Browsing by Author "Omar Zakaria"

Mobility and real-time traffic such as Voice over IP are two expanding areas within communication systems. The concept of combining these two areas contains several challenging problems. Providing real time data traffic to the real time application which cannot tolerate delay has become ever more important feature of mobile networks in recent years. Fast Hierarchal mobile IPv6 (F-HMIPv6) reduces the packet loss and improves the handover consequences in case of micro-mobility movement, by reducing the network registration time and the address resolution time. However it does not suit in case of macro mobility when the mobile node moves to different network with different MAPs (Mobility Anchor Points). More efforts are needed to support both macro and micro mobility (inter/intra domain movement). This dissertation aims to present and evaluate a new approach to support macro mobility movement. The evaluation of this proposed approach is done using simulation and analytical approach. The proposed enhancement is benchmarked with the standard Mobile IPv6 (MIPv6) and Hierarchical Mobile IPv6 (HMIPv6) which are proposed by Internet Engineering Task Force. The results obtained show that the proposed scheme has better performance than MIPv6 and HMIPv6 in term of packet loss and handover latency

Wireless Mesh Network is one of the promising architecture for providing last-mile broadband Internet connectivity to network users. The network capacity in 802.11-based single channel wireless mesh network is highly affected by interference caused by backhaul wireless links' transmissions. This makes it inadequate for the new deployment scenarios with high number of users and traffic demands. To increase the network capacity, mesh routers are equipped with multiple radio interfaces. As a consequence, various wireless links can simultaneously operate within a set of orthogonal channels instead of a single channel. Routing and channel assignment are fundamental challenges in such networks, where the two functions determine how the traffic distributes over different links and channels. Therefore, for a given traffic load distribution, both channel assignment and routing need to be efficiently determined. The interdependent nature of routing and channel assignment has attracted researcher's attention to address these two issues jointly. In addition, re-configuration is required in dynamic traffic loads to ensure optimal network resources utilization. Frequent re-configuration degrades the network performance. This is because re-configuration of channels and routes disrupt the network traffic and increase the packet loss and delay. The main objective of this research is to develop an efficient joint state-aware algorithm, which is capable of adapting the traffic load variation with less traffic disruption. In developing the proposed solution, the re-configuration cost should be identified and considered. Firstly, the problem is formulated as a multi-objective optimization problem. The aim of this optimization problem is to minimize four objective functions, namely the maximum channel-link utilization, average network contention, channel re-assignment cost and re-routing cost. Then a heuristic algorithm called State-Aware Joint Routing and Channel Assignment (SA-JRCA) is proposed to address these challenges. The proposed algorithm is compared with the proposal of Avallone et al., 2013 and the proposal of Raniwala et al., 2004. The ns-2 simulator is used for evaluation. The proposed and compared works are evaluated and analyzed based on various metrics, such as maximum channel-link utilization, average network contention, channel re-assignment cost, re-routing cost, average throughput, and average end-to-end delay. The proposed algorithm shows better performance compared with the other two proposals. A new metric is proposed to evaluate the network performance. The proposed average network contention metric shows more correlations with network performance than maximum channel-link utilization. The results show that the proposed algorithm achieved the highest packet delivery ratio with more consistency with the traffic variation. In contrast, the other two algorithms show degradation in the performance with higher traffic variation and their achieved packet delivery ratio, reduced by 13%, 21% respectively when the traffic load varied from 10% to 50%.