“…UAVs horizontal and/or vertical placement, distance, cost, UAV numbers, coverage rate, and users–UAV connectivity are important factors needed to be considered in the deployment problem (Gao et al, 2021; Ghazal, 2021; Lahmeri et al, 2021; Q. Liu et al, 2018; Masroor et al, 2021a; Rahimi et al, 2021; J. Yang, Liang, et al, 2021; C. Zhang et al, 2021). ML algorithms are used to predict the optimal position of the UAVs by identifying the overloaded traffic areas by predicting users' demands and positions (Nouri, Abouei, et al, 2021; Nouri, Fazel, et al, 2021; Oliveira et al, 2021). According to Oliveira et al (2021), among ML approaches tested for predicting users' positions, Gradient Boosting and Random Forest provide the best result, while Lasso, Ridge, and ElasticNet are tied at the last place. - Channel estimation : Channel state information (CSI) highly impacts the performance of the UAV communication systems.
…”