Font Size: a A A

Performance evaluation of hierarchical ad hoc networks

Posted on:2006-12-11Degree:M.A.ScType:Thesis
University:University of Windsor (Canada)Candidate:Yu, TaiFull Text:PDF
GTID:2458390005994432Subject:Engineering
Abstract/Summary:
Ad hoc networking is one of the most challenging areas of wireless communication. Theoretical analysis and experimental results show that QoS (Quality of Service) for each node degrades rapidly while the number of nodes increases in the network. One way to solve performance degradation is to use hierarchical network architectures.; In this paper, we investigate performance improvements offered by hierarchical ad hoc networks over flat (non-hierarchical or conventional) ad hoc networks for QoS parameters, namely throughput capacity, delay and power efficiency. We investigated and identified trade-offs among those QoS parameters via computer simulations carried by Network Simulator 2 of University of California (NS-2). In those simulations, we created hierarchical ad hoc networks by clustering the networks using cluster head nodes. Initially network is static (no mobility). Results of static network simulations act as benchmark for the performance parameters. Later mobility scenarios are added into the network to observe how mobility affects the performance. In order to compare two architectures, hierarchical and flat, we systematically changed number of nodes, data packet generation rates, number of clusters, node densities and transmission ranges for the nodes. At the same time, we compared hierarchical ad hoc network architecture with WLAN architecture, which has full infrastructure.; Simulation results state that throughput performance is linear with numbers of clusters; and in hierarchical architecture, power efficiency is doubled and delay is significantly lower than flat architecture. Our simulation results conclude that clustering schemes in wireless ad hoc networks can solve the scalability problem that exists in flat architectures.
Keywords/Search Tags:Ad hoc, Hierarchical ad, Performance, Results, Flat, Architecture
Related items