K-Means Clustering-Aided Power Control for UAV-Enabled OFDM Networks

In this paper, we consider an unmanned aerial vehicle (UAV)-enabled orthogonal frequency division multiplexing (OFDM) network where the UAV acts as an aerial base station and employs the OFDM to transmit individual data to multiple users on the ground. We propose a
$K$
-means clustering-aided power control algorithm to maximize the sum-rate of the network under the total transmit power and the minimum rate constraints. The proposed algorithm categorizes
$N_{\mathrm {s}}$
subcarriers into
$K$
clusters and optimizes the transmit power for the clustered subcarriers with a power splitting factor to decompose the multi-user power allocation problem. Since the number of clusters is far less than the number of subcarriers, the proposed algorithm requires low-computational complexity compared to the conventional water-filling algorithm. Simulation results show that the proposed algorithm provides near-optimal performance with
$K=4$
clusters even for a large number of subcarriers
$N_{\mathrm {s}}=1024$
.