Beamforming Vector Quantization for Cooperative Transmission

Event Date: June 22, 2010
Speaker: Chun Kin Au Yeung
Speaker Affiliation: Ph.D. Candidate
Sponsor: CNSIP Seminar Series
Time: 2:00 pm
Location: MSEE 239
Contact Name: David Love
We study the beamforming vector quantization and feedback problem for a
cooperative transmission setup, which can be modeled by a potentially large and
varying number of transmit antennas M. The conventional codebook approach has a
complexity that scales exponentially with M, unfavorable when M is large. We
propose to use a trellis-based quantization technique with linear complexity in
$M$ and good performance for moderate values of M. Codebook design criteria is
derived, and three enhancements giving 0.2-0.8 dB gain each are shown. These
enhancements address issues relevant to feedback error, trellis termination,
and unequal path-loss from different transmission points.

Chun Kin received his B.A.Sc degree in computer engineering from University of
Waterloo, Canada, in 2004. He had been pursuing his Ph.D. under Professor David
Love in Purdue University since. His primary research interest is in feedback
design in next generation MIMO-enabled cellular systems. Chun Kin interned with
Mitsubishi Electric Research Laboratory in Cambridge, MA, during the fall of
2008 and with Huawei Technologies, Dallas, TX, during the summer of 2009. Chun
Kin has 17 journal and conference publications, 6 patents, and was the
recipient of the Outstanding Graduate Student Award within the ECE department
in 2009.