Treffer: User pairing and location planning in multi-UAV assisted downlink NOMA for emergency communications.

The heterogeneous Internet of Things (IoT) network has received significant attention in 5G and Beyond 5G systems. However, IoT is more susceptible to disasters since the network is mains powered and the devices run with high computational power. Meanwhile, the network needs to meet technical challenges such as spectrum scarcity, massive connections and high power consumption. To address such issues, a more systematic tool is employed, utilizing unmanned aerial vehicle (UAV) and non-orthogonal multiple access (NOMA) scheme. In this article, we establish a multi-UAV assisted downlink NOMA framework to support emergency communications required in a disaster area. The proposed system optimizes UAV placement and dynamic user clustering to maximize energy efficiency (EE) while mitigating interference. A novel scheme is proposed for user clustering based on the geometric distribution of users to reduce inter-UAV interference. An optimization problem that maximizes the sum rate has been formulated for the location planning of UAVs. This developed problem is solved by introducing Lagrangian multipliers which are updated using the gradient-descent method. Meanwhile, Karush–Kuhn–Tucker conditions are utilized to deal with inequality constraints. In an effort to propose a unified solution for the problem of user clustering and UAV location planning, we develop an algorithm using the block coordinate descent method. Extensive simulations are conducted to illustrate the superiority of the proposed unified solution. EE performance of different conventional benchmarks such as orthogonal multiple access scheme is considered for comparison. The optimal number of clusters and optimal number of users per cluster are also computed to frame the unified solution more efficiently. [ABSTRACT FROM AUTHOR]