Full text
| |
| Author(s): |
Jorge Lima de Oliveira Filho
Total Authors: 1
|
| Document type: | Doctoral Thesis |
| Press: | Campinas, SP. |
| Institution: | Universidade Estadual de Campinas (UNICAMP). Instituto de Computação |
| Defense date: | 2013-06-21 |
| Examining board members: |
Edmundo Roberto Mauro Madeira;
Jose Ferreira de Rezende;
Michele Nogueira Lima;
Islene Calciolari Garcia;
Nelson Luis Saldanha da Fonseca
|
| Advisor: | Edmundo Roberto Mauro Madeira |
| Abstract | |
The mobile nodes (MNs) have become popular nowadays, the rapid increase in the use of wireless technologies has changed the behavior of individuals and businesses in the way they communicate. Currently, there is a demand for people to be connected all time and everywhere by using their MNs to perform their tasks. It is desirable to have a ubiquitous Internet, available at any moment and everywhere. However, the integration of heterogeneous wireless networks introduces several challenges. One of the most challenging issues is service continuity during handovers. As a result of the high demand for mobility, IEEE has created a set of services and protocols in order to integrate heterogeneous networks, called IEEE 802.21. However, there are no mechanisms in the standard in order to support mobility in heterogeneous networks, such as decision algorithms and mechanisms to perform the handover. Moreover, the mobility demand requires a special management mechanism due to the elements' diversity and the dynamic environment of the heterogeneous wireless networks. In order to tackle these challenges, this work presents cognitive mobility management architecture for heterogeneous wireless networks. This architecture is based on the IEEE 802.21 and is able to integrate different networks with different technologies and supporting mobility while maintaining service continuity when the MN performs a handover. A handover decision algorithm was developed for this architecture to select networks based on the previous experiences of MNs. This algorithm uses as decision metric the TCP loss in the MNs. The cognitive module in the architecture is an evolution of research results of our proposed algorithm for reconfiguration of cognitive parameters of the IEEE 802.11, called CogMAC. The CogMAC is a completely distributed proposal capable to monitor the MAC layer performance and react to changes to avoid performance degradation. The CogMAC is evaluated by simulations and real testbed and the results are compared with the first-WiFi algorithm. This algorithm prioritizes the connection of WiFi networks. The results also confirm the good performance of the proposed architecture (AU) | |