Search

Inspired by the solid foundations from information theory, non-orthogonal multiple access (NOMA) has rekindled the interests of researchers as a benefit of the recent advancement in signal processing and silicon technologies. However, comprehensive performance analysis on NOMA and practical resource allocation designs to exploit potential gains of NOMA in terms of spectral and energy efficiency have not been fully studied and investigated in the literature. In this thesis, we study performance analysis and resource allocation designs for NOMA in wireless communication systems. First, we investigate the ergodic sum-rate gain (ESG) in uplink communications attained by NOMA over OMA in single-antenna, multi-antenna, and massive-MIMO systems with both single-cell and multi-cell deployments. The distinctive behaviours of ESG under different scenarios are unveiled which provide some important and interesting insights for the practical implementation of NOMA in future wireless networks. Second, in practice, the successive interference cancelation (SIC) decoding of uplink NOMA suffers from the error propagation. To secure the performance gain of NOMA, we jointly design the power allocation for pilot and payload to enhance NOMA's robustness against channel uncertainty. Third, we study the robust power-efficient resource allocation design for downlink multi-carrier NOMA (MC-NOMA) systems, with taking into account the imperfect channel state information and heterogeneous traffic demands. A globally optimal design is proposed as the performance benchmark and a low-complexity suboptimal iterative algorithm is developed which can converge to a close-to-optimal solution rapidly. Fourth, we propose a multi-beam NOMA framework for hybrid millimeter wave (mmWave) systems and investigate its resource allocation design. Compared to the conventional single-beam mmWave-NOMA scheme, our scheme can flexibly pair NOMA users with an arbitrary angle-of-departure distribution. A two-stage resource allocation design is proposed to maximize the system sum-rate. Fifth, we propose a beamwidth control-based mmWave-NOMA scheme and study its energy-efficient resource allocation design. Enjoying the widened analog beamwidth, our proposed scheme can increase the number of potential NOMA groups, which improves the system energy efficiency. A user grouping algorithm and a low-complexity iterative digital precoder design algorithm are developed for maximizing the system energy efficiency.