Channel Estimation using In-Band Pilots for Uplink Massive MIMO

In this paper, we extend the channel estimation technique using in-band pilots (IBP) for uplink DFT-Precoded OFDM single input single output (SISO) to massive multiple input multiple output (MIMO) system and compare its performance with that of using time domain superimposed pilots (SP). We derive analytical expression for the mean square error (MSE) of least square channel estimation using IBP and show that IBP outperforms SP when the data frequencies are removed for the accommodation of pilot frequencies. Moreover, we compare the signal-to-interference ratio (SINR), bit error rate (BER), and peak-to-average power ratio (PAPR) performance of the DFT-precoded OFDM signal with IBP and SP. It was found that the removal of a small portion of the pilot frequencies has negligible effect on the PAPR performance and that IBP gives slightly better BER and SINR performance over SP, depending on the number of antennas and the total number of users in the system.