HKUST Library Institutional Repository Banner

HKUST Institutional Repository >
Electronic and Computer Engineering  >
ECE Doctoral Theses >

Please use this identifier to cite or link to this item:
Title: Adaptive downlink multi-user MIMO wireless systems
Authors: Wang, Cheng
Issue Date: 2007
Abstract: Recent years have witnessed fast and tremendous growth in the wireless communication industry. Future wireless communication systems are expected to provide a wide variety of new services, from high-quality voice to high-definition video and wireless multimedia for anyone, anywhere at anytime. This results in ever increasing demands for high data rate through harsh wireless channels over scarce frequency spectrum and is the key challenge in wireless system design. In this thesis, new and novel downlink multi-user multiple-input multiple-output (MU-MIMO) systems are described and investigated. By multi-user system we refer to systems that can serve multiple users simultaneously in both frequency and time. The ultimate goal of this work is to 1) investigate the impact of various realistic considerations, e.g. spatial fading correlation, imperfect channel state information (CSI), on system performance and develop adaptive algorithms to improve system performance under these realistic considerations; 2) investigate methods to further improve the spectrum efficiency by exploiting the synergy between the PHY (physical) layer and the MAC (Multiple Access Control) layer. In total five new and novel major contributions are presented in the thesis. Firstly, the impact of spatial fading correlation on the performance of downlink MU-MIMO systems is described and investigated. By exploiting the subspace structure of the channel, our work focuses on the impact of the channel characteristics on the performance of MU-MIMO formulations and suggests two adaptive algorithms to improve performance. The next contribution concerns jointly considering the queueing state information from the MAC layer and the channel state information from the PHY layer. Two optimal cross MAC-PHY layer MU-MIMO scheduling algorithms are proposed, where the optimality of the algorithms lies in that they can achieve maximum system throughput while at the same time guaranteeing the stability of the system. Motivated by the fact that CSIT has to be sent back to the BS through capacity-limited feedback channels in FDD systems, we propose a realistic framework for MU-MISO transmission with limited feedback, where the useful information that enables the BS to perform efficient multi-user transmission is identified and two feedback strategies are proposed accordingly. We then analyze the impact of imperfect CSIT on the system performance in terms of sum mutual information scaling law for downlink MU-MIMO decomposition transmission and it is found that imperfect CSIT has different impacts on the mutual information scaling law for different categories of system configuration. Finally, due to the fact that imperfect channel estimates arise in any practical estimation scheme, it is therefore important to be able to characterize the performance of MIMO receivers in the presence of channel estimation error. As a preliminary result, we evaluate the impact of channel estimation error on the performance of single-user MIMO Zero-forcing receiver. In particular, the post-processing SNR, outage probability and BER expressions for MQAM and MPSK modulation are derived in closed form. Numerical results and comparisons are provided to demonstrate the potential of our proposed techniques and the tightness of our analysis. In particular, it is shown that our proposed algorithms substantially improve the spectrum efficiency of the downlink MU-MIMO systems. In addition, with efficient adaptive scheme and appropriate system setting, multi-user MIMO system is still promising even if CSIT is not perfectly known to the BS.
Description: Thesis (Ph.D.)--Hong Kong University of Science and Technology, 2007
xiv, 150 leaves : ill. ; 30 cm
HKUST Call Number: Thesis ECED 2007 Wang
Appears in Collections:ECE Doctoral Theses

Files in This Item:

File Description SizeFormat

All items in this Repository are protected by copyright, with all rights reserved.