Superimposed Pilots based Estimation of Sparse Multipath massive-MIMO NOMA Channels

This paper presents two new schemes for the estimation of channels encountered in the scenarios exhibiting frequency-selective sparse-structured channel response in uplink of massive Multiple-Input Multiple-Output (MIMO) Non-Orthogonal Multiple Access (NOMA) systems. A spectrally efficient Superimposed Training (SiT) based estimation approach is adopted, coupled with Compressive Sensing (CS) techniques to exploit the sparsity of the channels. Moreover, a Minimum Mean Square Error-Successive Interference Cancellation (MMSE-SIC) equalizer design is exploited to counter the nonorthogonality between the users. The performance evaluation and comparative analysis of the proposed techniques are conducted in terms of Normalized Channel Mean Square Error (NCMSE) and Bit Error Rate (BER) under various channel, noise, Training-to-Information power Ratio (TIR) conditions. The conducted comparative analysis of the proposed techniques has demonstrated a significant gain over the baseline Least Squares (LS) channel estimation technique. Furthermore, it is established that MMSE-SIC outperforms the MMSE equalizer in the considered communication scenario.