Pilot decontamination based on superimposed pilots assisted by time-multiplexed pilots in massive MIMO networks

Pilot contamination has been recognized as one of the main limiting factors in massive multiple-input multiple-output (MIMO) since it damages the acquisition of accurate channel state information (CSI) at base stations. In this work, we propose a novel pilot decontamination approach in a massive MIMO network where the time division duplex operation is assumed. In our proposed approach, user equipments are allowed to use superimposed (SP) pilots in combination with time-multiplexed (TM) pilots to estimate the CSI prior to downlink data transmission. Specifically, the contaminated channel estimates obtained by TM pilots are used to reduce the amount of interference caused by transmitting pilots together with uplink data, which is a typical problem in SP pilots-based channel estimation. Through numerical results, we show that the interference that emerges from transmitting pilots with data is substantially reduced by our proposed approach. This reduction, in combination with the reduced overhead in TM pilots transmission, leads to an improved spectral efficiency achieved in massive MIMO networks.