Adaptation in Multiple Input Multiple Output Systems with Channel State Information at Transmitter

The thesis comprises two parts: the rst part presents channel-adaptive techniques to achieve high spectral e ciency in a single user multiple-input multiple-output (MIMO) system; the second part exhibits a programmable and recon gurable software-de ned-radio workbench (SDR-WB) in the Matlab/Octave environment that accommodates a variety of wireless applications. In an attempt to achieve high spectral e ciency, an adaptive modulation technique is applied at the transmitter to vary the data rate depending on the channel state information (CSI). To further enhance the spectral e ciency, adaptive power allocation schemes are applied in the spatial domain to adjust the power on every transmit antenna. We analyze several power control schemes subject to a peak power constraint to maximize the spectral e ciency given an instantaneous target bit-error-rate (BER). A novel power allocation strategy is proposed to achieve high spectral e ciency with relatively low complexity. In addition, adaptive techniques that switch across di erent MIMO schemes enables even higher spectral e ciency by choosing the scheme with the highest spectral e ciency for a given signal to noise ratio (SNR). We propose a new method to switch between spatial multiplexing with zero-forcing (ZF) detection and orthogonal space-time block coding (OSTBC). This is done by exploiting closed form expressions of the spectral e ciencies discrete rate spectral e ciency (DRSE) and nding the crossing point of the two curves of spectral e ciency. The proposed adaptation scheme adds little complexity to the transmitter since it requires only statistical information of the channel, which does not change as time evolves. Software De ned Radio (SDR) has received more and more interest recently as a promising multi-band multi-standard solution for transceiver design. In order to support various wireless applications, we build a programmable and recon gurable workbench, namely the SDR-WB, in Matlab/Octave environment. The workbench is functionally modularized into generic blocks to facilitate fast development and veri cation of new algorithms and architectures. The modulation formats that are currently supported by the SDR-WB are MIMO, Orthogonal frequency-division multiplexing (OFDM), MIMO-OFDM, DS-CDMA. keywords: MIMO, adaptive modulation, power allocation, spectral e ciency, CSI.