Optimization Schemes for UAV Data Collection with LoRa 2.4 GHz Technology in Remote Areas without Infrastructure

Abstract

Recently, the use of unmanned aerial vehicles (UAVs) and LPWANs (low-power wide-area networks) has been a good solution to the problem of data collection for environmental monitoring in remote areas without infrastructure, and there are many valuable research works in this field. UAV data collection for sensor nodes is becoming a challenge, that is, the amount of data will affect the UAV’s communication time and flight status, especially in LPWAN systems. In this paper, the optimization schemes are proposed to improve the efficiency of UAV for collecting data in LoRa network monitoring systems. Firstly, an improved clustering algorithm for the LoRa network is proposed, which considers the influence of distance between the cluster heads and the UAV take-off point. Secondly, we present an improved Genetic Algorithm for path planning to reduce the UAV flight distance, which introduces the Teaching–Learning-based Optimization (TLBO) and local search optimization algorithms to improve convergence speed and the path solution. Then, a LoRa 2.4 GHz adaptive data rate strategy with a dual channel is designed based on distance and link quality, to reduce the data transmitting time between the UAV and the cluster head nodes. Finally, we carry out the simulations and experiments. The results show the performance of the proposed schemes, which means that these can improve the efficiency of UAV data collection with low cost LoRa networks in remote areas without infrastructure.