Hybrid Computation Approach for Performance Evaluation of Broadband Wireless Networks based on Tethered High-Altitude Unmanned Platforms

This paper discusses the advantages of implementation of a broadband wireless network based on a tethered drone, and describes a novel hybrid approach based on a combination of analytical modeling, simulation and machine learning to the computation of its performance characteristics. The paper presents the calculation results for the increase in the telecommunications coverage area (line-of-sight zone) and the parameters of the communication channel between the base station (BS) located on the drone and the ground station (GS) within line-of-sight. A stochastic polling model with batch packet servicing is proposed to evaluate the network performance. A description is given of the interaction protocol between the BS and the GS for obtaining initial data for carrying out numerical calculations.