Energy-Efficient Precoder Design for Multi-User MIMO Systems With a Full-Duplex Base Station

Abstract

In this paper, we investigate the downlink (DL) and uplink (UL) precoder design for a multi-user multiple-input multiple-output (MU-MIMO) network consisting of a full-duplex (FD) base station (BS) and multiple users operating in the half-duplex mode. Our goal is to find the precoders that maximize the overall energy efficiency (EE). However, the design problem is intractable to solve due to the fractional optimization formulation. Besides, the self and co-channel interferences also aggravate the design difficulty. To solve the problem, we first apply the UL-DL duality to have the reformulated UL and dual-UL precoder optimization problems. Applying the Dinkelbach's algorithm, we show that both the UL and dual-UL-precoders can be derived with the proposed iterative water-filling (IWF) algorithm. Finally, the DL precoders are obtained by exploiting the duality property again. Note that the proposed precoders are of closed-form expressions. Simulation results justify the superiority of the proposed scheme over the existing design.

Keywords

• Multi-user multiple-input multiple-output (MU-MIMO)
• full-duplex (FD)
• self-interference (SI)
• co-channel interference (CCI)
• fractional programming
• iterative water-filling (IWF)