N grad school, it seems, you are supposed to do this thing called "research". So I've done a little bit of that. Below are some of the class projects that I've done. You can read the final reports and/or presentations.

ast year I decided to graduate from Berkeley with a Masters degree, so I did. For my thesis, I invented a technique for reducing the delay due to crosstalk on a digital bus by encoding the data with a "self-shielding" code. I developed a rigorous theoretical framework for describing and analyzing self-shielding codes and calculating their performance. This involved making up a lot of terminology and proving a lot of theorems. All that is available below.

resume is available.

Masters Research:
Bus Encoding to Prevent Crosstalk Delay

ICCAD paper: A short paper, written for and accepted by the ICCAD-2001 conference. Gives a summary of the research and results, with cursory proofs of the theorems. (7 pages, 91Kb pdf)

Thesis, Official Version: Version of my thesis that was submitted to the school. Dry and humorless. (132 pages, 1053Kb pdf)

Thesis, Director's Cut: Version of my thesis with the good stuff left in. Still doesn't read quite like a Tom Clancy novel, but it's a bit more entertaining. (132 pages, 1058Kb pdf)

Class Projects

Software Optimization Using Hardware Synthesis Techniques: EE 219b project. An exploration of how various hardware design techniques can be applied toward optimizing software control structures, and implementation of a package that attempts to optimize control structures through boolean network manipulation. (pdf) (ppt)

Variable Word Width Computation for Low Power: CS 252 project. Analysis of a technique for reducing power consumption in a general-purpose embedded processor by allowing computation to be performed on shorter word widths. (pdf) (ppt)

Design of a High-Bandwidth Sigma-Delta Modulator: EE 247 project. Description of the design of a Sigma-Delta analog/digital converter for wireless communication applications. (pdf) (ppt)