Results 1 to 10 of 10

Thread: Any ideas for a computer sci proj(final year)

  1. #1
    Just burned his ISO
    Join Date
    Mar 2007
    Posts
    15

    Default Any ideas for a computer sci proj(final year)

    Hello everyone, I am currently in my final year of engineering, Comp Sci, and as part of the curriculum we have to complete a project on any topic(CS related obviously) for course completion.

    I was hoping some of you could pitch some ideas for the topic of the project cause I am drawing a complete blank.......... anything interesting, even if its a bit tricky its ok.
    Considering the fact that I have an interest in computer networks, ideas for topics in this field will be highly appreciated.

    ....any help will be appreciated, tks in advance.

    P.S : I'm good wid c,c++,java,html,php, have been using BT for almost 2 years now and i'm extremely comfortable with linux and know how to use quite a few of the pentesting tools provided in BT.

  2. #2
    Member
    Join Date
    Jan 2008
    Posts
    194

    Default

    Something that really is in high demand with little open source backing:

    Wireless Intrusion Detection Systems

    I've been researching this heavily for the past month or so and have decided to just start from scratch on a system. There are a lot of good docs out there if you whip out your google-foo.

  3. #3
    Senior Member Talkie Toaster's Avatar
    Join Date
    Jun 2008
    Location
    Scotland
    Posts
    131

    Arrow take it a bit further.....

    A wireless ids would be fantastic, but how about using multiple pickups (either old machines setup as kismet drones,or using linksys WRT54G routers) each assigned their gps location and then make a guess at where the 'target' device is....

    Imagine a large call centre with 200 employees per floor, over 4 floors. Have a WIDS box on each floor, with 3 or 4 routers/drones spread over the floor, and each WIDS box connects securely to a master box for the IT admin. When a suspicious client/AP appears he would be able to see what floor and more importantly what part of the floor the device is likely to be in, he can then head up with a laptop already knowing the channel,mac,name and a rough location and catch the offender.

    Most of the stuff needed to do this is available, it really just needs put together properly.

    Maybe if you design something for your uni they might help with equipment, if they have a no private wifi AP policy which many seem to nowadays.....

    TT
    Programming today is a race between software engineers striving to build bigger and better idiot-proof programs, and the Universe trying to produce bigger and better idiots. So far, the Universe is winning.

  4. #4
    Member
    Join Date
    Jan 2008
    Posts
    194

    Default

    Talkie,

    What you are describing is exactly my challenge except it's 1000+ employees in a 6 story building.

    Kismet actually does pretty well as a WIDS but doesn't coorelate the capture source with the interception. Therefore, having a distributed WIDS is somewhat of a pain. Kismet newcore assigns a UUID to the capture source but that isn't tied to the interception packets in the program. Dragorn is working on that as we speak.

    I plan on posting a message here sometime in the future once I get a working distributed WIDS.

  5. #5
    Senior Member Talkie Toaster's Avatar
    Join Date
    Jun 2008
    Location
    Scotland
    Posts
    131

    Lightbulb maybe i'm being stupid...

    maybe i'm being stupid but i'm sure while checking out gps with kismet i noticed a option to give the drones each a individual gps location, i don't know what it then does with that info though or if it would get sent to the 'master' box along with the rest of the drones traffic.

    Seems a bit silly to have it if its no real practical use.....

    Your project sounds really interesting though, hope to hear more about it soon as I'm getting really into this wireless thing!

    *he sits polishing his bi-quad!*

    TT
    Programming today is a race between software engineers striving to build bigger and better idiot-proof programs, and the Universe trying to produce bigger and better idiots. So far, the Universe is winning.

  6. #6
    Member
    Join Date
    Jan 2008
    Posts
    194

    Default

    This should give you a better idea as to where kismet stands in relation to being a distributed WIDS:

    http://kismetwireless.net/Forum/Gene...131421.4146891

  7. #7
    Just burned his ISO
    Join Date
    Mar 2007
    Posts
    15

    Talking

    Hmmm, seems very interesting. Will have to research a bit and check out its feasibility and then discuss with my prof. Still, thanks a lot guys, I feel this is a huge challenge cause I'm reading what you guys have written and some of it is going over my head.
    Any links to reading material will be helpful, I've already googled some stuff, but I guess you guys will know better sources..... so if any links, please post. Tks again.

  8. #8
    Member
    Join Date
    Jan 2008
    Posts
    194

    Default

    Quote Originally Posted by srinag View Post
    Hmmm, seems very interesting. Will have to research a bit and check out its feasibility and then discuss with my prof. Still, thanks a lot guys, I feel this is a huge challenge cause I'm reading what you guys have written and some of it is going over my head.
    Any links to reading material will be helpful, I've already googled some stuff, but I guess you guys will know better sources..... so if any links, please post. Tks again.
    It IS interesting and quite in depth. Defense is quite a bit more difficult IMO.

    But, this is YOUR research project If you decide to do it, I'll throw a few things your way. Just keep in mind, all I used was google.

  9. #9
    Senior Member Talkie Toaster's Avatar
    Join Date
    Jun 2008
    Location
    Scotland
    Posts
    131

    Default Totally of the wall idea....

    .... but if kismet doesn't work how it should relating to drones maybe you could fix it???

    Sounds extreme i know, but i don't mean a full re-write or anything like that! more getting the drone/server arrangement to work on YOUR equipment in YOUR situation. So that'll be some coding, some debugging and selecting easy to mod equipment for the purpose.

    If you keep all the equipment the same then it shouldn't take too long , this kind of mod would normally take ages with added testing and stuff if it was for general release, but if its for one project/contract and it isn't expected to work on everything that someone might plug into it it might be achievable.

    I'm afraid this is where my knowledge starts to run out, however there are many others on this forum who might be able to tell you if this is possible.

    If ^that^ is a bit to much you could always check out barry's box(!), click on his user name then through to his home page, not sure what your Comp Sci actually involves where you are, is it mostly software or hardware based? A wireless audit box with a passive ethernet tap built in, maybe with a secure 'dial-to-admin' so it can be left somewhere on a site would be cool.

    TT
    Programming today is a race between software engineers striving to build bigger and better idiot-proof programs, and the Universe trying to produce bigger and better idiots. So far, the Universe is winning.

  10. #10
    Just burned his ISO
    Join Date
    Mar 2007
    Posts
    15

    Default

    Quote Originally Posted by Talkie Toaster View Post
    .... but if kismet doesn't work how it should relating to drones maybe you could fix it???

    Sounds extreme i know, but i don't mean a full re-write or anything like that! more getting the drone/server arrangement to work on YOUR equipment in YOUR situation. So that'll be some coding, some debugging and selecting easy to mod equipment for the purpose.

    If you keep all the equipment the same then it shouldn't take too long , this kind of mod would normally take ages with added testing and stuff if it was for general release, but if its for one project/contract and it isn't expected to work on everything that someone might plug into it it might be achievable.

    I'm afraid this is where my knowledge starts to run out, however there are many others on this forum who might be able to tell you if this is possible.

    If ^that^ is a bit to much you could always check out barry's box(!), click on his user name then through to his home page, not sure what your Comp Sci actually involves where you are, is it mostly software or hardware based? A wireless audit box with a passive ethernet tap built in, maybe with a secure 'dial-to-admin' so it can be left somewhere on a site would be cool.

    TT

    We deal with both aspects, software as well as hardware. I have even studied advanced micro-processors....
    I looked into the whole WIDS thing.... I understand the logic and the work involved, but I cant use existing applications for the project, we have to write our own code.... And it dosen't have to do the whole detection and prevention thing.
    I'm just looking at writing the code for a sniffer which will pick up packets and if the packet is intended for the network its monitoring, then it will check its mac i.d against the list of known and authenticated i.ds and if it is not in the list then it will report it to the admin. This much functionality is sufficient.
    Will give it a gui to display the info regarding the packets its picking up and alert the admin if a packet from an unknown access point(mac i.d not in list) is attempting to join the network......
    Again, guys if you know any good links, pls do give it to me...... I defnly have to read way more before I can get some where.

    P.S : I just approcahed my friend in CISCO, said he might be able 2 get me an intern project from there.... keeping this option open as well.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •