Research On The Deployment Of RSU Based On Vehicle Density In The Internet Of Vehicles

VANET has broad application prospects in improving road safety and providing on-board entertainment services.Due to the rapid movement of vehicle nodes in the VANET and the special topology of the road network,the problem of wireless channel instability is generated.To solve this problem,some RSU need to be deployed to provide communication coverage for the vehicle.Because of the high cost of RSU deployment and maintenance,it is important to select the areas with cost-effectiveness and coverage value for RSU deployment.In terms of driving assistance and commercial promotion,the deployment of RSU focuses on covering the area where most vehicles are located,so vehicle density can be used as a factor to measure whether the area is worth covering.How to obtain vehicle density in real time and accurately is also a widely concerned issue in the field of IoV.Most of the existing vehicle density estimation methods need the assistance of infrastructure,these methods have drawbacks in terms of reliability,limitations,real-time and cost.In the vehicle density estimation method without infrastructure,the vehicle distribution under different road conditions is not taken into account,so the algorithm cannot adapt to the changeable traffic environment.In the research on RSU deployment,many methods mainly cover the target area completely,which will cause the redundancy of RSU,because the road is the main target to provide communication coverage.Vehicle density as an important factor in the transportation system should be considered when deploying RSU,but most researchers ignore this factor.To solve these problems,this paper studies vehicle density estimation and RSU deployment.The main research results are as follows:1.Put forward the fuzzy set based on Logistic traffic flow modelAccording to the Logistic speed density model and the road traffic manual,the fuzzy set corresponding to the speed and density of vehicles was proposed according to the fuzzy logic.With reference to the fuzzy set,vehicles on the road can be judged to be sparse or dense according to their own running speed,and then different vehicle density estimation algorithms can be selected adaptively.2.Proposed the infrastructure-free distributed adaptive vehicle density estimation algorithmIn the Distributed Adaptive Vehicle Density Estimation Algorithm(DAVDE),firstly,according to the fuzzy set,traffic conditions are divided into two categories: sparse and dense.Then,the Density Estimation methods based on beacon and clustering are used respectively for different traffic conditions.Simulation results show that compared with existing infrastructure-free density estimation algorithms SLDE and CoTEC,The DAVDE algorithm can better adapt to complex and variable traffic environments,maintain stable low error rate and communication overhead,and the density estimation error rate can be controlled within 4% under different vehicle density environments.3.Proposed the RSU deployment algorithm based on vehicle densityThe RSU Deployment Algorithm based on Vehicle Density(VDRD)uses the cover value model proposed in this paper and the adaptive aggregation point discovery model based on cover value to select the RSU Deployment location according to Vehicle Density.Simulation results show that Compared with the GBP and PMCP algorithm,the vehicle coverage rate is increased by 15% and 2% respectively when 10 RSUs are deployed;when vehicle density exceeded 40pcu/km,the average notice time was reduced by 27.97% and 21.82%,respectively;the number of vehicles’ contact with RSU and the time covered by RSU are superior to the comparison algorithm.