I. Introduction
Wireless transmissions through multiple transmit and receive antennas, or the so-called multiple-input multiple-output (MIMO) channel, have received great attention during the past decade due to its promising capacity gain as compared with transmissions through a single transmit and receive antenna, or the so-called single-input single-output (SISO) channel [1], [2]. On the other hand, orthogonal frequency-division multiplexing (OFDM) is a robust and effective technique for suppressing the intersymbol interference (ISI) inherent in time-dispersive channels [3]. As a result, the incorporation of MIMO with OFDM, termed as MIMO-OFDM, is a promising technique to support “broadband” transmissions over time-dispersive channels and has been suggested for high-speed wireless local area networks (WLANs) in IEEE 802.11n proposals (see, e.g., “wideband-eigenmodes” transmission proposed in [4]). In this paper, we consider wireless transmissions through a MIMO-OFDM channel with transmit antennas, receiver antennas and orthogonal subcarriers. The transmission is assumed to be continuous and on a time-frame basis. The wireless channel of interest is assumed to be “block-fading”, i.e., the channel remains constant within each time frame, but can possibly vary from one frame to another. Furthermore, we assume that the channel state information (CSI) is available at both the transmitter and the receiver. Therefore, a MIMO-OFDM channel can be decomposed into space–frequency subchannels through singular-value-decomposition (SVD)-based transmit and receive eigenbeamforming [5].