الفهرس 
Overview of VANETsOnly 14 pages are availabe for public view
 |  | from | 186 |  |  |
| | | from | 186 | | |
Abstract
Vehicular Ad-hoc Networks (VANETs) are a subclass of Mobile Ad-hoc Networks (MANETs) and the general characteristics of VANETs are typically inherited from MANETs. However, VANETs exhibit several unique characteristics such as the large scale of the network, high mobility, and dynamic network topology. Hence, VANETs unique characteristics might cause frequent disconnections of the communication links resulting in an increased overhead of the communication protocols in terms of extra messages and time delay. Therefore, VANETs face diverse considerable challenges that instigate developing efficient communication protocols.
Creating a hierarchical structure by clustering is widely introduced as an emerging research topic in VANETs to make the network topology less dynamic, maintain a level of coordination between neighboring vehicles, and hence improve the networks overall performance. Typically, clustering is achieved by partitioning the network into smaller groups that are called clusters. A cluster is composed of a cluster head (CH) and cluster members (CMs); the CH is chosen among the CMs to manage the cluster and coordinate between the members of the cluster, while the remaining CMs locally communicate with their associated CH.
from another perspective, predicting the future movements of the vehicles is possible due to the restrictions of the road topology, urban layout, and traffic constraints. Hence, the accurate prediction of the vehicles future movements could play a crucial role for both building efficient vehicular communication protocols and enhancing the vehicular transportation systems.
Thus, this thesis is concerned with two major contributions. The first one is to follow the guidelines of systematic literature reviews in order to provide a premier and unbiased survey of the existing mobility prediction-based protocols for VANETs and develop novel taxonomies of those protocols based on their main prediction applications and objectives. Whereas the second contribution of this work is to develop an efficient clustering scheme for VANETs that benefits from the ability to predict the vehicles future movements.
In this thesis, the mobility prediction in VANETs is briefly reviewed with reference to their prediction aims, techniques, use cases, and challenges. Also, each category is discussed with a focus on the prerequisites, advantages, and limitations. In order to figure out the potential usefulness of the mobility prediction in VANETs, usage and performance analysis of the mobility prediction are provided. Additionally, a derivation is provided for the appropriateness of utilizing each prediction aim/technique for the varied use cases.
On the other side, a novel mobility prediction-based efficient clustering scheme (MPECS) is proposed to provide a more stable cluster architecture for VANETs with minimal clustering cost. An analytical analysis is discussed to explore the parameters set that improve the overall performance of MPECS. Also, performance evaluation via simulation is presented to evaluate MPECS compared to four existing clustering schemes for VANETs. The conducted evaluations show a close agreement between simulation and analytical results and demonstrate that MPECS can significantly improve the stability of the clustering architecture with minimal overhead. Finally, our conclusion and the future work are presented.
Keywords:
Mobility prediction, Applications, Objectives, Clustering, Voronoi diagram, VANETs.