Optimal Training Design for Individual Channel Estimation in Two-Way Relay Networks

This correspondence considers the optimal training design in a classical three-node amplify-and-forward two-way relay network (TWRN) that targets at estimating the individual channel between each source node and the relay node. The transmission environment is assumed to be frequency selective and the orthogonal-frequency-division multiplexing (OFDM) modulation is adopted. We derive the Bayesian Cramér-Rao bound (CRB) for the individual channel estimation, from which the optimal training is obtained. Extensive numerical results are provided to corroborate the proposed studies.