A Semi-Blind Channel Estimation Algorithm for One-bit Massive MIMO Systems

Massive multiple-input-multiple-output (MIMO) uses a large number of antennas and radio frequency (RF) chains which incur a huge cost and power consumption. Since high-resolution analog-to-digital converters (ADCs) consumes a major portion of the RF chain circuit power, so one-bit massive MIMO systems are seen as one of the potential solutions to reduce the power consumption and cost associated with the RF chains. Our paper addresses channel estimation issue which is among one of the crucial needs for the practical realization of one-bit massive MIMO systems. The pilot-aided channel estimator demands additional pilots to improve the estimation accuracy which in turn reduces the spectral efficiency of the system. To overcome this, we propose an iterative semi-blind based channel estimator for one-bit massive MIMO systems. The proposed algorithm consists of two stages: initialization and iteration. The initial channel estimate is obtained from the pilot based initialization stage, which is refined further in the iteration stage with the help of both pilot and few data symbols. So, the semi-blind estimator improves estimation accuracy without the addition of extra pilot symbols into the system. Through simulations, we show that the proposed scheme achieves significant improvement against the existing pilot based estimators in terms of estimation accuracy and bit error rate (BER) at the cost of a nominal increase in the computational complexity. Further, the proposed algorithm attains convergence in almost one iteration for all the considered scenarios of one-bit massive MIMO system. Thus, the semi-blind estimator is spectral and power efficient in comparison to the existing pilot based algorithms.