In this paper, the symbol error rate (SER) and the bit error rate (BER) of orthogonal space-time block codes (OSTBCs) and their achievable capacity over an amplify-andforward (AF) relay channel with multiple antennas at source and destination and single antenna at relay node are investigated. Considered are receive antenna selection, transmit antenna selection, and joint antenna selection at both the transmitter and the receiver. The exact SERs of OSTBCs for M-PSK and square M-QAM constellations are obtained using the moment generating functions (MGFs). Also, we analyze the achievable capacity over such channels assuming antenna selection is done at the source and relay nodes. We show that a small number of selected antennas can achieve the capacity of the system in which no channel state information (CSI) is available at the source and relay nodes.

The degrading effect of RF impairments on the performance of wireless communication systems is more pronounced in MIMO-OFDM transmission. Two of the most common impairments that significantly limit the performance of MIMO-OFDM transceivers are IQ-imbalance and phase noise. Low-complexity estimation and compensation techniques that can jointly remove the effect of these impairments are highly desirable. In this paper, we propose a simple joint estimation and compensation technique to estimate channel, phase noise and IQ-imbalance parameters in MIMO-OFDM systems under multipath slow fading channels. A subcarrier multiplexed preamble structure to estimate the channel and impairment parameters with minimum overhead is introduced and used in the estimation of IQ-imbalance parameters as well as the initial estimation of effective channel matrix including common phase error (CPE). We then use a novel tracking method based on the second order statistics of the intercarrier interference (ICI) and noise to update the effective channel matrix throughout an OFDM frame. Simulation results for a variety of scenarios show that the proposed low-complexity estimation and compensation technique can efficiently improve the performance of MIMO-OFDM systems in terms of bit-error-rate (BER).