Cross layer design of network- based fully distributed mobility management for heterogeneous wireless networks

Current mobility protocols; such as MIPv6 and PMIPv6, are deployed in a hierarchical and centralized manner in which a single anchor at the core network handles all mobility signaling and data traffic forwarding. As the core of the mobile network is heavily loaded by inducing excessive traffic, C...

Full description

Saved in:
Bibliographic Details
Main Author: Murtadha, Muayad Khalil
Format: Thesis
Language:English
Published: 2016
Subjects:
Online Access:http://psasir.upm.edu.my/id/eprint/66883/1/FK%202016%20163%20IR.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Current mobility protocols; such as MIPv6 and PMIPv6, are deployed in a hierarchical and centralized manner in which a single anchor at the core network handles all mobility signaling and data traffic forwarding. As the core of the mobile network is heavily loaded by inducing excessive traffic, Centralized Mobility Management (CMM) suffers from several issues in scalability, reliability, signaling overhead and non-optimal routing. Therefore, the IETF introduced a Distribution Mobility Management (DMM) working group to overcome these issues. The DMM paradigm involves a flattened IP network architecture in which the mobility anchor is moved closer to the users and the control and data planes are distributed at the network edge. The DMM is divided into two categories, partially and fully DMMs. The aim of this thesis is to design and develop network-based fully DMM solutions for flat IP architecture by removing any centralized mobility anchor from network infrastructure. Several solutions for heterogeneous wireless networks have been proposed based on the cross layer design of layer2 (data link) and layer3 (network). The IEEE Media Independent Handover (MIH) framework and Logical Interface (LIF) concept are used to abstract the heterogeneity of wireless networks. First scheme is developed using modified MIH framework to carry the addresses of active anchored flows, meanwhile; a modified and extended version of PMIPv6 has been used in the second scheme to carry the addresses of anchored flows. Third scheme is developed based on further modifications in MIH and PMIPv6 protocols by excluding client participation in any L2 or L3 wireless mobility signaling. The LIF concept has been used in the proposing of fourth scheme. Last proposed scheme develops more flattened architecture by distributing both mobility management and authentication process during vertical handover procedure. The analytical modeling and simulation implementation have been used to evaluate the proposed fully DMM solutions. Analytically, the DMM reports 80% lower data cost compared to CMM, while simulation shows 37% reduction in the end to end delay compared to CMM in heterogeneous networks. Moreover, MIH based fully DMM solutions are more complex and show higher signaling cost, handover latency and packet loss compared to LIF based solutions. However, the MIH based solutions can provide Quality of Service (QoS) provisioning of future networks. In particular, MIH based scheme gives an average of 53% lower signaling cost than PDMM, while LIF based scheme reports 102% reduction in signaling cost compared to MIH based scheme with client participation. In addition, MIH based scheme without client participation produces 50% lower handover latency compared to MIH based with client participation. Moreover, distributed mobility and distributed authentication scheme reports 52% and 24% packet loss reduction compared to MIH based without client participation and LIF based schemes, respectively.