QoS Oriented Network Integration Scheme For Heterogeneous Vehicular Network

Based on intra-vehicle network, inter-vehicle network and vehicular mobile internet, IoV(Internet of Vehicle) has been developed into a large system through wireless communication and information exchange. The goal of IoV is to realize the integration of “Human-Vehicle-Road-Internet” which can satisfy vehicle users’ personalized information demands. However, a variety of wireless networks with different network performances composed heterogeneous wireless networks in the complex urban environment, such as cellular networks, WiFi, VANET(Vehicular Ad hoc Network). How to realize vehicular network integration to ensure vehicle users’ quality of service(QoS)by selecting reasonable network is one of the most challenging and important issues in the research of IoV.Face to the former question, this thesis studied the network integration scheme with two scenarios, 1) urban viaduct environment and 2) urban tra?c environment,respectively, based on the analysis of the main characteristics of IoV. The following contributions are made by this thesis.1) For the urban viaduct scenario, we divided the road into several segments atfirst, then the mobility of vehicles can be represented by the segments transitions using a Markov chain model. Combined with channel model and transmission mechanism, we analysed the collision probability and date rate when the access subnetwork is transferring data. In the second, a performance guaranteed optimized network integration algorithm is provided. With this algorithm, the communication of vehicles can not only to reach overall load balance among all access points with the minimization of cost, but also to maximize the data rate of the whole networks as well as the vehicle users’ demands. At last, simulations are conducted to demonstrate the performance of the proposed network integration algorithm.2) For the urban tra?c scenario, we proposed a network integration architecture of vehicular heterogeneous networks at first. The distribution model of vehicles is established through the analysis of real tra?c data. Then, we developed an intelligent network access system(INAS) using control theory and fuzzy model to provide seamless vehicular communications. Specially, the INAS network integration mechanism comprises two components,(a) access recommender console and(b) automatic access engine. The first component recommends vehicle an appropriate network to access by employing an analytic framework which takes tra?c status, user preferences, service applications and network conditions into account. In the second one, an automatic access engine is developed, which enables individual vehicle to make access decision using a trained mapping relationship. This mapping is conducted by fuzzy model which trains the information of access recommender, local observation and QoS demand. Lastly, the simulation results show that INAS mechanism can effectively select the optimum network to ensure the QoS of vehicle.In summary, this thesis designed network selection schemes in connection with the main characteristics of different scenarios. Through these schemes, vehicular heterogeneous networks can be merged to satisfy vehicle users’ QoS demands. Researches of this thesis are helpful for IoV technologies’ development and applications.