Spectrally Efficient Channel Estimation for Asynchronous Amplify-and-Forward Two-Way Relay Networks

In this paper, we consider the problem of channel estimation for asynchronous amplify-and-forward (AF) two-way relay networks. We propose a novel semi-blind channel estimation algorithm for this problem based on the expectation-maximization (EM) framework. The proposed EM algorithm has low complexity, and only a small number of EM iterations are needed to achieve convergence. The semi-blind Cramer-Rao bound (CRB) for channel estimation in asynchronous AF two-way relay networks is also obtained. Using simulations, we show that the proposed EM algorithm significantly outperforms pilot-based estimation by using only a limited number of received data samples in addition to the pilot samples. Furthermore, the achieved mean-squared error performance almost overlaps with the obtained CRB. Finally, a semi-blind generalized likelihood ratio testing (GLRT) method is proposed to tackle sequence arriving order (SAO) detection at the terminals and is shown to yield a higher probability of detection than the pilot-based GLRT for SAO detection.