Selection diversity techniques in MIMO and cooperative relaying
Shiva Prakash Premananda
Date of Issue2011
School of Computer Engineering
Centre for Multimedia and Network Technology
The primary objective of this thesis revolves around investigating MIMO (multiple-input multiple-output) transmit antenna selection (TAS) in delay constrained networks. In this thesis, for various different scenarios of TAS arrangement, channel prediction is explored as a tool to recover some of the system gain lost due to delayed switching. In use, TAS requires at least partial channel knowledge at the transmitter in order to perform selection. In this work, performance degradation due to outdated channel knowledge is determined analytically, and related to channel characteristics. Performance gains of several different antenna placements are presented and trade-offs are analysed with respect to use of prediction. Furthermore, adaptive modulation is also explored in a TAS scenario (allied to maximal ratio combining), and again channel prediction is employed at the receiver to provide estimates of future best transmission states, which include selecting the best antenna as well as the best supported modulation scheme, optimised for the given objectives. This thesis then extends these ideas into the interesting area of cooperative relaying. The principles of selection diversity are applied as we investigate partial relay selection in amplify-and-forward relaying for a variety of system configurations, and the effects of the multiple antennas are analysed when used either at the source, relay or destination terminal. Since selection is usually delay limited, we also consider the benefit of incorporating a predictive relay selection scheme to improve performance. To validate and test advanced MIMO algorithms over real-world channels, an experimental architecture was designed for a reconfigurable multi-antenna test-bed. This 4 × 4 MIMO FPGA (field programmable gate array) base-band architecture is presented, and verified by MIMO loop-back testing. The setup is useful in validating future research work relating to MIMO and antenna selection, and also in developing efficient reconfigurable architectures that are demonstrable over physical channels.
DRNTU::Engineering::Electrical and electronic engineering::Wireless communication systems