Low-Complexity Joint Power and Spectrum Management for Non-Terrestrial Networks

Non-terrestrial networks (NTNs) play an essential role in the 6G multi-layer architecture to provide ubiquitous coverage as well as guarantee heterogeneous requirements from vertical services. Compared to terrestrial gNodeB, flying base stations (F-BSs) in NTN are limited in terms of computation capability as well as energy budget. Therefore, it is of great importance for F-BSs to have computational and energy-efficient radio resource management (RRM) strategies. In this paper, we propose a low-complexity algorithm that jointly optimizes the transmit power and bandwidth allocation of OFDM-based multiuser downlink in NTN under flat fading scenario. The proposed algorithm exploits the flexible bandwidth design methodology to tackle the binary selection challenges, followed by a bandwidth adjustment step to force the allocated bandwidth to a multiplication of sub-channel bandwidth. More importantly, the proposed algorithm is robust against the channel estimation error. It is shown that the proposed algorithm retains the optimal solutions while significantly reduces the computational complexity, compared to both the optimal brand-and-bound (BnB) and the popular difference-of-convex (DC)-based subchannel allocation solutions.