dan wells

Intel has recently released their Intel Atom processor and motherboards. I specifically like the Dual Core one. While these motherboards are relatively easy to get your hands on by ordering online (and cheap!) in SA, I’m having serious trouble finding any decent Mini-ITX cases for it.

With my project I’m hoping to deploy to small form factor PC’s, most of them to be simple routers which do a little packet crunching. Ideally I’m looking for a case which I can put in an additional network card, that’s it. Smallest form factor as possible really, and low cost.

Will keep you posted if I find something decent :P

My supervisors and myself have decided that I’ve been doing too much technical work lately and that I must start my Masters thesis write up. I could not agree more.

I have constructed my thesis layout using LaTeX, nicely laid out into seperate files for each chapter, the bibliography, appendices and the glossary.

Currently I’m working on my Chapter 2, which is the Related Work or the Literature Survey. It’s a good start and I’m slowly getting back into writing again. Things are going faster after each consecutive day that I work on it.

Watch this space, but Chapter 2 is presently sitting at 3338 words.

Just a quick update on how things are going! (Sorry it’s been a while!)

So, in my virtual network which I posted about before, we have our ACR which is the gateway to our network, under this we have multiple routers and under those there are multiple machines. After much fussing with Squid Caching Proxy Server and LightSquid (displays Squid logs nicely, served off Apache) and FreeRADIUS authentication and DaloRADIUS to manage it. These are all running rather nicely but have been put aside for the moment (as our first deployment will not be needing Squid as we have an already well maintained cache here at Rhodes).

Our first deployment of the project will be to the SCW network, the Settler City Wireless project here at Rhodes, which is fully functioning test network. As the connections are wireless, and therefore rather susceptible to eavesdropping, we’ve set up PPPoE tunnels to secure and encrypt the wireless data.

Ok, so we’ve mimicked what is currently in service in the SCW, now to extend it to be easily managed and also monitor traffic for more information on how things are working. We’re going to be using FreeRADIUS combined with DaloRADIUS for managing user accounts (and going to be looking into the Billing and Accouting sections of DaloRADIUS). Cacti has been installed but it needs to be configured for the specific data we require and then graph it. Also we must get ahold of the Alvarion MIBS so the WiMAX devices can be monitored too.

Also, I want to look into using VMware Studio (released this month) which “provides convenient management features such as a web-based console with quick-start templates to streamline the authoring process of virtual appliances, and a command-line interface to enable the automation of management tasks and empower VMware Studio 1.0 to act as an extension of appliance vendors’ existing build and source control systems.”

More to come…

I need a network to do my implementation and testing of my Masters project. I created a virtual network as a backbone to base my project on. It’s a very simple network, partly based on how the Dwesa test network is implemented. Dwesa is a rural community in the Eastern Cape where many developmental projects using ICTs are researched.

The virtual network that I constructed has three ‘levels’. At the lowest level there is a typical ‘workstation’ which can be any PC, with any operating system, at a site. That workstation sends packets destined for other networks through its router. The router sits at the edge of the site and passes packets in that have been sent to it or routes packets for other networks to the access concentrator. The access concentrator either sends those packets to the Internet or to another site via their routers. The routers and the access concentrator are using FreeBSD 6.3.

For more detailed information see the document on my Virtual Test Network. Any feedback about how I’ve set this up would be great.

“On behalf of the SATNAC 2008 Programme Committee, I am delighted to inform you that the following submission has been accepted to appear at the conference:
Mapping the Location of 2.4 GHz Transmitters to Achieve Optimal Usage of an IEEE 802.11 Network”

Cool! Another conference! Really nice reviews from them as well, this paper was based on similar work to the ISSA paper but it has been completely reworked. Now I must just do the reviewers changes and submit the camera-ready paper.

Still waiting to hear about my Research-in-Progress paper from SATNAC, will keep you posted!

I guess I can say that I am now published due to an interesting turn of events!

I posted before on how my paper was unfortunately not accepted, but I wasn’t too disappointed as I was still going to be attending the conference and I designed an A1 poster for the event. Anyway, two days before leaving Grahamstown I found out that my paper had been upgraded to presentation material and that I would be presenting in a Research-In-Progress slot instead! Cramming an entire years work into a 15 minute presentation time slot was rather difficult, but I think it went ok and I received some interesting comments. The audience seemed very interested.

I am published in the proceedings as well, however it is not my “camera-ready” paper, which I am fixing up now and will be mailing to them to put on their website instead of the pre-review paper.

I also have a very shiny poster which was never seen…

Happy Birthday Dad! :)

Last Week:
Additions to project proposal.
SATNAC Work-In-Progress paper completed and submitted.
Research website up and running (this.blog)
[COE--Zenoss email notification set up and the spam fest has commenced]

To Do:
Finalise project proposal for full hand in on Friday. Get changes and finish by Monday.
ISSA Poster.
Daily progress reports to supervisors.

Future:
Get up and running implementation of project by latest end of September 2008 (from now this provides +-90days of development). Deploy to SCW network for data collection.

A full paper titled Mapping the Location of 2.4 GHz Transmitters to Achieve Optimal Usage of an IEEE 802.11 Network has been submitted to the Southern Africa Telecommunication Networks and Applications Conference (SATNAC) for 2008. This paper is based on the work I completed in my honours year.

I am finishing writing a second paper for SATNAC, a Work In Progress (WIP) paper which is more concise and based on the work I have completed so far in my Masters research. This needs to be submitted by 7th June.

Unfortunately my paper on Location and Mapping of 2.4 GHz Transmitters for the ISSA Conference 2008 was not accepted. But I will still be attending the conference and instead presenting a poster on my honours work.

So I will be heading to Johannesburg on the 6th July 2008 and I’ll be getting back to Grahamstown on the 10th July 2008.

Last year I completed my BSc Honours in Computer Science. My project was about discovering the locations of 2.4 GHz transceivers (like Wi-Fi devices) using multiple spectrum analysers (and the method of trilateration). The project was aptly titled IEEE 802.11 Signal Source Mapping using Low Cost Spectrum Analysers.

The project was rather successful and I developed a proof of concept application to demonstrate its uses. A small pamphlet (70 page thesis) was written discussing the finer points of the project :) For more details on the project see my honours research website.

The application produced some interesting results, and some really interesting signal maps. Each circle is displaying data from the application interfaced to the spectrum analyser, the circle radius is an estimate at how far a received signal has been transmitted from. Devices can be discovered by noting where the overlapping signals intersect.