Power Allocation for IRS-Aided Two-Way Decode-and-Forward Relay Wireless Network

Abstract

In this paper, an intelligent reflecting surface (IRS)-aided two-way decode-and-forward (DF) relay wireless network is considered, where two users exchange information via IRS and DF relay. To enhance the sum rate performance, three power allocation (PA) strategies are proposed. Firstly, a maximizing sum rate method based on successive convex approximation (Max-SR-SCA) is proposed to jointly optimize the PA factors of user1 ( <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$U_{1}$</tex-math></inline-formula> ), user2 ( <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$U_{2}$</tex-math></inline-formula> ) and relay station (RS). To further improve sum rate performance, a method of maximizing determinant (Max-Det) with a higher sum rate enhancement is presented. Considering the asymmetry of two-way rates caused by channel quality and two users' demand, maximizing sum rate with rate constraint (Max-SR-RC) is put forward. Simulation results show that the proposed three PA methods not only outperform the equal power allocation (EPA) method, but also the sum rates corresponding to the three PA methods are slightly lower than that of optimal exhaustive search (ES) method. Especially for Max-SR-RC method, its rate gain over EPA is up to 64.7% and its sum rate gap with optimal ES is less than 0.1bits/s/Hz. Furthermore, it is verified that the total power and random shadow variable <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$X_\sigma$</tex-math></inline-formula> have a substantial impact on the sum rate.