Joint Precoding and Predistortion Scheme for Massive MIMO Transmission Systems With Nonlinear Power Amplifiers

Digital predistortion (DPD) technology can compensate for the nonlinear distortion when the power amplifier (PA) works in the saturated region. To achieve linearity of each radio frequency (RF) link, it is necessary to configure the DPD separately for each PA, which results in the increasing hardware complexity and power consumption. In this paper, a novel nonlinear distortion suppression scheme combining precoding and predistortion is proposed for maximizing spectral efficiency (SE) as well as reducing hardware complexity and power consumption in massive multiple-input multiple-output (MIMO) transmission systems. To realize predistortion, a partially connected intermediate frequency (IF) architecture is designed, where DPDs are configured for partial RF links based on a distortion power-aware configuration (DPAC) method. To compile precoding, the SE maximization problems with total power constraints (TPC) and per-link power constraints (PLPC) are formulated based on the partially connected IF architecture. Moreover, a local optimal solution is obtained by a gradient projection method. Compared with the traditional zero-forcing precoding scheme, the proposed scheme can reduce the number of DPDs by up to 86.2% with the same SE and the RF front-end energy efficiency (EE) can be improved by up to 39.5% with the same number of DPDs.