Research On Transceiver Design Of MIMO Relay Systems

MIMO (Multiple Input Multiple Output, MIMO) relay technology combined MIMO technology and relay technology, which can provide the system spatial multiplexing gain and spatial diversity gain, while also utilizing the relay technology to improve wireless communication system reliability and communications coverage. In the MIMO relay system, the transceiver design includes the design for the precoding matrix at the source node, the estimate matrix at destination node, and the precoding matrix at the relay node. When take the multi-user MIMO relay system into account, in addition to the noise, there also exists the multiuser interference (MUI). the architecture of the system become more complex and the constraint of the relay node antennas will also bring difficulties to the transceiver design. In this paper, we focus on the multiuser MIMO (MU-MIMO) relay systems, design transceiver schemes for the MU-MIMO relay downlink system and the MIMO multi-pair two-way relay systems.In the MU-MIMO relay downlink system, the traditional Block Diagonalization (BD) algorithm used at relay can reduce the MUI completely, however this algorithm doesn’t take the noise into account, when at the low Signal-to-Noise Ratio (SNR) region, the noise become the major influence on the system performance. In order to reduce the Bit Error Rate (BER) at the low SNR region, we propose a linear transceiver optimization design based on the Regularized Block Diagonalization (RBD) algorithm. Firstly, we decompose the relay transceiver matrix into two parts: the estimate matrix for base station (BS)-relay link and the precoding matrix for relay-multi user link. For the BS-relay link, we design the transceiver scheme in order to minimum mean square error (MMSE) at the first hop. For the relay-multiuser link, the RBD algorithm is used at the relay to reduce the noise as well as MUI, then we design an iterative algorithm which joint the estimate matrix at the users and the relay precoding matrix. Simulation results show that the proposed algorithm can converge to a fixed value in a few iterations, and has a good BER performance at the low signal-to-noise ratio (SNR) region as well as the ergodic capacity performance.In the multi-pair MIMO two-way relay system, system structure is more complex, and the MUI also exists. The existing methods need to know the overall channel state information (CSI), which lead to a high complexity and feedback overhead. We design a transceiver design with low complexity and low feedback overhead. Firstly we use the BD algorithm to transform the multi-pair system into multiple single-pair systems, by this way, when we design the transceiver method for a single-pair users, each user only need to know the CSI of its own pair. Secondly, the precoding design for the single-pair users need to be jointly designed by the iterative algorithm, we present an simplify algorithm for the precoding matrix design at the user, each user design its own precoding matrix independently, and needn’t iterate with the other users. Simulation results show the proposed algorithm speed up the convergence rate and reduce the complexity. Also the system BER and capacity performance used the proposed simply design can be approximate to the joint users precoding design.