Research On Optimization Method Of Roadside Unit Deployment In Internet Of Vehicles Based On Evolutionary Computation

With the rapid development of science and technology and the continuous progress of society,the number of automobiles is increasing explosively,which leads to many problems in the traditional transportation infrastructure.Therefore,people put forward the internet of vehicles,using vehicle-to-vehicle,vehicle-to-road and other information exchange methods to reduce the pressure of traditional transport facilities.Road-side unit(RSU)is one of the core parts of communication mode in internet of vehicles.As a bridge connecting vehicles and external networks,its importance is self-evident.However,the cost of RSU deployment is high,so it is very important to design a reasonable RSU deployment scheme to give full play to its unit efficiency in the internet of vehicles.The main contents of this paper are as follows:(1)Firstly,the communication mode and channel model in the internet of vehicles are studied,and the RSU deployment problems in the two application scenarios of highway and urban road network are analyzed respectively.Specifically,single-objective optimization of RSU deployment in highway scenario and multi-objective optimization of RSU deployment in urban road network scenario are proposed.(2)Secondly,an improved discrete CS algorithm is proposed to solve the single-objective optimization problem of RSU deployment constructed in highway scenarios.In order to solve the problem of RSU deployment optimization in discrete highway scenarios,this paper studies Cuckoo Search(CS)algorithm in swarm intelligence optimization algorithm,analyses its basic principle and main work flow,proposes a discretization method for each part of CS algorithm,and obtains a discrete version of CS algorithm,which is suitable for solving the problem proposed in this paper.In addition,in order to improve the performance of the discrete CS algorithm,an improved discrete cuckoo search(IDCS)algorithm is proposed.In IDCS algorithm,the population is divided into two parts.Different step sizes are used to update the location,which improves the local search ability of cuckoo algorithm.In order to test the effectiveness of the improved scheme,CEC2014 test set function is used to test the performance of IDCS algorithm,and compared with other swarm optimization algorithms in solving accuracy and convergence speed.The comparison results show that IDCS algorithm has better solving ability.(3)Thirdly,an improved non-dominated sorting genetic algorithm Ⅱ(INSGA-Ⅱ)is proposed to solve the multi-objective optimization problem of RSU deployment in urban road network scenarios.The main operation and working process of the traditional non-dominated sorting genetic algorithms Ⅱ(NSGA-Ⅱ)are analyzed.INSGA-Ⅱ discretizes NSGA-Ⅱ to make it suitable for solving the optimization problem proposed in this paper.At the same time,a multi-point crossover strategy is proposed for the crossover operation of the algorithm to increase the diversity of the population.In order to improve the utilization of the population,a de-duplication operation of the solution set is introduced.In addition,the performance of INSGA-Ⅱ is tested by using ZDT test function,and the effectiveness of INSGA-Ⅱ is proved by comparing with the experimental results of other multi-objective optimization algorithms.(4)Finally,the simulation results of two optimization strategies in different application scenarios are given.Firstly,the parameters of the two algorithms are optimized to make them have better performance on the corresponding optimization problems.Then,the effectiveness of their improvement factors is verified separately.Next,the optimized algorithm is used to solve the corresponding optimization problems.Through the analysis of the experimental results,we can see that IDCS algorithm and INSGA-Ⅱ algorithm have obtained the best optimization results respectively.Finally,the stability of the two algorithms is analyzed.