CABRAL, Y. K. S.; http://lattes.cnpq.br/9033486953815495; CABRAL, Yngrid Keila Silva.
Resumo:
Wireless networks that follow the IEEE 802.11 standard, popularly known as Wi-Fi networks, are present in the most diverse sectors, whether academic, industrial, commercial, etc. And due to the widespread use of wireless devices such as cell phones, notebooks and tablets, the adoption of this type of network is also likely to grow. However, the use of this technology can suffer from the effects of poor deployment or configuration planning, or even both. So, dealing with wireless network management in an environment where there are a lot of other devices operating, or even dealing with an environment where the signal coverage has not been properly planned, can present a big challenge for network administrators. For this reason, tools that provide access and visualization to network monitoring information are widely adopted by these professionals. However, the moment a failure occurs, the operator may need to perform a series of investigative analyzes on the available data, in addition to needing to carry out tests and verifications until the root cause is identified. So, depending on the failure witnessed, the time taken to find the cause, as well as to make the necessary corrections, can have significant negative effects on the use of the end user. Therefore, this work seeks to propose a mechanism that aims to detect four failures that are common in networks of this type, through the analysis of some metrics that are commonly used for monitoring. In addition, this work also presents an approach for the planning and distribution of access points between channels, based on the neighborhood information provided by these devices, as well as the overlap factor and the channel utilization weight. The results obtained through the experiments conducted show that the mechanism was able to correctly identify the failures at times when customers were exposed to them, and it was also observed that the approach to channel allocation was able to benefit the network elements, since it successfully led them to the best available channels.