Optimum training sequence design for multi-input multi-output (MIMO) two-way wireless relay networks

In this paper, we consider a multi-input multioutput (MIMO) two-way wireless relay network in which two users, U₁ and U₂, exchange their information via a relay station, R. Single carrier cyclic prefix (SCCP) is used in the system to combat intersymbol interference (ISI) in frequency-selective fading channels. The transmission process is divided into two phases. At phase 1, both users send their information to R simultaneously. R then amplifies and transmits its received signals at Phase 2. We propose an algorithm to estimate the channel information at end users based on the least-square principle. To further minimize the mean square error (MSE) of the estimate, a method to design the optimal training sequences is also proposed. The proposed method does not estimate individual channels between each end user and R but incorporates them into composite channels. Simulation results show that the performance achieved by our optimal design is close to that with perfect channel information.