Effects of imperfect CSIT on downlink MU-MIMO fair SLNR scheduling algorithm

Indoor downlink communication contributes to a large part of the data traffic generated in today's world. The demand for high rate wireless indoor coverage while providing complete support to the existing wireless technologies is a big challenge. The future proof centralized optical fiber indoor back haul for efficient indoor coverage is gaining a lot of attention recently. Physical layer techniques like multiuser MIMO (MU-MIMO) is becoming an inevitable approach in this regard. Quality of service serves as the most attractive feature that should be ensured among the mobile terminals (MTs). In this work we try to study the effects of imperfections in channel knowledge owing to estimation errors. We study the effects of it on our previously proposed throughput fair successive signal to leakage and noise ratio (SLNR) precoding algorithm [1] for such a fiber-wireless (Fi-Wi) MU-MIMO indoor. Our algorithm has been shown previously to achieve maximal fairness. In this paper we use MMSE alterations to significantly reduce the effect of estimation errors for our scheme. We provide a lower bound on the sum rate achievable with imperfect CSIT. In this paper we also compare different power allocation policies. Power allocation policies plays an important part in improving the performance of the system in terms of BER of the worst case user. We show that channel adaptive power allocation guarantees approximately 3dB SNR improvement over the equal power allocation policy for BER of 10-7.