Research On Intelligent Nodes Deployment Planning Algorithms For Internet Of Things Applications

With the deep integration of the new generation of information technology and urban modernization,various intelligent Internet of things(Io T)applications are emerging,and there are more and more types of services.With the explosive growth of mobile data traffic and the number of connected devices,wireless network becomes more and more complex,which requires the network not only be able to meet various service requirements,but also reduce network costs.Therefore,how to implement intelligent and efficient green deployment planning for smart nodes such as base stations and gateways with large coverage for various Io T applications is one of the key issues.To solve the above problems,this paper first introduces the base stations deployment planning and Io T gateways deployment planning methods in different scenarios,and then mainly researches from the following three aspects:(1)A three-dimensional dense network planning method based on adaptive variable-length particle swarm optimization is proposed for the three-dimensional dense network planning problem.Firstly,the network dimension is estimated according to the characteristics of signal propagation and hybrid distribution of services in the indoor environment.Then,the three-dimensional dense network planning model is established.Under the condition of satisfying the constraints of coverage and capacity,the adaptive variable-length particle swarm optimization algorithm is used to obtain the optimal deployment scheme of small base stations.The simulation results show that this method can effectively improve the service ratio and coverage,save the deployment cost and improve the deployment efficiency,which has a certain reference value for dense network planning.(2)A deployment method of edge gateways based on simulated annealing is proposed for the edge gateways deployment problem in the Io T.Firstly,this paper analyzes the conditions of covering and serving terminal traffic generators,the factors affecting the offload delay of the computing tasks,and the constraints to be satisfied by the capacity allocation of edge gateways.Then,an optimization model of edge gateways deployment is established.Finally,the simulated annealing algorithm is improved to obtain the optimal deployment scheme.The simulation results show that this method can minimize the deployment cost,improve the resource utilization of the edge networks,realize the load balance of the edge gateways,and provide some guidance for the placement of the edge gateways.(3)A big data visualization platform for base stations deployment based on data mining is designed and implemented for the base stationsdeployment problem in Low Power Wide Area Network(LPWAN).The core of the platform is a data-driven base stations deployment framework.Firstly,the received signal values of all terminal test points are predicted by the received signal prediction module based on Light GBM.Then,the prediction results are transformed into the weight values of all terminal test points during clustering,and the terminal test points are clustered by the base station location optimization module based on weighted K-means.Finally,the simulation of data-driven base stations deployment is established on the big data platform.The results show that the data-driven base stations deployment framework achieves efficient base stations deployment and improves the coverage performance of LPWAN.The big data visualization platform demonstrates the whole process of base stations deployment,which has a certain reference value for base stations deployment in LPWAN.