Skip to Main Content
In this paper, we consider optimal power allocation (OPA) for distributed space-time coded two-way relay networks. Each relay transmits a scaled version of the linear combinations of the received symbols and their conjugates, and the scaling factor is based on automatic gain control (AGC) at the relays. We show that solving the OPA across relays to minimize the average conditional PEP of the destination terminals is a generalized linear fractional programming problem, which can be resolved by the Dinkelbach-type procedure. We also prove that at most two relays are active while the others keep silent if the sum power constraint is made across the relays. This motivates us to propose a new low-complexity relaying scheme that uses distributed Alamouti codes on the selected two-best relay nodes. Simulation results show that the distributed space-time codes (DSTC) with the OPA and the proposed scheme have significant performance gains over the DSTC with the equal power allocation.