CSCI 7000 -- Project Page
Each student should select a paper
from the list below. If you prefer a paper not on this list, you
must submit it for approval by me. The only real requirements for the
paper you choose are that it is relevant to the course and it contains
non-trivial material. After you have selected a paper, you
must do the following:
- Tell me via email (not in person) which paper you have selected.
This is an individual report, and you cannot team up with other
students. You must send this email by Friday, 4/15. If
you are selecting a paper not on the proposed list, get approval
well in advance of this date in case it takes a while to settle
on something.
- By Tuesday, 5/3, you must submit a document which summarizes the
contents of the paper. You do not have to do an oral presentation
of the paper.
The Report Itself
This isn't meant to be a hard project. It shouldn't take more than a
few days to complete. After you have selected a paper, read it, study it,
look up references if you have to, or come to me if you cannot understand
some central part of it. Then write your report. Here are the requirements:
- It must be typeset in LaTeX.
- It should be about 5 to 7 pages in length, but completeness and clarity
will determine my evaluation more than length. Don't babble on
endlessly just trying to fill out the page requirement; be terse
but complete. And if you come up short, add some more background
material (even if it's redundant with the class material).
- You should have the following sections in your report:
- What paper did you read: Title, Authors, One line abstract.
- What is the paper about: setting, problem, goal.
- What are the main results?
- What is your impression of the impact of this paper (you don't have
to go to the library and do a lot of research to answer this; give
me your informed opinion, try a web search, etc.)?
- Suggest related questions you think might be interesting which are
related to the topic of this paper. (You are not responsible for
ensuring that these questions have not already been answered by
other papers.)
- Extra Credit: try to extend the results, even if only in an incremental
fashion.
The score on your report will be based on the difficulty of the paper you
select (obviously choosing an easy paper means you wouldn't get as many
"difficulty" points, but you can make up for this in other ways), clarity
of the write-up, and the depth of your personal insights. I know, it's
all very vague, isn't it?
Another approach, rather than reporting on a paper listed below or one of
your choosing, would be to select a published cryptographic protocol and
break it. After all, that's what this class is about!
Papers
- Easier Papers:
- Moderate Papers:
- Generalized Birthday Paradox by
David Wagner. Mostly talks about probability theory and attacks based
on n-way collisions in the context of various cryptographic protocols.
- L-Collisions paper by Semanko.
Mike was a master's student at UCSD when he wrote this very nice paper.
It describes an attack on MACs which use a randomized string to get by
the birthday bound.
- RSA backdoors by Crepeau. He
investigates generating keys for RSA which provide backdoor access to
the system. Read this and suggest safeguards against it.
- Difficult Papers:
- SASAS attack by Birykov and
Shamir. This paper is basically an extension of the Square attack (tho
they don't really admit this) and assumes a good deal of knowledge about
linear algebra (you should know what "rank" means before delving in, for
example).
- TWIRL. This paper is about factoring
integers and should only be attempted if you have a decent background in
number theory. But what a cool paper.