|
Closing the gap with Limited Feedback
Why Limited Feedback?
A MIMO channel
promises multiplicative gain in capacity and matching gains in channel
reliability. This means that 2 antennas at the transmitter and 2 at the receiver
can achieve 2 times the data rate compared to a single antenna system and 4
times the reliability. When the
transmitter is informed about the propagation channel, the benefits of MIMO
communication can be more easily achieved with lower complexity. When the
channel is known at the transmitter, communication is possible on the eigenmodes
of the channel. Thus information can be tailored to the subspace structure of
the matrix channel. In time division duplex systems it is reasonable to assume
channel state information is available though this information is rarely
perfect. Calibration errors also take their toll. Therefore it is of interest to
study MIMO communication systems where limited feedback is available. In our
research, we have developed a framework for informing the transmitter about the
channel state using a low-data-rate feedback channel. Our approach estimates the
critical parameters of the channel at the receiver, quantizes then, and then
sends them to the transmitter using a low-rate feedback channel. Limited
feedback MIMO can fulfill the MIMO promise!
Principle
- ãWe
donât need all the information from the channel all the time!ä
á Identify the channel information ö what to
feedback and how often to feedback?
á
How to
compress the channel information? öwe can compress
the channel information upto 10 fold!!
á
How to
use the channel information at the transmitter? ..optimally
What
do we do?
Currently we are investigating a number of methods for
quantizing channel state information using subspace quantization techniques and
relaying this information to the transmitter. We are investigating a variety of
space-time coding techniques including space-time block codes, beamforming, and
spatial multiplexing systems. We use tools from comm. theory, signal processing, computer
science and applied mathematics in our research focused in the following
directions:
á
Limited
Feedback Precoding Framework to
compress channel information and utilize at the
transmitter.
á
Multi-mode
Antenna Selection to
exploit the independence of multiple channels and match the transmission
to channel conditions.
á
Broad band MIMO (OFDM) and Multi-user
MIMO extension of limited feedback.
What
have we done?
Our
group has demonstrated superior gains in channel performance using limited
feedback precoding as well as multi-mode antenna selection techniques. We have
introduced Grassmannian Subspace Packing based limited feedback methods and have
demonstrated their performance in broad-band and multi-user MIMO.
Follow the links to the left for more details in limited feedback research and related publications.
This material is based upon work supported by the National Science Foundation under Grant No. 0514194, Freescale, and the Office of Naval Research under grant number N00014-05-1-0169.
Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation, Freescale, or the Office of Naval Research.
|
