Research On Collaborative Deployment Of Urban Environmental Emergency IoT Based On Unmanned System

In the city’s emergency relief work,we need to deploy the emergency networks to restore the city’s communication system so that the rescue work can be carried out quickly.The traditional way to deploy base stations is time-consuming and costly.With the development of unmanned system technology,the emergency Internet of Things based on unmanned system has become an important choice.Different urban scenes with different characteristics have different network deployment requirements.Currently,there are so many cities,and the methods to classify cities as objects are more and more extensive,but the standards for city classification are mainly formulated for various industries and vary from person to person.Therefore,these methods cannot be applied to the collaborative deployment of emergency networks of unmanned systems under urban scenes.Therefore,thesis first proposes a city classification standard based on building density and floor area ratio,and optimizes the SVM prediction model.Secondly,this thesis proposes an adaptive coverage optimization algorithm for UAV network in urban disaster areas,which is applied to the optimization of UAV emergency network deployment in urban scenarios.The main work and innovation of this thesis are as follows:(1)Considering the different collaborative deployment methods of unmanned systems in different urban disaster areas and the influence of building density and floor area ratio on the A2G channel model,a new city classification standard is proposed.The city sample data were labeled by this classification standard.Based on this standard,aiming at the problem of low classification accuracy of SVM classification model,the classification accuracy was taken as the objective function,and the sparrow search optimization algorithm was used to optimize two important parameters of SVM classification model,and the classification model S-SVM was obtained.Compared with the SVM classification model and the SVM classification model optimized based on gray wolf algorithm(G-SVM),the accuracy of S-SVM classification model is effectively improved compared with the SVM model.Compared with the G-SVM classification model,the accuracy is higher,the time is shorter,and the classification effect is stable.(2)Aiming at the problem of emergency network coverage optimization in urban disaster areas,an adaptive coverage optimization algorithm of UAV network in urban disaster areas was proposed.Firstly,key areas are divided from urban disaster areas,and the movement of ground user nodes in urban disaster areas is simulated and applied to the improved cuckoo algorithm.The coverage rate of key areas is taken as the objective function,and finally the coverage rate of mobile user nodes in key areas of urban disaster areas is improved.The simulation results show that the proposed algorithm under the same experimental environment standard,cuckoo algorithm(CSA)and simulated annealing algorithm(SAA),compared to the coverage of all key areas has been effectively improved,and the network coverage,connectivity and the path loss of unmanned aerial vehicle(UAV)network are stable.As the simulation time change,several evaluation indexes of emergency network are also relatively stable.