Nyhetsflöde

  Project submission deadlines
* Topic and contact person: Sept 247, MonThursday
* List of papers, with one-one sentence motivation: Sept 27, ThursOct 1, Monday
* Outline, max one page, paper structure, content of sections: Oct 34, WedneThursday
* Final report, max 5 pages: Oct 9, Tuesday 
* Presentation day: Oct 11, Thursday
How does the project work?
* You have to work in groups – three or four students per group. The groups are defined by the instructor.
* Select subject from the list or any subject you like. I would like to know which subject you have selected. If you suggest your own subject, you have to get my agreement.
* You will have to overview 3-4 papers. The papers should discuss/use mathematical models for performance evaluations. Send me the paper references (authors, title, where it was published) once you have them.
* You have to prepare a written report of ca. 5 pages, comparing the selected papers (not just discussing them one after the other!). Send me the outline of the report before you write a detailed version.
* Each project will be presented on the scheduled exam in a talk of about 20 minutes.
Which sources to look at? The following journals and conferences are good sources of high quality papers. Check some papers form the last years, read the related work chapters and look for often referenced early publications. Often they are the ones with basic results in an area.

You can find the full papers through www.lib.kth.se, if you are in the campus.

* IEEE Communication Magazine – publishes mainly overview papers
* IEEE Infocom, IEEE ICC, IEEE ICC – the major networking conference
* IEEE Transaction on Networking – heavy research results, often hard to read…..
* ACM Sigcomm – very selective networking conference, basic Internet research, the mathematical models are usually possible to follow.
Submission intructions: Topic: please submit the topic you have selected as follows:

* subject: EP2210 topic {group #}
* in the body: motivation why the sopic was selected. Also, please give the contact person from the group. I will answer to him/her.
List of papers: please submit the list of selected papers as follows:

* subject: EP2210 list of papers {group #}
* in the body: please write simple text, no attachments. For each selected paper: give the exact reference and a max 10 lines motivation on why the paper was selected.
* how to give full reference: list of authors, title of the paper, place and year of publication. For example: L. Massoulie, J. Roberts, “Bandwidth sharing: objectives and algorithms,” IEEE Infocom 2000.
* Note, do not send me the papers themselves, I can find them on the web.
Outline: please submit the outline of your report as follows:

* subject: EP2210 outline {group #}
* in the body: please write simple text, no attachments. Follow the structure of scientific papers. Compulsory parts are Abstract, Introduction and Conclusion. Between Introduction and Conclusion structure the paper according to your needs. For each section write max 10 lines or a couple of bullets on what the section will contain.
Final report: please submit your report as follows:

* subject: EP2210 final report {group #}
* attached: the paper, in pdf format, if possible. Follow the structure of scientific papers. Compulsory parts are Abstract, Introduction, Conclusion, References. Between Introduction and Conclusion structure the paper according to your needs. The paper can not be longer than 5 pages.
Presentation: Thursday Okt 11, E35

20 minutes presentation per group. All group members should be there. Make sure that your presentation is not longer than 20 minutes.

Two time-slots of ca. 2 hours. You have to be present in the entire time-slot where your group gives the presentation.

Select one of the listed topics, or define your own. If you define your own topic, talk to me before you start to work.

One paper is listed for each topic. Select 2-3 other, closely related papers to address the topic of the project. You can find related papers by checking the references in the paper, or you can google the paper, and then you will see other publications where this one is referenced.

Topics are added continuously:¶ Model
* How can we model the wireless channel for packet level performance evaluation? From communication theory to teletraffic theroy.
* Is long-range dependence present ing the packet loss process on wireless channels.¶ Modeling the effect of hidden and exposed terminals.¶ 802.11 RTS/CTS model (A. Bianchi, Performance analysis of the IEEE 802.11 distributed coordination function, IEEE JSAC, 2000)¶Internet? T. Karagiannis, M. Faloutsos, R.H. Riedi, "Long range dependence: Now you see it, now you don't!", Proc. of IEEE Globecom, 2002.
* Is it complicated to model the performance of a WLAN access point? A. Bianchi, "Performance analysis of the IEEE 802.11 distributed coordination function," IEEE JSAC, 2000
* The precence of hidden and exposed terminals. How bad is it?
* How does TCP perform over wireless links where packet losses happen due to signal propagation probelms? F. Anjum and L. Tassiulas, "On the behavior of different TCP algorithms over a wireless channel with correlated packet losses," SIGMETRICS Perform. Eval. Rev., May 1999.
* What happens with the very short flows in a network with TCP flows? Liang Guo, I. Matta, "The war between mice and elephants," International Conference on Network Protocols, Nov. 2001
* What do they mean with proportional fairness in the wireless world? D. Piazza, L.B. Milstein, "Multiuser diversity-mobility tradeoff: modeling and performance analysis of a proportional fair scheduling," IEEE Global Telecommunications Conference, 2002. Nov. 2002.
* Is peer-to-peer working for live streaming, when users come and leave all the time? R. Kumar, Yong Liu, K. Ross, "Stochastic fluid theory for p2p streaming systems," IEEE Infocom, 2007.
* What is Dynamic Spectrum Sharing and why is it good for everyone?
* What service quality can we expect from a cloud computing server?