Connection probability in reference point group mobility model
In ad hoc wireless mobile networks, the mobility of wireless nodes plays a significant role in evaluating the network’s performance. As the nodes are mobile there is a frequent transition between the connections of nodes from up to down state. It is essential to know the likeliness of having a connection between different nodes and the parameters on which the connection probability depends in order to have a better performance of the mobile networks. This thesis presents such an approach to evaluate the likeliness of having connections amongst the nodes in ad hoc mobile network. In this study a mathematical model is proposed for evaluating one hop connection probability between nodes of different groups. The mathematical model is devised for a specific group mobility model, known as Reference Point Group Mobility model (RPGM) which can be used to simulate the networks where node co-ordinate with each other. The mathematical model is based on the results of ongoing research in the field of mobile ad hoc networks. It provides a relationship between different network parameters which sufficiently determine the connection probability in reference point group mobility model. This work demonstrates the impact of different parameters on connection probability. Furthermore, the graphical results are used to illustrate the behavior of connection probability in RPGM. It is found that connection probability of directly connected nodes of two different groups’ increases with decrease in the difference between mean locations of the two groups. Second observation is that as the mobile nodes move with higher velocity, the nodes break their intergroup connections quickly which result in lower connection probability and vice versa. The study also suggests that as range of nodes increases, the connection probability increases. The mathematical model developed in this thesis work can be used to determine the values of various network parameters for required connection probability in a mobile wireless network. The practical use of this thesis work can be found in determining the connection probability in real world scenarios like disaster recovery where two or more groups need to interact with each other. Another useful application is in military scenario where member are divided in different groups and they interact with each other. In such scenarios the devised model can be used to set values of different network parameters such as velocity of mobile nodes, transmission range etc. to have required connection probability in the network.
Thesis (M.S.)--Wichita State University, College of Engineering, Dept. of Electrical and Computer Engineering.