GUIDELINES FOR PRESENTATION PROJECT

Each of you will present a mini-lecture (25 mins, 5 mins break between
mini-lectures) explaining a key concept on the blackboard in
sufficient detail that listeners could potentially solve a problem
based on the material.  You will also hand in a copy of your lecture
notes (handwritten or typewritten) to me.  These do not need to be
long and should not be an additional burden--you will be making notes
anyway for your presentation.  You just need to write them out neatly
enough that you can share it with me.

My goal this quarter is for us to understand some of the ideas behind
recent work on topics that have potential for future research.  I am
skewing it towards areas I am curious about, but have not had time to
explore in depth.  If there is something else that you are intensely
curious about, let me know.

TEAMING

Teams of two or more, seeking to provide a coherent picture of a paper
or set of papers.  Team members will be giving consecutive
mini-lectures, or pooling their time into a time-interleaved regular
lecture (e.g. a 50 minute lecture from a team of two, with each person
alternating in some fashion, is also acceptable)

SCHEDULE

STARTING NOW: Form teams, browse through papers, decide topic.  Do preliminary consultation
with me about your plans.

THURSDAY APRIL 23: Each team hands in a project proposal (at most one
page) describing their topic, and listing the papers they will cover.
It would be good to convey some preliminary idea of what key concepts
you will be teaching us, but I understand that this may change as you
study the papers further.

THROUGH THE QUARTER: Consult with me and each other as needed as you
study the papers.

PRESENTATIONS: We will schedule them in two sessions during the last
week of classes.  If I manage to cover the material I want to cover
before that, we can do it during regular lecture slots.  If not, we
will schedule separate slots in the evening. (You owe me 3 hours in
any case, since there is no final exam.)

SUGGESTED TOPICS

Cognitive radio: I am particularly interested in the work coming out
of Berkeley Wireless Research Center, which includes both work on
fundamental limits (involving Anant Sahai) and testbeds (involving Bob
Brodersen).  More generally, the Dyspan workshop is a good source for
papers.

Network coding: While the concept came from information theorists
(Raymong Yeung), I am particularly interested in work coming from the
computer science community on its practical applications to wireless
networks.  Specifically, a series of papers by Dina Katabi from MIT
look interesting, and appear to contain simple ideas that should be
easy to explain.

Flash memory: Flash memory is becoming increasingly popular, but has
not received as much attention from communication theorists as, say,
the magnetic recording channel.  Explaining the channel models and
typical impairments is a necessary prelude to coming up with clever
ideas for improving it.  I saw a good session on this at the 2009
Information Theory and Applications (ITA) workshop at UCSD--many of
these papers involved Jehoshuah ("Shuki") Bruck of Caltech.  These
papers are not on IEEE Explore yet, but I will try to get preprints.
But there are papers already out there on this topic.

Wireless channel modeling: There's been a lot of work on this over the
years, but the work by Massimo Francechetti at UCSD (A random walk
model of wave propagation) has an interesting perspective that may be
worth discussing.