Traffic-engineering based routing and channel allocation in wired and wireless networks

• Main Author: Khan, Junaid Asim
• Language: English
• Published: 2009
• Online Access: http://hdl.handle.net/2429/16934

Goal of traffic engineering (TE) in packet networks is to improve the network performance by providing support for congestion management, higher bandwidth utilization (or throughput), and QoS. There are two ways to provide congestion management, either by avoiding congestion before routing packet flows or by eliminating congestion after routing packet flows. Congestion can be eliminated in a network by capacity re-planning, however in wired networks it is not possible to perform capacity planning periodically. Therefore, wired networks rely on congestion avoidance that can be accomplished by using explicit path support in MPLS. This thesis proposes a fuzzy logic based TE routing algorithm to calculate these explicit paths. Simulation results have shown that proposed algorithm outperforms the well-known widest shortest path (WSP) algorithm and minimum interference routing algorithm (MIRA). The thesis also provides a TE solution in broadband fixed wireless networks with directed (or physical) mesh topologies. The solution approach exploits the fact that in wireless networks it is possible to perform capacity re-planning by re-planning the frequency channel allocation to links in a network. Unlike wired networks, wireless networks do not require any infrastructure upgrade to support channel reallocation in a short scale of time. The proposed solution is based on a distributed dynamic channel allocation algorithm that is capable of finding a solution at the time of network initialization and also dynamically fine tunes the channel allocation to eliminate congestion to provide traffic engineering. The proposed distributed dynamic channel allocation is highly scalable and hence is suitable for large networks. Simulation results have shown that channel allocation based on distributed dynamic channel allocation provides much better results than a fixed channel allocation based scheme. === Applied Science, Faculty of === Electrical and Computer Engineering, Department of === Graduate