The unique contribution of this thesis is a resource allocation solution capable of handling mixed media applications within the constraints of a 60 GHz wireless network. The challenges of multimedia wireless transmission include high bandwidth requirements, delay intolerance and wireless channel availability. A performance analysis for a 60 GHz wireless network using the IEEE 802 .1S.3c protocol considers the impact of variable rate traffic on network throughput. The impact of network size on the optimal contention access period and superframe duration of the 802.1S.3c MAC layer is identified. A series of optimization techniques applied to the resource allocation problem are explored and compared against the criteria of execution time/network overhead and ability to support applications with both convex and non-convex utility functions. The Particle Swarm Optimization (PSO) method • emerges as the best solution for quality-rate optimized network utility maximization (NUM). A new channel time allocation PSO (CTA-PSO) is proposed with network-l inked swarm size, an increased diversity function and a learning method based on the personal best, Pbest, results of the swarm. CTA-PSO demonstrates improved Convergence speed while maintaining the QoS improvement of the NUM. A novel PTV utility function is also proposed in this work to describe the relationship between allocated channel time and the user perceived quality of the transmission. A further extension to CT A-PSG overcomes the issue of link blockage due to the short wavelength at 50 GHz. CTA-PSO demonstrates improved performance compared with resource allocation on the basis of fixed bandwidth allocation to individual services 50 that pre-planning of network resources is not required. The multimedia resource allocation solution presented in this thesis provides a practical solution for real time wireless networks.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:603066 |
| Date | January 2013 |
| Creators | Scott-Hayward, Sandra Dean |
| Publisher | Queen's University Belfast |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
Page generated in 0.0018 seconds