Thesis follow-up december

For my thesis follow-up of december, I had to make a detailed planning of my work, for the next 6 to 7 months. This planning can be found here for reference.

Intermediate thesispresentation

I did an intermediate thesispresentation this week, the pdf files of which can be found here.

During this presentation, I showed screen prints of my first application. What this application does is explained in the presentation. The .jar files to run the application can be found here:

• the application itself (needs to run to have a GUI);
• the “person” plugin;
• the location plugin;
• the reasoner;
• version 5.0 of the MUSIC context middleware;

To run the application, install each of these .jar files in equinox, and start them (if you want to see the plugins appear in the GUI of the application, start the application first, but the other way around works too).

If there are any questions, feel free to leave a comment or send me a mail.

Thesis follow-ups october and november

To be able to go back to my thesis follow-ups (to see whether I’m still on schedule), I will post them here from now:

• Thesis follow-up for october 2009;
• Thesis follow-up for november 2009.

Popular article in the making

The new version of my popular article (now entitled “Het leven delen op een need-to-know basis”) can be found here.

At the time of this post, it is not completely finished yet. I will update the file while I continue writing, so the link will end up pointing to the finished article.

summary of popular article

Based on the subjects that I want to write about in my popular article, I made a summary of my popular article to be sent in with my november thesis follow-up:

“Ubiquitous systemen, waaronder context-aware systemen, worden alsmaar meer wijdverspreid. Dit soort systemen die op de hoogte zijn van hun omgeving moeten efficiënt werken, zonder daarbij de batterij van de gebruikte apparaten “dood” te bloeden. Daarom wordt middleware zoals MUSIC gebruikt bij het ontwikkelen van context-aware applicaties.
Bij het gebruik van context-aware systemen ontstaan wel een aantal security en privacy problemen, aangezien heel veel informatie gedeeld kan worden met de omgeving. Dit kan voor Big Brother situaties zorgen wanneer de omgeving altijd op de hoogte is van waar de gebruiker zich bevindt. Of voor situaties waarin mensen dingen over u te weten komen, die eigenlijk privé zijn en dus niet met hen gedeeld mochten worden. Het is dus belangrijk dat informatie dus enkel op een “need to know” basis wordt gecommuniceerd. In mijn thesis zal ik dus systemen zoeken om dit te bekomen. Onder anderen zal gebruikt gemaakt worden van Quality of Context, machine learning om sociale relaties en regels te beschrijven, en certificaten.”

Popular article, take 2

The first draft of my popular article ended up not being “scientific” enough. So I decided to start anew, making sure I knew which aspects of my thesis should be mentioned in the article. In the new version, I will try to describe the following points:

• Short introduction about Ubiquitous systems, and specifically Context Aware systems;
• Explanation of MUSIC;
• Example of why one would only want to share their life on a “need to know” basis;
• Based on the above example, a description of the security/privacy problems that can arise in context aware systems;
• Possible solutions to solve these security/privcacy problems, that I will try to use in my implementation;
• Conclusion, recapitulation of the paper.

[I have to give a short summary of this article in my november thesis follow-up, so I will soon post it here]

Trust in ubiquitous social networks

While reading papers and thinking about how the ideas described in them, could be applied to my thesis, I decided it would probably be best to split the communication in two groups. By splitting it up, a different approach can be taken for each group (see part 1).

A first draft of possible ways to implement the security/privacy in the context scenario I will be developping, is also given in part 2.

summaries of literature, part 3

Because I needed to find a way to respect the user’s privacy in my scenario, I continued my literature study.

First I started looking explicitely for more information about “Gaia”, a middleware that allowed authentication for portable devices, based on Kerberos. I ended up having to let this idea go, since this project apparently doesn’t seem to be supported anymore (the latest information I found about it on it’s website and in papers, dates from 2002).

So I started searching for other possible solutions. While doing this, I read several papers on authentication in ubiquitous systems (summaries of the interesting papers can be found here).

After reading these papers, and realising that authentication mechanisms designed specifically for portable devices are not widespread, my conclusion for now is that I will probably use a combination of existing methods to determine which information the user’s device will communicate to another device:

• indicators of quality of service:
  • precision (the granularity of the information that is given out);
  • freshness (is the information still valid);
  • spatial resolution (can the device I interact with really be it says it is? For example, a device in a coffee shop that tells me it is a hospital computer, will probably be wrong/malicious).
• certificates: windows mobile has SSL certificates support.

Summaries of papers for literature study

A summary of papers I read about Security and Authentication in distributed systems (particularly context-aware and ubiquitous systems) can be found here.

First part of the “popular” article

I have started writing the popular article I will present at the end of november. The current version of this article can be found here.

Note: This is not the final draft of the article. The file will be updated in the future, and I will try to always make it link to the most recent version of the article.

Update: the paper has been translated to Dutch, so the link will point to a Dutch text.