An Intelligent UAV based Data Aggregation Algorithm for 5G-enabled Internet of Things

Unmanned Aerial Vehicle (UAV) has become a significant part of 5G or beyond 5G (B5G) paradigm, and is used in various scenarios, including cargo delivery, agricultural application, event surveillance, etc. Although plenty of studies have been proposed on UAV-based data aggregation, how to ensure security and energy-efficiency of the data aggregation process in 5G-enabled Internet of Things (IoT) is an open problem. In this paper, we propose an Intelligent UAV-based Data Aggregation Algorithm, named IDAA for 5G-Enabled IoT. Specifically, IDAA applies v-Support Vector Regression (v-svr) to predict the data collection rate. Then, a security level based task decomposition mechanism is designed that allows UAVs to accept the tasks of corresponding security levels. Finally, energy efficient routes for UAV are planned utilizing a deep reinforcement learning method to achieve the trade-off between the sinking ratio and the energy cost. The theoretical analysis and simulation results indicate that (i) IDAA improves the security of data aggregation; and (ii) IDAA enables UAVs to collect more data and consume less energy compared with baseline strategies.