| With the development of mobile cloud computing, content aware and dedicated short-range communication technology, vehicles network with the ability of decision-making and independent control has been upgraded for the Autonomous Vehicular Cloud (AVC). The AVC can provide resource centralized service dynamically, so that authorized users can enjoy the service sharing and management in real time. When the users in AVC have a task request, how to allocate tasks efficiently and how to receive the results of the task accurately are still problems in AVC. The biggest obstacle of task allocation in AVC is the mobility of vehicle node and the dynamicity of network, which will lead that the task executing node may leave the network at any time, especially in highway scenarios, this situation will become more frequent. Therefore, it is necessary to monitor the task executing node continuously and reallocate the task while the node leaves the network. In order to solve the above problem, the checkpoint server mechanism is introduced, but the checkpoint server selection scheme based on neighbor vehicles node still has its shortage, so this thesis proposes the selection scheme of the checkpoint server based on link stability, in order to improve the efficiency of task allocation and the result receiving in highway, and also comparing the two checkpoint server selection methods-the choice of relative task initiating node stability of the vehicle node and the choice of relative task executing node stability of the vehicle node.This thesis uses OPNET network simulation tool to evaluate the two schemes mentioned above in different scenarios with different node densities. Finally, the empirical conclusion is that the selected checkpoint based on link stability methods is more stable, can improve the efficiency of task allocation and the result receiving in highway, and vehicle node is more sparse, the enhanced effect is more obvious. |
PDF Full Text Request