Beamformer design for interference alignment in full-duplex cellular networks with imperfect CSI

In this work we focus on a cellular network where a full-duplex (FD) base-station (BS) communicates with multiple half-duplex (HD) downlink (DL) and uplink (UL) users. The introduction of FD capability at the BS causes a surge in the amount of interference compared to the HD network counterpart. In particular a new type of co-channel interference arises between UL and DL data, since they are transmitted simultaneously. Here, we consider the use of linear interference alignment (IA) to manage interference in such networks, under the availability of imperfect channel state information (CSI). We design two novel IA algorithms, the first is based on maximizing the signal-to-interference-plus-noise ratio (Max-SINR) and the second minimizes the mean squared error (MMSE). Both algorithms, 1) take into effect statistical knowledge of the CSI error for added robustness, 2) follow specific design principles that distribute the different interference components amongst the various beamformers, and 3) result in unitary precoders and receivers. Additionally, we show that under certain conditions the proposed Max-SINR and MMSE designs obtain identical beamformers.