Download or read book Markov Decision Process Based Vertical Handoff Management in Heterogeneous Wireless Networks written by Shusmita Anwar Sharna and published by . This book was released on 2012 with total page 280 pages. Available in PDF, EPUB and Kindle. Book excerpt: Vertical handoff management is a very complex process in heterogeneous wireless environment. Maximization of user satisfaction often conflicts with the capitalization of network utilization as native users face increased competitions from the foreign users to access limited radio resources of a network and, more importantly, multiple networks compete to serve the same user pool. This thesis exploits these competitions to develop an efficient vertical handoff management system in heterogeneous wireless environment using a powerful modelling tool named, Markov decision process. A new vertical handoff decision algorithm is developed with novel competition-sensitive reward function formulation, which outperforms the existing algorithm in terms of improved expected values of QoS parameters such as bandwidth, delay, and jitter. A novel weight estimation technique is then defined, which can adaptively control the width of the spanning of estimated weights, to provide the ultimate flexibility for users to control the effective range of QoS parameters. Vertical handoff management should not be considered in isolation as the impact of call admission control policy's priorities to new, vertical handoff, and horizontal handoff calls is significant in the long run for maximizing user satisfaction. A network-centric Markov decision process based call admission control scheme is developed for multi-class traffic to analytically model the expected call dropping rates of vertical and horizontal handoff calls of a network. According to the model, these rates are primarily influenced by the priorities assigned by a network to different types of calls. In order to provide an effective feedback mechanism to the competing networks so that their priorities can be adaptively adjusted accordingly, the reward function of mobile-centric vertical handoff decision algorithm is further refined by considering the effect of call dropping rates of vertical and horizontal handoff calls on long-term user satisfaction. This is the first successful step in designing an integrated vertical handoff management system capable of significantly improving user satisfaction and widening the span of QoS operability, while concomitantly allowing networks to adapt call admission control policies to optimize network utilization.