首页    期刊浏览 2024年11月22日 星期五
登录注册

文章基本信息

  • 标题:WiFi Access Point Selector Based on the Network Delay
  • 本地全文:下载
  • 作者:Pedro Mestre ; Alexandre Fonseca ; Paulo Salgado
  • 期刊名称:Lecture Notes in Engineering and Computer Science
  • 印刷版ISSN:2078-0958
  • 电子版ISSN:2078-0966
  • 出版年度:2018
  • 卷号:2235&2236
  • 页码:409-414
  • 出版社:Newswood and International Association of Engineers
  • 摘要:In WiFi, the algorithm used to choose the best candidate Access Point can vary from vendor to vendor, however it always involves the analysis of the Received Signal Strength Indication (RSSI) value. The simplest algorithm that can be implemented consists in choosing the Access Point that has the highest RSSI value. However, such an algorithm can lead to bad choices, for example when the Access Point with the best RSSI value has too many clients associated to it, which will be sharing the available bandwidth, while a further Access Point (with lower RSSI value) has no clients connected to it. Probably in this case the latter might offer a better network performance. To assess which is the best Access Point to connect to, in this paper it is presented a simple algorithm that is based in the network delay. Before choosing the best Access Point, the mobile station measures the network delay of the available Access Points, and chooses that with the lower delay. Te drawback is that this cannot be done off-channel, a connection must be made to the Access Point for the delay to be measured. Nevertheless, taking into consideration that that the overall network performance will be better, and if the user is using a nomad equipment (such as a laptop), this method is feasible. Results show that using network delay will lead to better choices. In one of the testing scenarios presented in this work, the method based on the network delay was able to obtain more than the double of best decisions in comparison to that using RSSI values (87.50% against 41.67%).
  • 关键词:WiFi; IEEE802;11; NTP; handoff; network delay; access point selection
国家哲学社会科学文献中心版权所有