Thinh P.Q. Nguyen
34212
Aspen Loop Phone
: (510) 378-3456
Union
City, CA 94587 E-mail
: thinhq@eecs.berkeley.edu
Research
Interests
·
Multimedia networking
·
Signal processing
·
Machine learning
·
Networking
·
Wireless communication
·
Data analysis and mining
Education
Ph.D. Electrical
Engineering and Computer Science,
University
of California, Berkeley, CA December 2003
Dissertation
Title: Path
Diversity Media Streaming
Research
Adviser: Avideh Zakhor
Thesis
Committee: Avideh Zakhor, Kannan Ramchandran, Ion Stoica, Stanley
Klein
Major
Areas: Signal Processing, Multimedia
Networking, and Data Analysis
Minor Areas: Statistics, Machine Learning, and Networking
M.S. Electrical Engineering and Computer Science,
University
of California, Berkeley, CA May 2000
Thesis
Title: Performance Analysis of an
H.263 Video Encoder for IRAM
Thesis
Committee: Avideh Zakhor, Kathy Yelick
B.S. Electrical Engineering and Computer Science,
University
of Washington, Seattle May 1995
Professional Experience
Video and Image Processing Laboratory,
Department of Electrical Engineering and Computer
Science, U.C. Berkeley
Graduate Student Researcher, 09/98 – present
q
Path diversity media
streaming
The current ``best-effort'' Internet does not
guarantee Quality of Service (QoS) such as minimum bandwidth, packet loss rate,
and delay, which are critical to many multimedia applications. As such, many
significant challenges remain to design and deploy delay sensitive multimedia
applications over the Internet effectively.
In my dissertation, I develop a path diversity framework for
concurrent media streaming to a receiver using multiple routes. Without requiring QoS, the path diversity
framework improves the quality of the streamed media via multiple routes
created using either multiple senders or relay nodes, in order to increase
available bandwidth, reduce packet loss and delay. The proposed path diversity framework combines many
approaches from the network architecture and protocols, to source and channel
coding to in order to improve the overall quality of the streamed media.
From an architecture point of view, the
path diversity framework combats packet loss, delay, and insufficient
bandwidth for pre-recorded streaming media by sending packets simultaneously
from multiple senders to a single receiver. For interactive and live streaming
applications, the path diversity framework allows a single sender to
send packets simultaneously on both default and redundant paths to the
receiver. To create a redundant path,
the sender sends packets to the appropriate relay node, which then forwards the
packets to the receiver. The relay node
selection algorithm is designed to ensure that packets traveling through the
relay node take a different underlying physical path than that of the default
path, hence, providing redundancy and protection against network outages and
congestion.
To facilitate concurrent streaming to a receiver from
multiple senders, I develop a transport protocol to synchronize simultaneous
transmission of a media. In particular,
the protocol employs the rate allocation and packet partition algorithms.
The rate algorithm determines the sending rate on each route in order to
minimize the packet loss, while the packet partition algorithm ensures that
each packet is sent by one and only one sender and at the same time, minimizes
startup delay.
From a channel coding perspective, I
show theoretically and experimentally that using Forward Error Correction (FEC)
in streaming the media simultaneously over multiple mostly independent routes
at appropriate sending rates is more effective than using FEC with the
traditional uni-path streaming. I also
provide a robustness analysis of the sending rates, and implement the optimal
FEC strategy in the path diversity framework under various network
conditions.
From a source coding point of view, I design a network
adaptive matching pursuits based multiple description video coding scheme for
the proposed path diversity framework.
Multiple description coding is an error resilient source coding scheme
that generates multiple encoded bit streams of the source with the aim of
providing an acceptable reconstruction quality of the source when only one
description is received, and improved quality when multiple descriptions are
available. The proposed network adaptive multiple description matching pursuits
scheme is designed to optimally split the source into descriptions that are
adapted to the network characteristics of each route, hence providing superior visual
quality.
q Video, Image, and Network Protocol Standards
Video standards: 5+ years of experience with
H.263, H.26L, MPEG2, MPEG4,
Image standards:
JPEG, JPEG2000 standards.
Media Network Protocols: Implement RTP/RSTP for video streaming.
Microcomputer Research Lab
Intel Corporation
Research Engineer, 7/1995
– 8/1998
Graphics and volume visualization: - Design and
develop graphics and volume visualization algorithms for interactive navigation
in human body.
Computer architecture: - Design snooping protocol for caches in
multi-processor Pentium II board.
Computer aid design software: - Design and
develop database tools for circuit designs and VLSI layout.
Graphics Lab
Intel Corporation/Microsoft Corporation
Software
Engineer, 4/1998 – 8/1998
DirectX
: - Optimize geometry engine in
DirectX for Pentium pipeline.
Siemens
Software Engineer, 1/1994
– 9/1994
PBX Switches: - Verify correctness of software/hardware
for advanced PBX switches for central telephone office.
Teaching Experience
Digital Image Processing
Department of Electrical Engineering and Computer
Science, U.C. Berkeley
Graduate-Student Instructor (Spring 2001)
This
graduate-level course covers fundamental multi-dimensional signal processing
concepts, together with practical algorithms for image enhancement,
restoration, analysis, and compression. As a teaching assistant, I hold office
hours, grade homework and projects.
Introduction to Computer Architecture and Programming
Language
Department of Electrical Engineering and Computer
Science, U.C. Berkeley
Graduate-Student Instructor (Fall 1998)
This course is aimed to introduce computer
architecture and programming languages to undergraduate students. As a teaching assistant, I hold weekly
discussion and lab section, design homework and grade examinations.
Computer Graphics
Department of Electrical Engineering and Computer
Science, University of Washington
Teaching Assistant (Spring 1995)
This undergraduate course covers fundamentals of
modeling and rendering in computer graphics, together with various algorithms
such as ray tracing and radiosity. As a
teaching assistant, I hold office hours, grade homework and projects.
Undergraduate Physics and Math
Educational Opportunity Program, University of
Washington
Tutor (1992-1994)
As a tutor,
I help students with homework and understanding of the key concepts in physics
and calculus.
Selected Publications
Books:
·
T. Nguyen and A. Zakhor, “Video Communication Over
Packet-Switched Networks”, Claypool/Morgan, in preparation.
Journals:
·
T. Nguyen and A. Zakhor, “Video Communication Using Path
Diversity System and Matching Pursuits based Multiple Description Video Codec”;
Submitted to IEEE Transactions on Multimedia.
·
T. Nguyen and A. Zakhor, “Protocols and Algorithms for Distributed
Video Streaming”; Accepted for publication in IEEE Transactions on Multimedia.
·
M. Brady, K. Jung, H.T. Nguyen; T. Nguyen, “Interactive
Volume Navigation”; IEEE Transactions on Visualization and Computer Graphics,
vol.4, (no.3), IEEE, July-Sept. 1998. p.243-56
Conferences:
·
T. Nguyen and A. Zakhor, "Matching Pursuits Based
Multiple Description Video Coding for Lossy Environment" (Invited Paper); International
Conference on Image Processing 2003, Barcelona, Spain.
·
T. Nguyen, P. Mehra, A. Zakhor, "Path Diversity and
Bandwidth Allocation for Multimedia Streaming" (Invited Paper);
International Conference on Multimedia and Expo 2003, Baltimore, MD, USA.
·
T. Nguyen and A. Zakhor, "Path Diversity with Forward
Error Correction (PDF) System for Packet Switched Networks"; INFOCOM 2003,
San Francisco CA, USA.
·
T. Nguyen and A. Zakhor, "Distributed Video Streaming
with Forward Error Correction" (Best Paper Award); Packet Video
Workshop 2002, Pittsburgh PA, USA.
·
T. Nguyen and A. Zakhor, "Protocols for Distributed
Video Streaming"; International Conference on Image Processing 2002,
Rochester NY, USA.
·
T. Nguyen and A. Zakhor, "Distributed Video Streaming
over the Internet"; SPIE, Conference on Multimedia Computing and
Networking, San Jose, California, January 2002.
·
T. Nguyen, A. Zakhor, and K. Yelick, “Performance Analysis
of an H.263 Video Encoder for IRAM''; International Conference on Image
Processing, Vancouver, Canada, September 2000.
·
T. Nguyen, “
Performance Analysis of an H.263 Video Encoder for IRAM”; UCB Master Thesis
1999.
·
M. Brady, K. Jung, H.T. Nguyen; T. Nguyen, “Two-phase
Perspective Ray Casting for Interactive Volume Navigation”; Visualization '97
New York, NY, USA: IEEE, 1997.
·
T. Nguyen, “Parallel Implementation of Shear-Warp Technique
for Volume Visualization on Pentium II”; Intel Technical Report, 1996.
Honors and Organizations
·
Best
Paper Award at Packet Video Workshop, 2002
·
Invited
Talk at International Conference on Image Processing, 2003
·
Invited
Talk at International Conference and Multimedia Expo, 2003
·
Berkeley
First Year Fellowship
·
University
of Washington High Scholarship: Freshman, Sophomore, Junior, and Senior
·
University
of Washington Dean’s List: Freshman, Sophomore, Junior, and Senior
Professional Activities
Reviewer for
·
IEEE Transactions On
Circuits and Systems for Video Technology
·
IEEE Transactions On
Multimedia and Networking
·
IEEE Transactions On
Signal Processing
·
IEEE Transactions On
Image Processing
·
Computer Network
Magazine
References
Professor Avideh Zakhor
Office: 507 Cory Hall
Phone: (510) 643-6777
Fax: (510) 642-2739
Email: avz@eecs.berkeley.edu
EECS Dept, U.C. Berkeley, CA 94720
Professor Kannan Ramchandran
Office: 269 Cory Hall
Phone: (510) 642-2353
Fax: (510) 642-9160
Email: kannanr@eecs.berkeley.edu
EECS Dept, U.C. Berkeley, CA 94720
Professor Ion Stoica
Office: 645 Soda Hall
Phone: (510) 643-4007
Fax: (510) 643-1289
Email: istoica@cs.berkeley.edu
EECS Dept, U.C. Berkeley, CA 94720
Dr. Bob Liang, Director of Future Platform Lab
Intel Microprocessor
Research Labs
Phone: 408-765-8925
Email: bob.liang@intel.com
2200 Mission College Blvd
Santa Clara, CA 95052