Analysis of a semi blind pilot decontamination method in massive MIMO

Recently, a semi-blind pilot decontamination based method is proposed in [1] for multicell multiuser massive multiinput multioutput (MIMO) systems. In [1], to reduce the effect of the pilot contamination, a semiblind channel estimation method that does not require cell cooperation or statistical information of the channels is proposed. The blind pilot decontamination method in [1] is reformulated as a constrained optimization problem. The constrained minimization problem of the mixing matrix is then carried out by a generalized robust independent component analysis (ICA) based deflation method. The key demerit of the proposed deflation method is that it suffers from error accumulation. In this paper, we investigate semi blind pilot decontamination method in [1] aiming to analyze the role of the deflation technique used in this method. Our analysis reveals that the error accumulation is a major drawback found in this algorithm. Furthermore, the overall performance improves without the deflation method in the semi blind pilot decontamination. The effectiveness of our analysis is verified by extensive simulations considering different noise levels.