Wireless Network Congestion Management Using Predictive Analytics
Computer Science and Engineering
StatisticsView Usage Statistics
Wi-Fi Access Points (APs) deployed publicly are facing serious demands due tothe proliferation in Wi-Fi enabled devices. This becomes more prominent whenthe user crowd moves dynamically in space creating a sporadic usage pattern. Inorder to cater for the dynamically changing spectrum demands, we need to identifyareas with high spectrum usage that needs betterWi-Fi coverage. In this thesis,we aim to understand the dynamic spectrum usage over space, time and channels.The temporal and spatial analysis helps us to identify places that are highly congestedat any given time. The channel usage pattern determines channels that areover utilized and under utilized in the congested areas.The usage data from user devices can be analyzed to answer a number of possiblequestions in regards to congestion, access point load balancing, user mobilitytrends and efficient channel allocation. Using this data, we attempt to identifyWi-Fi usage trends in a dynamic environment and use it to further predict thecongestion in various locations. To accomplish this, we have used University ofNevada, Reno (UNR) to conduct our experiments where we use various supervisedlearning algorithms to find the existing patterns in spectrum usage insideUNR. Using these patterns, we predict the values for certain key attributes thatdirectly correlate to the congestion status of any location. Finally, we apply unsupervisedlearning algorithms to these predicted data instances to cluster them intodifferent groups. Each group will determine the level of congestion for any buildingat any time of any day. This way, we will be able to ascertain whether or notany place at any time in the future might require additional resources to be able todeliver wireless services efficiently.In an attempt to deliver wireless services in a resourceful manner, we also talkabout self-coexistence among networks where the secondary networks can accesslicensed bands without interfering with the primary networks. This technologyis referred to as dynamic spectrum access that allows the underutilized frequencybands to be used avoiding the need for additional resources. With all the secondarynetworks trying to access an available channel, there arises a game theoretic competitionwhere they want to get a channel for themselves by incurring as minimumcost/time as possible. We implement a predictive strategy in the networks forthem to land on an available channel in the shortest time possible minimizing thecollisions among themselves. Thus, we investigate various predictive algorithmsand observe how a self-learning approach can be helpful in maximizing utilities ofthe players in comparison to traditional game theoretic approaches.