SAGE based semi-blind channel estimation technique for massive MIMO system

In this paper, we have addressed the problem of channel estimation in massive MIMO systems, because the large scale benefits of massive MIMO relies on the accuracy of channel state information (CSI) available with the system. We have proposed an iterative space-alternating generalized expectation maximization (SAGE) based semi-blind channel estimation technique. The proposed method iteratively updates the initial estimate (obtained by the pilot based maximum-likelihood estimation (MLE)) with the help of a SAGE algorithm on pilot and data symbols. The method improves the CSI accuracy without the addition of extra pilot symbols, and converges in almost two iterations which is shown through simulations. The performance of the proposed method is compared with the existing estimators in terms of computational complexity, mean-squared error (MSE), and bit error rate (BER). Simulation results show that the proposed semi-blind estimator achieves significant gain over the existing pilot based estimators. Further, Cramer-Rao lower bound (CRLB) is derived to validate the performance efficacy of the proposed estimator.