Research On Wireless Sensor Network Communication Based On Unmanned Aerial Vehicle

With the Internet of things,Wireless Sensor Network(WSN),as an important means of big data acquisition,has become an important part of the Internet of things.However,In complex scenarios,due to the obstacles between nodes,the positioning and communication of sensor network nodes are severely restricted.With the development of small unmanned aerial vehicles(UAVs),the research on the combination of UAVs and WSN is becoming more and more popular.Especially in the field of wireless sensor network emergency communication,uavs can play a significant role.This paper focuses on the UAV-assisted wireless sensor network communication and works from the following aspects:(1)in the complex 3D scenarios,the positioning of UAV-assisted wireless sensor network nodes is studied,which is mainly divided into two parts:RSSI based multilateral positioning and UAV dynamic path planning algorithm of RT-TF(real-time Topology-Free).With Linear,Spiral and LMAT comparison of three path algorithm,The experiment proves that the UAV auxiliary node positioning solution can overcome the shortcomings of traditional positioning in complex scenarios,and RT-TF algorithm proposed in this paper compared with the Linear,Spiral,LMAT path planning algorithm to locate all the blind node,the total average number was reduced by 62.22%,27.88%and 46.46%respectively,algorithm robustness has reached nearly 90%.(1)In the 3D complex Scene,It is mainly divided into two parts:RSSI based multilateral positioning and uav path planning.RT-TF(Real-Time Topology-Free)UAV dynamic path planning algorithm has been proposed,and compare with three path algorithm of Linear,Spiral and LMAT.The experiment proves that the UAV auxiliary node positioning solution can overcome the shortcomings of traditional positioning in complex scenarios.Moreover,the RT-TF algorithm proposed in this paper is 62.22%,27.88%and 46.46%lower than Linear,Spiral and LMAT path planning algorithms in locating all blind nodes,respectively.,and the algorithm robustness reaches nearly 90%.(2)The research on the establishment of self-organized network of wireless sensor nodes assisted by UAV.Improvement of k-means clustering algorithm based on regional segmentation mode selection,residual energy and average distance between nodes,the UAV can calculate clustering,broadcast and guide the ground node to quickly establish the self-organized network.In the communication stage,the UAV communicates with the cluster head to collect cluster information and balance network load through the cluster head rotation strategy.The experimental simulation results show that with the assistance of UAV,the KM-CF(k-means Cluster Formation)algorithm proposed in this paper is superior to the traditional clustering routing protocol and can reduce the overall energy consumption of WSN and prolong the life cycle of nearly 200 rounds.(3)A wireless sensor network visualization analysis system based on UAV is constructed by Nodejs and React technology.It is mainly divided into three parts:data parsing,data analysis and decision,and data visualization.Data parsing includes the data packets sent to the server by UAV through LoRa remote transmission module,after stored in Mongodb database.Meanwhile,WebSocket technology is adopted to realize real-time update of view.The data analysis and decision part mainly calls the interface of KM-CF algorithm to clustering,establishing an self-organized network,and graphicalize the cluster information with statistical analysis and Canvas technology.The data visualization part mainly realizes the visualization of node location,WSN topology and UAV flight path through the study of baidu map API.