Research On Performance Analysis And Deployment For Vehicular Ad Hoc Networks

With the rapid development of automotive industry and transportation systems, the issues of road safety and traffic congestion have gradually drawn wide attention. As a consequence, vehicular ad hoc networks (VANETs) emerge as a new type of network and support the inter-vehicle communications through wireless technologies. VANETs play a key role in improving the safety of road traffic and enhancing the efficiency of transportation systems. In order to ensure reliable data dissemination for safety warnings and traffic information, the system performance of VANETs should be guaranteed and improved. Due to the characteristics of VANETs, such as high mobility of the vehicles, dynamic change of network topology, random positions of the vehicles and time variation of radio environment, the network performance is always seriously affected, and reliable inter-vehicle communications may not be guaranteed all the time. Therefore, it is of importance to investigate the impacts of various factors on the system performance of VANETs.Based on the probability theory and stochastic process, this dissertation comprehen-sively investigates the effects of different factors on the network performance in highway scenarios for VANETs. The key factors that involved in this dissertation include dynamic topology, high mobility, fading channels and user behavior. Analytical results can pro-vide guidance for protocol designing and network deployment of VANETs. The main contributions of this dissertation are listed as follows:1) As for continuous transmission requirement of VANETs, an analytical model is developed to study how continuous connectivity performance would be influenced by the dynamic topology in highway scenarios. Based on the assumption that the vehicles arrive at the highway according to Poisson process, the distributions of inter-vehicle distance and vehicles’relative speed are obtained, thus continuous connectivity probability is further derived. Moreover, the link connectivity perfor-mance can be expanded to the whole network. Then the analytical expressions of network transmission capacity, network connectivity probability and vehicle isola-tion probability can be obtained, respectively.2) Due to high mobility of the vehicles and time-varying radio environment in the highway scenarios, wireless link of inter-vehicle communication would be easily obstructed by other surrounding vehicles. Considering this issue, this dissertation proposes an analytical model to study the effects of both traffic flow and radio en-vironment on the connectivity performance under obstacle-based channel model. Combining measurement-based dual-slope path loss model with channel state tran-sition, the closed-form expressions of inter-vehicle connectivity probability and ve-hicle isolation probability are obtained. Considering the effects of vehicles’relative movement, the statistic characteristics of link duration are also derived.3) The user behavior has a significant impact on the connectivity performance in VANETs. An analytical model based on inter-vehicle cooperation is established in this dissertation to study the tradeoff between the cooperative ratio and connectiv-ity probability under realistic fading channels, with independently identically dis-tributed (i.i.d.) and non-identically distributed (non-i.i.d.) interferences. Tractable results of connectivity probability for both i.i.d. and non-i.i.d. cases are derived, and the benefits of cooperative ratio on inter-vehicle connectivity probability are analyzed. Moreover, the optimal cooperative ratio is obtained and its relation with system parameters is derived under Nakagami and Rayleigh fading channel models, respectively. Analytical results shed light on the effects of cooperative ratio on con-nectivity and provide useful guidelines for designing wireless vehicular networks of high performance.4) In order to enhance the connectivity performance for highway VANETs, this dis-sertation addresses infrastructure-based and relay-vehicle-based vehicular network-s, respectively. The system performance of relay transmission is also analyzed. Based on dual-slope path loss model from realistic measurements and small-scale fading channel, we analyze the enhancements of channel capacity and link connec-tivity probability by the relay transmission in the infrastructure-based networks. According to Poisson Point Process, the global connectivity performance of relay-vehicle-based networks is investigated. Considering the transmission power and interference, the relay vehicle ratio is further studied, and the results can provide useful guidance for parameter configuration in the network design and deployment.