A Centralized and Scalable Uplink Power Control Algorithm in Low SINR: A Case Study for UAV Communications

Interference management through power control is essential to optimize the system capacity. With the introduction of aerial user equipments in cellular networks, resulting in an increase of line of sight links, power control is becoming more and more vital to enable the (uplink) high-throughput data streaming and protect the users on the ground. The investigation in [1] shows that in the high signal-to-interference-plus-noise (SINR) regime, geometrical programming (GP) can be used to efficiently and reliably solve the problem. In the low SINR regime, a series of GPs are solved by condensation. However, the condensation method proposed in [1] is non-scalable, which hinders its application to a large-scale network, e.g. a densified network, where many more cells could be jointly optimized. In this communication, by transforming the original problem into a standard form introducing auxiliary variables, a new condensation method is proposed. Its complexity linearly increases with the number of links increasing, which makes the power control practically solvable for both small- and large-scale networks. A case study for the up-link UAV communications in cellular networks is performed using the proposed algorithm.