Non-orthogonal multiple access (NOMA) and millimeter-wave (mmWave) are two crucial techniques of 5G to meet the explosive capacity demands. On the other hand, UAVs deployed as aerial base stations are potential to provide ubiquitous coverage and satisfy users' multifarious requirements due to their flexibility and mobility. Nevertheless, the finite onboard energy is a fundamental limit of UAVs, which can deter the performance of UAV communication networks. Therefore, the researchers Xiaowei PANG and Nan ZHAO from Dalian University of Technology, Jie TANG and Xiuyin ZHANG from South China University of Technology, and Yi QIAN from University of Nebraska-Lincoln have focused on designing energy-efficient transmission schemes for mmWave-enabled NOMA-UAV networks. The network model is illustrated in Fig. 1, where the UAV equipped with multiple antennas serves K single-antenna ground users who are grouped into M clusters in the downlink.
"Although a large amount of research has contributed to integrating mmWave or NOMA with UAV communications, respectively," the five researchers wrote, "few of them investigated on mmWave-enabled NOMA-UAV networks."
In mmWave networks, the number of supported users conventionally cannot be larger than the number of RF chains at the same time-frequency resources. To break this fundamental limit, users are grouped into multiple clusters according to their channel correlations, and NOMA is employed in each cluster to serve the users simultaneously. To achieve a good balance between system complexity and performance, a hybrid precoding architecture is adopted to reduce the hardware cost and energy consumption. The authors aim to maximize the energy efficiency of mmWave-enabled NOMA-UAV networks by optimizing the UAV placement, hybrid precoding and power allocation. Due to the fact that the overall energy efficiency maximization problem is intractable, it is divided into several sub-problems. First, they optimize the UAV placement considering the total channel strength of all UAV-served users. And then, the hybrid precoding schemes with user clustering are proposed to better reap the multi-antenna gain. The last step is to optimize the power allocation among users to maximize the energy efficiency with users' quality of service requirements and an efficient algorithm is presented to solve the problem iteratively.
It's worth mentioning that three hybrid precoding schemes are introduced in the network, all of which perform user clustering and design the analog and digital precoding to improve the multiplexing gains and suppress the inter-user interference. Particularly, the major distinction among them is that they can achieve different performance of user fairness, spectrum efficiency and energy efficiency, which are further demonstrated by simulation results. The effectiveness of the proposed energy-efficient design is verified through numerical comparisons with other schemes without UAV placement optimization and without energy efficiency requirement. Moreover, numerical results also reveal the effects of the maximum UAV transmit power and the number of RF chains on the energy efficiency.
For more details, please refer to the upcoming paper "Energy-efficient design for mmWave -enabled NOMA-UAV networks", to be published in SCIENCE CHINA Information Sciences, 2021, 64(4): 140303.
Science China Information Sciences