Multiple-input multiple-output wireless system designs with imperfect channel knowledge

Abstract

Employing multiple transmit and receive antennas for wireless transmissions opens up the opportunity to meet the demand of high-quality high-rate services envisioned for future wireless systems with minimum possible resources, e.g., spectrum, power and hardware. Empowered by linear precoding and decoding, a spatially multiplexed multiple-input multiple-output (MIMO) system becomes a convenient framework to offer high data rate, diversity and interference management. While most of the current precoding/decoding designs have assumed perfect channel state information (CSI) at the receiver, and sometimes even at the transmitter, in this thesis we will design the precoder and decoder with imperfect CSI at both the transmit and the receive sides, and investigate the joint impact of channel estimation errors and channel correlation on system structure and performance. The meansquare error (MSE) related performance metrics will be used as the design criteria. We begin with the minimum total MSE precoding/decoding design for a single-user MIMO system assuming imperfect CSI at both ends of the link. Here the CSI includes the channel estimate and channel correlation information. The closed-form optimum precoder and decoder are determined for the special case with no receive correlation. For the general case with correlation at both ends, the structures of the precoder and decoder are also determined. It is found that compared to the perfect CSI case, linear filters are added to the transceiver structure to balance the channel noise and the additional noise caused by imperfect channel estimation, which improve system robustness against imperfect CSI. i Furthermore, the effects of channel estimation error and channel correlation are coupled together, and are quantified by simulations. With imperfect CSI at both ends, the exact capacity expression for a single-user MIMO channel is difficult to obtain. Instead, upper- and lower-bounds on capacity have been derived, and the lower-bound has been used for system design. The closed-form transmit covariance matrix for the lower-bound has not been found in literature, which is referred to as the maximum mutual information design problem with imperfect CSI. Here we transform the transmitter design into a joint precoding/decoding design problem. The closed-form optimum transmit covariance matrix is then derived for the special case with no receive correlation, whereas for the general case with non-trivial correlation at both ends, the optimum structure of the transmit covariance matrix is determined. The close relationship between the maximum mutual information design and the minimum total MSE design is discovered assuming imperfect CSI. The tightness and accuracy of the capacity lower-bound is evaluated by simulation. The impact of imperfect CSI on single-user MIMO ergodic channel capacity is also assessed. For robust multiuser MIMO communications, minimum average sum MSE transceiver (precoder-decoder pairs) design problems are formulated for both the uplink and the downlink, assuming imperfect channel estimation and channel correlation at the base station (BS). We propose improved iterative algorithms based on the associated Karush-KuhnTucker (KKT) conditions. Under the assumption of imperfect CSI, an uplink–downlink duality in average sum MSE is proved, which is often used to simplify the more involved downlink design. As an alternative for solving the uplink problem, a sequential semidefinite programming (SDP) method is proposed. Simulations are provided to corroborate the analysis and assess the impacts of channel estimation errors and channel correlation at the base station on both the uplink and the downlink system performances.