Distributed versus centralized zero-forcing precoding for multiceli OFDM systems

In this paper we propose distributed linear multiuser precoding techniques for the downlink of multicell MIMO-OFDM systems. The precoder is designed in two phases: first the intercell interference is removed by applying a set of precoding vectors based on zero-forcing criterion and computed in a distributed manner at each BS; then the system is further optimized through power allocation. Three power allocation algorithms with per-BS power constraint and different complexity tradeoffs are proposed: one optimal to minimize the average BER and two suboptimal. The performance of the proposed schemes is compared to the full centralized multicell based approaches recently proposed, and evaluated considering typical pedestrian scenarios. Numerical results show that the proposed suboptimal schemes achieve a performance very close to the optimal. Also, it is shown that for the same degrees of freedom the performance is close to the one of centralized precoding schemes but with much lower feedback load over the backhaul network.