Events

 
PhD Thesis Presentation
Feedback Strategies and Transceivers Design in Multi-User Multi-Antenna Systems with Limited Feedback

by Miss Yuxian ZHANG

Date
 :  31 Jul 2013 (Wed)
Time
 :  10:30am
Venue  :  Room 4503, 4/F (Lifts 25-26), HKUST

Examination Committee
Prof Lilong CAI, MECH/HKUST (Chairman)
Prof Roger S K CHENG, ECE/HKUST (Thesis Supervisor)
Prof Angela Ying-Jun ZHANG, Department of Information Engineering, The Chinese University of Hong Kong (External Examiner)
Prof Ross MURCH, ECE/HKUST
Prof Chi Ying TSUI, ECE/HKUST
Prof Brian K W MAK, CSE/HKUST
 
Abstract
Multi-antenna wireless systems can provide more reliable transmission and higher spectral efficiency. Many techniques that realize these advantages rely heavily on the channel state information (CSI) at the transmitter (CSIT). CSIT can be obtained through a feedback channel. In our research, we focus on developing feedback strategies and transceiver designs in different multi-antenna systems with limited feedback.

We consider the point-to-point multiple-input single-output (MISO) channel and assume that the receiver performs transmit beamforming using a subset of antennas with power gains exceeding a threshold. The optimal threshold that maximizes the capacity is characterized when the number of transmit antennas is asymptotically large. We also obtain numerically the optimal threshold for systems with a finite number of antennas. Inspired by the asymptotic case, we propose a scheme to select the number of transmit antennas for beamforming.

The multi-user (MU) MISO interference channel is also studied. For the two-user case, a transmit antenna selection scheme based on the channel gains of desired links is proposed to select the subset of transmit antennas. At high signal-to-noise ratio (SNR) region, we analyze the expected signal-to-interference-plus-noise ratio (SINR) and the rate performance. Both the SINR analysis and simulations suggest that using two transmit antennas achieves best performance when the number of feedback bits is large enough.

Finally, we investigate the MU multiple-input multiple-output (MIMO) systems with interference alignment (IA). Previous works seldom pay attention to the issue of quantization complexity. We propose a quantization approach with manageable complexity and evaluate the rate loss relative to the perfect CSI case. This shows how the numbers of feedback bits affects the numbers of transmit and receive antennas that should be used. An iterative algorithm is introduced to jointly optimize the design of the front-end receive filter and the allocation of the feedback bits among different interference links.
 
 
*** ALL ARE WELCOME !! ***