Adaptive signaling for closed-loop MIMO communications

Multiple-input and multiple-output (MIMO) communication systems that adapt to the channel state at the transmitter (called closed-loop MIMO) offer numerous rate and reliability benefits. The channel state information (CSI) is typically delivered through a feedback link from the receiver to the transmitter. While feedback is well studied for independent Rayleigh block fading channels, there has been only limited work addressing how the closed-loop MIMO techniques should be modified when the channel is non-independent Rayleigh block fading. First, we discuss single user MIMO transmission schemes that account temporal correlation statistics and antenna polarization effects into feedback design when partial CSI is available at the transmitter. These include an adaptive limited feedback beamforming technique for dual-polarized MIMO channels optimized by minimizing the power leakage caused by polarization mismatch between the transmit and receive antennas, an adaptive differential feedback design for temporally correlated MIMO channels where the "amount" of the perturbation added to the previous precoder is characterized by directional variation statistics, and optimal allocation of the feedback bits and feedback period to maximize the average beamforming gain. Moreover, we broaded our scope to network MIMO to explore efficient MIMO signaling schemes with a perfect CSI assumption. To understand the fundamental limitation of the network MIMO, we first characterize the degrees of freedom outer bound of the multicell MIMO multiple access channel (MAC) and develop linear achievable schemes that attains the degrees of freedom outer bound.