Abstract
M.Ing.
Mobile adhoc networks (MANETs) are multihop wireless topologies that have rapidly changing node structure and limited connectivity. Since MANETs are not deployed on a wide scale, the research community still depends on the simulators such as the network simulator (Ns2) to evaluate MANET protocols.
The topic of how to accurately model an indoor environment in the MANET research community is explored in this dissertation. We take an empirical and simulative approach to model our mobility pattern. Our mobility model is based on activity patterns drawn from the transport science.
A comparison with the random way point is made in order to understand the weighty discrepancy between the two models. Our contribution in this research is three fold:
1. We argue that mobility modelling should be based on activities other than stochastic process that have got no realistic backing;
2. We model our network using by putting up an algorithm and take an empirical approach to model the radio frequency propagation. To show the difference of the two mobility models, the behaviour of the signal strength on the two mobility models is drawn; and
3. Finally an implementation of our mobility pattern and RF measurements in ns2 is done.