote:
I was looking at an SAM9M10 for video decoding, but the BFGA package with +300 pins is something difficult to work with.
TI has a Cortex-A in a 4-layer PCB friendly BGA package (via channels). Freescale i.MX233 (ARM9) has a QFP; an older Chumby (or variant) is an off-the-shelf example (sometimes inexpensive at surplus).
OpenCV with BeagleBoard and BeagleBone from video cameras (USB too); PandaBoard may be best for this (Cortex-A9 IIRC). Some video with Freescale/Genesi i.MX5 series (Cortex-A8 ).
The One Laptop Per Child, latest version, is an inexpensive ARMv6 SoC with a camera.
Some of these ARM SoCs come in PoP (stacked BGA-like, SoC+DRAM) so it's usually easier to add to the off-the-shelf main board your custom interface board.

Project - PV-powered wireless web videocam for bird nest watching (that'll push multiple AVRs ;-). Think New York City redtail hawks :shock: ; btw, that's a great documentary video I paid to watch.

Thank you gchapman!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:

It doesn't need a robust response Cliff, but....I pretty much hate my Sky HD box.

I don't work for Sky any more - when they bought out Amstrad that was pretty much the ond of it for me - I actually lead a team of about 30-40 engineers who were trying to fix all the issues you listed but Sky would never authorise release of the updates on the basis it could actually get worse. That and the fact that it took ten meetings, 3 committees and a user group to ever get anything authorised.

Having said that I have two SkyHD+ at home - one Amstrad, one Thomson. I've never lost a recording with the Amstrad and only a couple of times with the Thomson and my channel change time is sub 1 second. (with a 15 frame GOP it can take the best part of a second to get an I-frame).

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

MartinM57 wrote:
clawson wrote:
It took about 50 people two years so call it one hundred man years. This was using engineers with an average 10-15 years experience. It involves 57000 .c source files, more than 100 concurrent tasks with about 50 queues and over 500 semaphores.
It doesn't need a robust response Cliff, but....I pretty much hate my Sky HD box. It's mighty clever, and the HD picture quality is amazing, but when it takes 4 seconds to change a channel or respond to remote key presses (yes, pulled the plug/waited the ridiculous 30 seconds/plugged it back in again - accompanied by the zztt as it makes contact [why hasn't it got a proper power switch?], rebuilt the software blah blah blah), doesn't record things you settings it to record and breaks completely every 2 years or so and needs to be replaced, then I do wonder why I've paid Mr Murdoch about £8000 in the last 10 years (yes, not all HD years of course)

It's not until recently that we have gotten the faster ARMs.

My older receivers all had those annoying lag issues that didn't get fixed until years later with faster processors.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The lag is nothing to do with CPU speed. Even a 10 year old box had plenty of CPU speed.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

andrianu wrote:

I mean, even teachers they never sow what is inside that big box, and at circuit level, they can't explain how is made.

Why don't you build a basic 4 line telephone exchange as your project - one which you can plug standard analog phone handsets into and make calls with them. It'll be a good opportunity to learn what goes into making the "black box".

- S

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:
Why don't you build a basic 4 line telephone exchange as your project - one which you can plug standard analog phone handsets into and make calls with them. It'll be a good opportunity to learn what goes into making the "black box".

This seems like a good idea for a beginner..

-------------------------------------
When the champions rise from the abyss, even the sun will be out-shined.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Except I'd make the handsets 16Kbit ADPCM and make em work over a radio.

Imagecraft compiler user

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

About the small telephone central, is of course, a very nice project, but i am very afraid that the teachers will take this as plagiarism (piracy), because there are a lot of open source projects for this, and many student's (from past years) already made such a device ( I remember a 16 line one, and the telephones could communicate with each other, and could call an external number), but as I said, they were accused, that they might copied the project from somethere ( and I know for sure that they copied it :D )

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:
About the small telephone central, is of course, a very nice project, but i am very afraid that the teachers will take this as plagiarism (piracy), because there are a lot of open source projects for this, and many student's (from past years) already made such a device ( I remember a 16 line one, and the telephones could communicate with each other, and could call an external number), but as I said, they were accused, that they might copied the project from somethere ( and I know for sure that they copied it Very Happy )

You must be Indian :D

-------------------------------------
When the champions rise from the abyss, even the sun will be out-shined.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

You can try this project : Portable USB to USB transfer..

The model has two USB ports, 1. Parent USB 2. Child USB
,a LCD display, few keypad buttons and a microcontroller..
You can insert pen drives/flash drives at these ports and you can copy data from parent pen drive to child pen drive... Concepts of serial communication is covered here... :)

-------------------------------------
When the champions rise from the abyss, even the sun will be out-shined.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Whats the points formula? 25% for requirements and spec and documentation, 25% for hardware coolness, 25% for software coolness, and 25% for overall coolness? In other words, its cooler to have a purpose built board and custom sw rather than a COTS Arduino with their libraries? (I like my digital walkie talkie idea still)

Imagecraft compiler user

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

How about a device that spits out random project ideas?

Four legs good, two legs bad, three legs stable.