Dissertation Ideas - Networking

TomElZ

TomElZ

TomElZ

Associate
Joined
7 Aug 2008
Posts
302
Hey lads.

Unsure whether this is in the correct place but I guess it is relevant.

Basically I'm in the last year of Uni and the time has come to do a Dissertation/Final Year Project. Now I have selected Networking as my area of study however the amount of stuff out there to do is overwhelming.

Just wanted to know what topics people think would be interesting and I kind of want to research into something that is 'Fresh' if you get me. IPv6 seems like a possibility and stuff to do with CISCO but really don't know where to start.

Cheers
Tom.
 
The problem with IPv6 (at least when I was at uni) is that the Simulation/Modelling tools don't really support it well if at all, which can make it a huge pain in the ass to try and demonstrate anything like large scale performance advantages/disadvantages. You could do something like an IPv6 migration guide which would give you plenty to talk about without needing much in the way of experimentation.

Play it safe, pick something you know you can do and aim to finish it by Christmas - Don't leave it until the end of the last semester when you have tons of coursework and exams to revise for.
 
Last edited:
Is there any other aspects that you would recommend or think would be of interest - just gathering thoughts at the moment as I haven't even met my supervisor yet.

Yeah I'm going to do my best to make sure I don't leave it all to last minute ahha!

What about Wireless networking technology?
 
Last edited:
I recently just finished my OU final project and as I have an interest in networking I had similar initial ideas. When I started my project it was recently annouced that the last /8 of IPv4 had been allocated, so although initially I thought looking into IPv4 and the need to transition into IPv6 could be interesting there is just a lack of technical depth, IPv6 just isn't that complicated.

Eventully I ended up doing a project on Wi-Fi, looking at how 802.11g has evolved into 802.11n, researching the individual improvements and using 2 linksys E3000 routers with DD-WRT to create a wireless ethernet bridge and try and isolate and test each of the advancements in 802.11n. It turned out to be a good project, you can go into a lot of depth in WiFi, although WiFi is a pain to work with and the majority of the time did not play nice and work how I wanted it to.

Just to throw a few other potential ideas at you based on what some of the others on my course have done that are ICT related, you could look at mobile communications such as LTE/WIMAX, mobile sensor networks and steaming media (which would touch on topics such as bandwidth requirements, encoding, DRM etc..)
 
Last edited:
I know this isn't very helpful but my brother did his dissertation on the How VOIP can benfit a business, using my dads electrical company as a case study
 
Dist - thanks for the input there. Sound's like quite an interesting area and maybe something I could look into more. Just out of interest you mentioned you used 2 linksys routers - did you have to purchase these yourself or did the uni contribute to the costs?

Mobile related stuff could be of interest I guess as well considering it's increased useage - would that not fall under Communications areas though?

Jay - Another useful point but haven't got any company's that I could use.

Cheers for the help.

Any other points would be appreciated as I'm just gathering thoughts at what I can do at the moment.
 
For my final year project in Computer Science I played with wireless ad-hoc networks. I never actually did anything physical unfortunately, it was more research. Aimed at defining trust relationships within a cluster of ad-hoc nodes and looking at the effect on the range/reliability of the "cloud"

What constitutes the trust? Length of time in the cluster, proximity, previous history, that kind of thing. Wrote a couple of simulations, fairly interesting but would have liked to move it over to the physical side for demonstrations.

I'll also echo the leaving it past Christmas statement (like I did) - mistake! Exams + coursework + final year project = lots of red bull fuelled all nighters.

Good luck :)
 
I was eligible for a grant from the OU, which wasn't much but it just about covered the costs of the equipment I needed and all of the printing/binding of my project report.

Mobile networks and streaming media is more 'communications' than 'networking', but you have to remember for a dissertation you need something with technical depth and when I was looking into a decent networking project it was supprisingly difficult to find something that would be the level required, a lot of networking type subjects are quite basic with the more technical stuff usualy being at the phsyical layer.

If networking is all you want to focus on then perhaps another idea would be analyse all of the phsyical layer technologies (focusing on a specific environment, such as in the home, office or enterprise) such as Wired Ethernet, WiFi, Powerline, Fibre Optic. Depending on what you have available to you could potentially start off with research and looking into the phsyical specifications of each technology and then create a practical experiment to test them.
 
Dist said:
Mobile networks and streaming media is more 'communications' than 'networking', but you have to remember for a dissertation you need something with technical depth and when I was looking into a decent networking project it was supprisingly difficult to find something that would be the level required, a lot of networking type subjects are quite basic with the more technical stuff usualy being at the phsyical layer.

If networking is all you want to focus on then perhaps another idea would be analyse all of the phsyical layer technologies (focusing on a specific environment, such as in the home, office or enterprise) such as Wired Ethernet, WiFi, Powerline, Fibre Optic. Depending on what you have available to you could potentially start off with research and looking into the phsyical specifications of each technology and then create a practical experiment to test them.
Click to expand...

Good advice. I did my dissertation on networking and found it really hard to find a semi-decent idea. I'm not going to tell you what I did as it would be too much of a give away but i'll give you a framework.

Find a layer on the OSI you like and pick a technology within. Pick 2 or 3 which are currently used in industry. Investigate how they work & show differences between them. Find a security flaw (current or old) that has been exploited and advise how to stop it. Backtrack linux has loads of pre-made software to exploit networks.

That was the basis of my dissertation. It wasn't ground breaking compared to a few of the other guys in my class but it got me a 2.1 which I was pretty happy with. Could have easily broken the 10k word limit aswell.
 
Hey again lads.

Well after a meeting with the supervisor today and me proposing a few general ideas. I've been told that I need to come up with something that I will be able to produce, test, simulate - along them lines.

So perform some research into some kind of area, use this research I guess to perform some tests, simulate some kind of action and then draw conclusions from this I guess.

Now the Networking area compared to other areas (Electronics, Programming etc) is probably a bit harder to come up with ideas for (IMO).

Referencing back to a post above about wireless standards and how they have evolved may be do some kind of comparison between a 802.11a router and a 802.11n router?

Bit stuck again for what practical stuff I could do to contribute.

Cheers lads - any ideas at this moment are greatly appreciated.
 
Well for wireless there is plenty of stuff you can perform practical testing on. You can look at whats important when using Wi-Fi, such as bandwidth, reliability, latency and jitter for exmaple, and then see how they vary using different technologies. Obviously you will haev to go into depth at looking at not just 802.11n is better than 802.11g, but the specific technologies involved in 802.11n that advance it over previous standards, and look at creating a test for each of those technologies if possible.

I don't want to give everything away as this is realy something you should be looking into yourself, and you may have to look into a lot of stuff before you come up with a good idea. But the best advice I can give is to think of a technology, break it down into its main parts, reasearch those parts, predict how those technologies should perform and then create a test for that, then compare your predictions with the test outcome. Also remember that context is important too, such as what sort of environment are you doing the testing in (home, business, enterprise), what is the need for such research/testing, how will these results impact real applications of the technology/the end user, things like that.

You have a basic idea of what is expected of you now, you have a general idea of what you want to do, so start looking through some academic papers (just a quick browse) to see what sort of stuff can be tested to get an idea of what is/isn't possible.
 
Hey.

Well I have submitted a few of my own ideas as well as stuff on WiFi but again I have come to a brick wall and have been told that I need to produce something that actually produces something at the end of it as opposed to just a report by my supervisor.

Original ideas included things such as social networking and the privacy aspects, some sort of research into cloud computing, P2P and security and just looking into a particular OSI layer.

Browsing around I am finding a lot of research based stuff but nothing with practicality involved which could help me or give me an idea of what I need to look at.

Any sort of electronics - i.e building circuits etc for a networking device is completely out the question as I don't enjoy electronics in the slightest.

I think that the ideas are there just coming up with something that could be implemented practically is the brick wall.

Cheers.
 
I am really quite shocked that some of those could be considered dissertation projects, is that what a degree level networking course is these days? Do something on mesh networks or something cutting edge.
 
I'm not doing a networking course - I'm doing a computing technology course but I have picked networking as my area for my final year.

Bear in mind I only have till the beginning of March so I've got to do something achievable and I doubt I can do something that is 'cutting edge' in the space of 5 months.
 
OK, but some of this is A level grade stuff. If you're at dissertation level and can't learn about, implement and document a mesh network setup in 5 months then I'm shocked at what constitutes a university education.

A couple of gears ago I bought half a dozen Linksys WAPS and hacked them to form a mesh network, it was imperfect and flawed but that's just stuff to write about for you. It took me a couple of sunday afternoons to get basic functionality working.

Or do ID/Locator split, that's topical and interesting and an appropriate level for a dissertation project. Less physical though.

Though I may be of the old fashioned opinion that a university course should possibly teach you a bit more than a CCNA would and your dissertation should reflect that.
 
I have heard of a few people mention Mesh networking.

I will do a bit of reading into it over a few days and see whether it is viable - anything that will get me a high mark and something I can demonstrate in a practical way and I'm a happy guy.

Cheers.
 
How about bonding? Channel bonding at the physical layer or higher level using VRRP? Plenty to break and go wrong and plenty of things that work well
 
$anch3z said:
How about bonding? Channel bonding at the physical layer or higher level using VRRP? Plenty to break and go wrong and plenty of things that work well
Click to expand...

That's, incredibly basic, CCNA material and everything you could possibly want to know about it has been documented twice already. Do people seriously do dissertations on such things?
 
You must log in or register to reply here.
Back
Top Bottom