M.S./Ph.D. Candidate
 Department of Computer Science
 UMass - Amherst
 140 Governor's Drive
 Amherst, MA 01003-9264

 b o u l a t @c s . u m a s s . e d u
 Telephone: (413) 545-3179
 [more details]

 

B. A.    Economics - Dartmouth College 2001
M. S.    Computer Science - University of Massachusetts - Amherst 2008

Research Interests:  Information theory, covert communications, cooperative wireless networks, Internet reliability, data management in sensor networks, implementation aspects of network algorithms, distributed computational geometry

Current work: Research Assistant in Advanced Computer Networking Group

Teaching: I was an instructor of CMPSCI691WS seminar on wireless network security in the Spring 2011 with Shane Clark and Dennis Goeckel and covered the information-theoretic aspects.

Advisors: Don Towsley and Dennis Goeckel (ECE)

Previous work: From March 2004 until August 2005 I was a Research Associate in the Web, Internet and Networking Group at Boston University Computer Science Department under the supervision of John W. Byers

Publications:

Square Root Law for Communication with Low Probability of Detection on AWGN Channels
Boulat A. Bash, Dennis Goeckel, Don Towsley
Full paper in submission [tech report]
Abbreviated version to be presented at ISIT 2012

Clustering in Cooperative Networks
Boulat A. Bash, Dennis Goeckel, Don Towsley
Proc. of IEEE INFOCOM 2011
Shanghai, China, April 2011
[pdf] [slides] [expanded techreport]

Informed Detour Selection Helps Reliability,
Boulat A. Bash.
Proc. of the 12th IEEE Global Internet Symposium (GI '09),
Rio de Janeiro, Brazil, April 2009
[pdf] [slides ppt pdf]

Exact Distributed Voronoi Cell Computation in Sensor Networks,
Boulat A. Bash and Peter J. Desnoyers.
Proc. of the International Conference on Information Processing in Sensor Networks (IPSN '07),
Cambridge, MA, April 2007
[pdf]

Approximately Uniform Random Sampling in Sensor Networks,
Boulat A. Bash, John W. Byers and Jeffrey Considine.
Proc. of the 1st Workshop on Data Management in Sensor Networks (DMSN '04),
Toronto, Canada, August 2004.
[ps] [pdf] [slides ppt pdf]

Full Curriculum Vitae

Research

I am primarily interested in the information theoretic aspects of wireless networking. Having derived the fundamental limit to covert communication over wireless links, I now seek to establish a framework for the covert network. Previously I did some work on cooperative routing in ad hoc wireless networks. In general I am interested in designing, analyzing, and implementing practical algorithms for ad-hoc, sensor, and peer-to-peer networks. I have also studied the path failures in the Internet as well as distributed computational geometry problems like Voronoi diagram computation and determination of nodes on the convex hull of the network in the realm of energy-constrained limited-range lossy wireless networks.

Previous life

My main research effort during and immediately after college was in financial economics. The empirical studies which I carried out required efficient design and implementation of data analysis algorithms and dealing with gargantuan data sets. While working as a research assistant for Prof. Kent Womack at the Tuck School of Business at Dartmouth College, I was interested in market microstructure-based long-term initial public offering (IPO) returns predictors. In other words, I looked at sequences of trades on stock exchange and tried to forecast returns of IPOs. This work resulted in an SSRN working paper.

When I joined Quantitative Strategies Group in Investment Management Division at Goldman, Sachs & Co. as an analyst, I conducted research into risk factors in long-short portfolios and took major role in managing an experimental long-short strategy.

I left financial economics because it primary focuses on explaining various existing phenomena and finding ways to back up these explanations with clever manipulation of (usually) noisy data. But I longed to invent solutions to problems. I saw only a limited opportunity to do this in economics.

Publication

Boulat A. Bash,
Post-IPO Flipping and Turnover: Predictive Factors for Long-Run Returns.
Social Science Research Network working paper 623502
May 2001
[ssrn page] [local pdf]

Downloads

TAQAccess is a tool to extract NYSE Transactions and Quotes data into data structures convenient for analysis. It works on all TAQ database CDs/DVDs from its inception in 1993 to present. The source includes examples of using the data structures. Stable version. [taz] [zip]

Excel models developed for Tuck core MBA capital markets class: BondBuilder [xls], FrontierBuilder [xls], and BetaBuilder [xls].

Personal

I am fond of watching baseball, and am a big fan of the Boston Red Sox. The incomplete (through 2007) listing of the Red Sox (and their minor league affiliates) games that I saw in person can be found here. In the summer of 2010, I saw a game at all of the MiLB baseball parks in New England.

I am also a Dartmouth and UMass sports nut, in particular ice hockey (and football, to a lesser extent). One of the things I wanted to do before leaving Amherst is to see Dartmouth and UMass hockey games at all of their respective league opponents' rinks. As of November 4th, 2011, I've seen Dartmouth hockey play at all ECAC arenas, and UMass hockey at all Hockey East arenas. You can see which other sports venues I have been to here.

Here are a couple of other Red Sox and baseball resources:

My other hobbies include cross-country skiing, running, swimming, listening to heavy metal, and reading about military history.

I also help out Green Decade Coalition/Newton with web work whenever I can.