Max-min fairness based linear transceiver-relay design for MIMO interference relay channel

In this work, the authors investigate a multi-user multi-input multi-output (MIMO) interference relay system, where several transmitter nodes simultaneously communicate with their respective receiver nodes through half-duplex MIMO amplify and forward (AF) relay nodes. For this system configuration, they address the problem of linear transceiver-relay design to achieve max-min fairness among all the users' data streams. Assuming perfect channel state information, this problem is formulated as the maximisation of the minimum signal-to-interference-plus-noise ratio (SINR) per stream among all the users subject to transmit power constraints at the transmitter and relay nodes. Since the formulated problem of jointly optimising all the transmitters, relays and receivers is non-convex, a globally optimal solution cannot be easily obtained. Therefore, they propose a quasi-optimal iterative algorithm to jointly optimise the transmitter precoders, relay precoding matrices and receiver filters for all the users and relays such that the minimum per stream received SINR is improved at each iteration. Simulation results show that the proposed algorithm improves fairness among data-streams of all users and outperforms existing strategies in terms of minimum user rate and bit error rate.