Capacity planning of inhomogenous radio mesh networks

OData support
Supervisor:
Dr. Fazekas Péter
Department of Networked Systems and Services

The mesh network system provides services for users every time and everywhere, and it gives opportunity for high quality and fast wireless communication. However, solutions concerning its creation and installation are not fully developed. Therefore, my aim is to get to know and map the system’s features so that I can establish a new mesh node placement algorithm based on which a future realization becomes possible.

In the first phase of my study, I map the opportunities in the structure, a technology that allows communication between the network’s units and a radio channel describing propagation model. The above analyses and their results help to better understand the network’s possibilities and the particular features of each configuration.

Secondly, in order to find the problems and limits that might emerge during the mesh network node installation process, I study the interrelation of the mutual effect of the nodes in a regular grid structure/system. As a result, solutions will be proposed for the possible problems and limits in order to create configurations which allow better exploitation of the network.

As the third step, I introduce parameters to describe network capacity, level of utilization, network system and satisfaction of user needs. With the help of these parameters, I can accurately compare the possible configurations.

Afterwards, I study how much a slight change of the regular grid improves the parameters described above. The grid is changed by using non-equal node distances. The analyses show that a network-wide improvement of the parameters requires appropriate calculation before choosing the distance measures.

Finally, based on the study’s observations and the possible solutions proposed, I create a node placement algorithm which starts out from a regular grid and is altered with certain steps. To implement the algorithm and to show its operation, I design an Android application. This application helps simulating the algorithm’s operation and provides enough data to prove the configurations positive effect on the designed parameters.

Downloads

Please sign in to download the files of this thesis.