Pilot-Decontamination in Massive MIMO Systems Using Interference Alignment

Pilot contamination (PC) is a practical drawback of massive multiple-input multiple-output (mMIMO) systems. It prevents the exploitation of the inherent spectral-efficiency (SE) of mMIMO systems, especially in dense networks. In this letter, we consider multi-antenna users in a multi-cell network and propose a novel approach based on interference alignment (IA) to remove PC. We show that the extra degrees of freedom (DoFs), provided by multi-antennas at the users' side, can be used efficiently when employed for pilot decontamination, especially in heavy-loaded networks. Thus, while the intra-cell interference is removed due to the properties of mMIMO the out-of-cell interference is eliminated by the proposed IA-based pilot decontamination. The proposed algorithm uses linear zero-forcing (ZF) and needs no continuous cooperation between the base stations (BSs). Moreover, the users need no channel state information (CSI). Our simulations show that the proposed method outperforms the conventional PC precoding algorithms significantly, particularly in heavy-loaded networks.