Channel estimation and training design for two-way relay networks with power allocation

Abstract

In this paper, we propose a new channel estimation prototype for the amplify-and-forward (AF) two-way relay network (TWRN). By allowing the relay to first estimate the channel parameters and then allocate the powers for these parameters, the final data detection at the source terminals could be optimized. Specifically, we consider the classical three-node TWRN where two source terminals exchange their information via a single relay node in between and adopt the maximum likelihood (ML) channel estimation at the relay node. Two different power allocation schemes to the training signals are then proposed to maximize the average effective signal-to-noise ratio (AESNR) of the data detection and minimize the mean-square-error (MSE) of the channel estimation, respectively. The optimal/sub-optimal training designs for both schemes are found as well. Simulation results corroborate the advantages of the proposed technique over the existing ones.