File(s) under permanent embargo
Reason: Restricted by author. A copy can be supplied under Section 51(2) of the Australian Copyright Act 1968 by submitting a document delivery request through your library or by emailing email@example.com
Intelligent vertical handover management in heterogeneous wireless networks
thesisposted on 28.02.2017, 01:05 by Haider, Ammar
The upcoming generation of wireless networks are characterized by seamless mobility across wireless access networks belonging to different radio access technologies. This requires the heterogeneous access networks to be integrated together, and the mobile terminals to be equipped with multiple radio interfaces allowing the users to roam across different networks. Vertical Handover (VHO) is the process of transparent switching from one access network to other. To achieve seamless heterogeneous mobility, simple, robust and effective vertical handover mechanisms must be devised. The entire vertical handover process can by divided into three stages: VHO Preparation, VHO Decision and VHO Execution. Ideally, a vertical handover should complete without causing any disruption in the communication and with a minimum overhead on the network. To achieve these goals, each of the three stages in VHO should be individually optimized. This research aims to devise a vertical handover implementation that performs as close to ideal behaviour as possible. For this purpose, we have proposed new or enhanced algorithms corresponding to all three VHO stages. A heterogeneous network model comprises of two major types of networks - cellular networks with wide coverage areas, and wireless local area networks (WLAN) with small coverage but providing higher bandwidth at a lower cost. To optimize the VHO preparation phase, we have developed a hybrid vertical handoff method for maximizing WLAN utilization in a WLAN-cellular integrated environment. We accurately determine the timing instants for initiation of vertical handover for the mobile terminals that enter the WiFi hotspots, stay there for some time and then leave the area. The proposed method ensures that transition between cellular and WiFi networks should be with minimum disruption while utilizing WiFi signals for as long as possible. As a result, the mobile nodes get advantages of lowest probability of handover failure and minimum number of unnecessary handovers. The second phase, VHO decision should incorporate multiple parameters to select the best network from a list of available candidate networks. The list of possible decision metrics includes link quality, system QoS parameters, user preferences, monetary cost, application type, movement speed etc. To arrive at a unified decision, as per our proposal, the best candidate network is chosen by employing a multiple-criteria decision making algorithm named PROMETHEE. This approach yields the desired benefit of providing better QoS experience to end users. For the third phase in vertical handover (VHO execution), we have focussed on the specific admission control problem. We have introduced concepts from Online Social Networking (OSN) Sites into admission control and handoff scenarios, by use of OSN proxy servers running on WLAN access points. Because of an ever-increasing popularity of OSNs this solution hold a lot of potential. Using the access points as OSN proxy servers leads to a substantial localization of social network traffic within WLAN which is a great advantage when considered globally. We have formulated and evaluated a Social-Connectivity-aware Vertical Handover (SCVH) algorithm for efficient dissemination of social network feeds. This algorithm introduces a novel measure of social importance of users requesting for admission to WLAN by calculating their 'graph centrality' measures. According to our proposed SCVH, users with high importance value are given a high priority for handover to WLAN network. Performance evaluation of our proposed techniques is carried out in a network simulation environment which models a heterogeneous network scenario. The simulation results demonstrate that our proposed schemes are successful in optimizing VHO performance, which should result in a nearly seamless handover across heterogeneous wireless networks.