Research On The Method Of Vehicle Network Roadside Unit Deployment Based On Subarea

In the Vehicular Ad Hoc Networks,vehicles cannot access the network smoothly because of their high speed movement.The roadside units are deployed beside the road as auxiliary communication infrastructures,which is more stable than the on-board unit.Vehicles can transmit the information to the suitable roadside unit,and then forward it to the corresponding vehicle at the right time,which makes the vehicles communication more enabling.Due to the high cost of deploying the roadside units and its limited coverage,it is significant to select the locations of the roadside units reasonably,which improve the service quality of the Vehicular Ad Hoc Networks.To solve the problem that submitted above,this paper proposed the strategy of deployment subarea partition firstly,and then explored the connectivity probability between vehicles in platoon and roadside unit.Finally,a multi-group genetic algorithm is designed to maximize the role of roadside unit in the road.In terms of roadside unit deployment,the strategy of traffic subarea partition based deployment is introduced.The clustering indices were selected according to their function of the roadside unit.We used Euclidean distance to calculate the similarity matrix of each subarea,and the adjacent relationship of the subarea in geographical location was taken into account.In addition,the F-test is adopted to determine the optimal partition results,which acts as a prerequisite for the selection of subsequent deployment positions.With regard to the co-relation between the roadside units and the connectivity of the vehicles,this section will,with the platoon considered as a standard,concentrate on the moving pattern of the vehicles in the platoon.Thereafter,the calculation will transform from the connectivity of the platoon and roadside unit to the length of the arrived platoon.The renewal theory is introduced to describe the distribution of the headways of vehicles in platoon,which consists with the log-normal distribution.On account of that,the probability density function of the platoon can be deduced in accordance to the convolution formula,and then the complex integral model is solved based on numerical method.Upon the mentioned subareas,the multi-group genetic algorithm has been designed to deploy the roadside units and sort out the deployment candidate according to the above connectivity.The connectivity and the degree of vertex are required to act as the fitness function of the algorithm in the designation of genetic elements.Besides,with the application of the adaptive genetic operator,the global search ability of the algorithm can be improved.To sum up,this paper first divides the target road network area into subareas,combines them with more consistent features through similarity calculation,which acts as a prerequisite for the selection of subsequent deployment positions.Then,the vehicle platoon is used to study the connectivity probability of vehicle and roadside units.Through a comparison between the simulation results and the model,the accuracy of the model has been proven,which can be used to calculate the probability of the vehicle’s communication with the roadside unit,and provide reference for the deployment interval of the roadside unit.Finally,this paper uses Multi-group genetic algorithm to solve the problem of roadside unit deployment,the simulation results show that the algorithm can upgrade the deployment in less iteration times.The deployment based on subareas can disperse the location of roadside units,avoiding the uneven distribution of traffic resource.