Diagnosing Wi-fi Association Issues And Optimizing Network Performance

Abstract

Extensive deployment of Wi-Fi networks at homes and public places has triggered research interest in several aspects of Wi-Fi technology like deployment strategies and optimization. People use Wi-Fi or 802.11 networking, to connect their computers at home or office for instant internet access and it is a common scenario to have faced association and network congestion problems. Finding appropriate solution to these issues is a challenging task. In this thesis, an attempt has been made to solve Wi-Fi related concerns through a novel approach. Diagnosing problems occurring during the association process with the network and getting better throughput from the Wi-Fi connection are the most common Wi-Fi related concerns for a common user. Several diagnostics tools and network performance monitors have been developed in this direction. An efficient diagnosing and optimizing solution has been discussed in this thesis to improve user's experience with Wi-Fi networks. Choosing a good channel for the transmitting access point is important in a crowded network with numerous access points occupying the Wi-Fi frequency band. Also, in any large work area with distributed wireless routers, the devices have to connect to an access point which can deliver better throughput. The channel assessment and AP selection algorithm discussed in this thesis aims at addressing this concern. A novel approach to predict channel occupancy using beacon shift analytical model, calculate the maximum achievable throughput and compare the performance of a channel or an access point on the channel is proposed. The performance of the algorithm with respect to channel selection has been experimented. The diagnostics framework has been implemented on Android and the channel assessment algorithm has been tested for accuracy and can be integrated into the software.