An Adaptive Finite Resolution Selection Scheme for RIS Assisted NOMA Networks

Non-orthogonal multiple access (NOMA) and reconfigurable intelligent surface (RIS) have recently been recognized as innovative technologies for future beyond fifth generation (B5G) wireless networks due to their ability to improve the spectral efficiency and the wireless propagation properties, respectively. This paper presents an integration between RIS and NOMA to communicate with distant vehicular users. To alleviate the complexity of coherent phase shifting design at RIS, we propose a dynamic gain adaptation RIS selection scheme based on discrete phase shifting (i.e., termed as GADPS) to serve two vehicular users paired to perform NOMA. Regarding channel state information (CSI), the adaptation mechanism determines which user can be considered as a near user and the other as a far user with respect to RIS. To characterize the benefits of the GADPS scheme, the system outage performance is derived and compared with the cases of random discrete phase shifting (i.e., termed as GARPS) and conventional RIS-NOMA schemes; then the diversity order is also obtained. Furthermore, the impact of network parameters on the throughput and asymptotic average sum rate is discussed in both delay-limited and delay-tolerant transmission modes, respectively. Monte Carlo simulations emphasize the accuracy of the analytical work and investigate that: 1) The proposed GADPS scheme is more flexible than conventional RIS-NOMA scheme; 2) By joint optimization of RIS selection and phase shifting, the proposed GADPS scheme is capable of achieving superior performance over GARPS and conventional RIS-NOMA schemes with a superior diversity gain; 3) The throughput performance of the proposed GADPS scheme outperforms GARPS, conventional RIS-NOMA and other relay-assisted NOMA schemes, whereas full-duplex (FD) relay selection NOMA schemes suffer from throughput floor values; 4) By increasing the number of reflecting elements of RIS, the proposed GADPS scheme achieves the enhanced average sum rate performance over benchmark schemes in all scenarios of relative distance difference between users and the selected RIS.