This final career thesis (Master thesis) is a contribution on the enhancement of
wireless communications, specifically WLAN multi-cell systems based on the
IEEE 802.11 standard. The objectives were to propose and study different
Cross-Layer AP selection mechanisms that include single, dual and multiple
metric based criteria using PHY-MAC interactions. These mechanisms are
designed in order to improve system efficiency through the increase of the
utilization of the available transmission resources. The key idea of these
mechanisms is to make use of certain PHY and MAC parameters, other than
the traditional RSSI measurements, in order to optimize the association to the
best AP, specially focusing on the innovative use of MAC level state metrics. In
this regard, of special interest is the inclusion of MAC level AP traffic load
estimations within these association decisions.
All the proposals are based on the use of a high-performance MAC protocol
called DQCA (Distributed Queueing Collision Avoidance), which is specially
fitted to include the proposed techniques. Computer simulations have been
carried out to evaluate and quantify the benefits of the proposed mechanisms
and techniques in representative scenarios. Moreover, a completely new
handoff procedure has been designed for the DQCA muti-cell operation. This
handoff process allows implementing each of the proposed AP selection
mechanisms.
Furthermore, the interaction between a Cross-Layer scheduling technique at
the MAC level and two proposed AP selection mechanisms has also been
studied. The performance of these techniques has also been assessed by
means of computer simulations.
The analysis of the obtained results show that the proposed mechanisms
perform differently under the considered scenarios. However, the main
conclusion that can be drawn is that AP selection mechanisms that are based
on joint multiple metrics considerations (SNR, AP load, delay, etc.) perform
significantly better than those that use only single or dual metric based
mechanisms.
After the study, we can conclude that the proposed techniques and
mechanisms provide significant efficiency enhancements for DQCA-based
WLAN multi-cell systems so that all of them may be taken into account in future
wireless networks.
