Cooperation among multiple multiple-input-multiple-output (MIMO) relays has great potential to improve the quality of service (QoS) for the future wireless networks. However, many properties of multi-relay multi-user (MRMU) networks remain to be eluci...
Cooperation among multiple multiple-input-multiple-output (MIMO) relays has great potential to improve the quality of service (QoS) for the future wireless networks. However, many properties of multi-relay multi-user (MRMU) networks remain to be elucidated. Two schemes to determine the signal processing matrices such as BS
transmitter, relay precoders and user receivers achieving sum mean square error minimization (SMSE) for a downlink MRMU MIMO network are proposed. We denote two schemes as joint BS and multiple relay (JBMR). They are classified according to whether the power
constraints are placed on total relays or individual relays. Since it is very difficult to determine the optimal solution due to the non-convex structure of the problem, we propose suboptimal iterative sequential designs. In the first case of relay sum power constraint (RSPC), we extend uplink-downlink SMSE duality of the single relay aided multi-user (SRAMU) MIMO network to the case of a multiple relay aided environment and exploit the Karush-Kuhn-Tucker (KKT) theorem in convex optimization theory to minimize the SMSE. In the second case of per relay power constraint (PRPC), the minimum SMSE design problem is solved directly by the KKT theorem. The numerical results verify that the JBMRs provide comparable performance to that of a full relay cooperation bound (FRCB) method while outperforming the simple amplify-and-forward (SAF) and minimum mean square error
(MMSE) relaying in terms of not only SMSE, but also the sum rate. In addition, we are interested in more practical transceiver design in a relay network. Thus, we propose a transceiver design using local channel state information (CSI) at relay in this thesis. We
denote the proposed algorithm as joint BS and distributed multi-relay (JBDMR). The design criterion is to minimize the user SMSE under RSPC where only local CSIs are available at relays. Local CSI at a relay is defined as the CSI of the channel between BS and
the relay in the $1^{\rm{st}}$ hop link, and the CSI of the channel between the relay and all users in the $2^{\rm{nd}}$ hop link. Exploiting BS transmitter structure which is concatenated with block diagonalization (BD) precoder, each relay's precoder can be
determined using local CSI at the relay. The proposed scheme is based on the sequential iteration of two stages: the stage 1 determines BS transmitter and relay precoders jointly with SMSE duality, and the stage 2 determines user receivers. The proposed
scheme can be theoretically demonstrated to always converge. We verify that the proposed scheme outperforms SAF, MMSE relaying, and an existing good scheme of \cite{ref:Cho} in terms of both SMSE and sum rate performances. Moreover, it is demonstrated that the gap of
sum rate between JBDMR and JBMR is small in low signal to noise ratio (SNR) region and thus JBDMR is useful.