Full-Duplex Relay-Assisted Macrocell with Millimeter Wave Backhauls: Framework and Prospects.
Shuai Han,Yi Zhang,Weixiao Meng,Cheng Li,Zhensheng Zhang
DOI: https://doi.org/10.1109/mnet.2019.1800305
IF: 10.294
2019-01-01
IEEE Network
Abstract:The explosive increase in the number of mobile devices and machines has led to the need for high data rates and massive multiple access, which are to be addressed by upcoming 5G communication systems. Millimeter wave (mmWave) wireless backhaul can be an efficient scheme to connect the small cells to the macrocell to take advantage of its immense bandwidth and high directional antennas. However, the channel characteristics of mmWave communications lead to a limited transmission range. Therefore, relays are necessary for mmWave to extend coverage. In addition, full-duplex, which supports transmitting and receiving in the same band simultaneously, can achieve twice the spectral efficiency of conventional half-duplex in ideal conditions. However, this process will lead to self-interference. In this article, we propose a framework of a multiple-small-cell-based macrocell system with full-duplex millimeter wave backhauls to further improve the spectral efficiency and achieve massive multiple access. Full-duplex mmWave relays are configured as backhauls connecting the macrocell and small cells. The combination of a full-duplex relay and millimeter wave transmission can mitigate the severe path loss issues presented by short wavelengths, yield wide bandwidths, achieve deployment flexibility and improve spectral efficiency. Through precoding design, the self-interference of the backhaul can be significantly suppressed, and the system performance can be enhanced. In addition, within the small cell, a three-step precoding design is proposed for the relay-assisted small-cell system to increase the system throughput and realize greater user access. Sources of complex interference, including inter-relay interference, base station (BS)-to-user equipment (UE) interference, relay station (RS)-to-UE interference and self-interference (SI), are efficiently suppressed. Future research on further enhancing system performance is also discussed.