In spite of the significant increase of the use of Wireless Local Area Network (WLAN)
experienced in the last years, design aspects and capacity planning are still systematically
neglected during the network implementation. Typically, a wireless local area network is
designed and installed by networking professionals. These individuals are familiar with wired
networks, but are often unfamiliar with wireless networks. Thus, wireless local area networks
installations are prejudiced by the lack of an accurate performance evaluation model and to
determine the location of the access point (AP), besides important factors of the environment
are not considered in the project. These factors become more important when several APs are
installed, sometimes without a frequency planning, to cover a unique building. Faults such as
these can cause interference among cells generated by each PA. Therefore, the network will
not obtain the QoS patterns required for each service. The present work provides a planning
proposal to wireless networks regarding the influence of interference using computational
intelligence just as Bayesian Networks. An extensive measurement campaign was done to
evaluate the performance of two access points (PAs) under a multi user and interference
scenarios. The data collected in the measurement campaign was used as input of the Bayesian
networks and confirmed the influence of the interference in the QoS parameters. A genetic
algorithm technique was used as a hybrid approach to wireless planning. Another technique,
called particle swarm optimization (PSO) was used to compare the optimizations results from
the QoS parameters to find the best distance from the AP to the receiver to guarantee the QoS
ITU-T recommendations.
|
