Font Size: a A A

Heterogeneous Vehicular Network Model And Handoff Method With Varying Network Performance

Posted on:2020-04-06Degree:DoctorType:Dissertation
Country:ChinaCandidate:X C LiFull Text:PDF
GTID:1362330626456750Subject:Traffic Information Engineering & Control
Abstract/Summary:PDF Full Text Request
The heterogeneous vehicular network is one of the key technologies and hotspots in intelligent transportation system research.Previous heterogeneous vehicular network studies merge wireless networks with fixed performance together through vertical handoff to provide "always the best connection" for collaborative awareness applications.When this kind of system is applied to real scenarios with varying network performance,there will be a ping-pong effect resulting in a decrease of network performance.The systems are unable to provide a stable and reliable service for collaborative awareness applications.For the above problems,a heterogeneous vehicular network model and handoff method with varying network performance are proposed in this paper on the basis of reviewing the domestic and foreign research.The characteristics of wireless network technology,network utility evaluation,network selection based on game theory and heterogeneous network testing technology are also carried out in the work.Specific contents are as follows:1.A model of heterogeneous vehicular network system with varying network performance is proposed.Firstly,the instability of the previous heterogeneous vehicular network system utilized with varying network performance is analyzed and the necessity to propose a varying network performance structure is verified.The previous heterogeneous vehicular network system model cannot provide network service with transparent transmission,practicability to transportation communication infrastructure and no extra loads at the same time.A structure of heterogeneous vehicular network system is proposed applying to varying network performance and solving the above issues.Then the heterogeneous vehicular network system is formulated referring to the static traffic assignment model.This model describes the varying network performance and defines the optimization objectives of User Equilibrium(UE),and System Optimum(SO).The model will guide the selection of heterogeneous vehicular network and its strategy effectively.2.A heterogeneous vehicular network handoff method based on the evolutionary game and two-layer game is proposed.The model of heterogeneous vehicular network system with varying network performance is used as the constraints,and the UE and SO are used as the handoff targets,respectively.The evolutionary game and the two-layer game are used to realize the heterogeneous vehicular network handoff.In the evolutionary game-based handoff method,the network handoff process is described as a repeated game utilizing the bounded rationality of players,which drives the system to achieve stable UE with the terminals’ repeated perception-decision process.Referring to the network resource allocation method adopted by Long-Term Evolution-Vehicle(LTE-V),the terminal is deployed to switch in and switch out of Dedicated Short-Range Communication(DSRC)to realize SO based on the two-layer game.The above methods are compared with the classical game-based network handoff method.The simulation results show that the handoff method based on the evolutionary game and two-layer game can realize the user equilibrium and system optimum in a heterogeneous vehicular network system with varying network performance and overcome the ping-pong effect of the classical game-based method.However,in the process of network handoff,several parameters have to be adjusted for different network utility evaluations to keep the system stable while facilitating rapid convergence of the system since the handoff probability is related to the network utility evaluation.3.A heterogeneous vehicular network handoff method based on the pre-set terminal number is proposed.According to the simulation analysis of the above two-layer game-based method,as well as the characteristics of DSRC network performance and relatively stable network load,the heterogeneous vehicular network handoff method based on the pre-set terminal number was established with the ideal number of DSRC terminals as the optimization objective in the system.This method can effectively maintain the number of DSRC terminals to an ideal state by taking advantage of the bounded rationality of players.This method and the proposed method based on two-layer game are simulated and compared.Simulation results show that the method based on the pre-set terminal number can realize the approximate performance with the two-layer game without using probability calculated by network utility evaluation,at the same time improve the stability and average performance in the randomly varying network performance condition.4.A prototype for heterogeneous vehicular network based on DSRC,Long-Term Evolution(LTE)and Wi-Fi was established.A heterogeneous vehicular network terminal based on a soft handoff of multi-network adapter,which can simulate the scene of 50 terminals broadcasting basic safety messages with 5 sets of equipment,realizes the test of the heterogeneous vehicular network.On this platform,the network handoff method based on the pre-set terminal number proposed in this paper and the previous network handoff method based on the classical game are compared and tested.The test results show that the proposed network handoff method based on the pre-set terminal number is significantly better than the classical game-based network handoff method on both network utility evaluation and system stability.The proposed method adapts to the varying network performance in real-world scenarios.The verification of the proposed method can promote the transformation of heterogeneous vehicular network technology from scientific research to practical application,and promote the development and application of connected and intelligent transportation-related technologies.
Keywords/Search Tags:Vehicular Network, Heterogeneous Vehicular Network, User Equilibrium, System Optimum, Evolutionary Game, Bounded Rationality
PDF Full Text Request
Related items