Recently, stochastic geometry has been shown to be a very powerful tool to model, analyze, and design networks with random topologies such as wireless ad hoc and sensor networks as well as multi-tier cellular networks. In stochastic geometry analysis, point processes are used to model the positions and the channel access behaviors of the nodes. The thesis develops analytical frameworks to characterize the performance of large-scale wireless networks with random topologies. In particular, I use stochastic geometry tools to model, analyze, and design ad hoc networks, star-connected sensor networks, and infrastructure-based two-tier cellular networks. I have optimized the tradeoff between outage probability and spatial frequency reuse efficiency in carrier sensing-multiple-access based ad hoc networks. I have developed a novel spectrum efficient design paradigm for star-connected wireless sensor networks. For downlink transmission in cellular networks with cognitive femto access points (FAPs), I have quantified the performance gain imposed by cognition and developed a paradigm to optimize the spectrum sensing threshold for cognitive FAPs. Finally, I have developed a novel modeling paradigm for uplink transmission in cellular networks and obtained simple expressions for network performance metrics including the outage probability and average rate. Furthermore, I have revealed a transition point in the behavior of uplink transmission in cellular networks that depends on the relative values of the network parameters.
| Identifer | oai:union.ndltd.org:MANITOBA/oai:mspace.lib.umanitoba.ca:1993/23349 |
| Date | 27 March 2014 |
| Creators | Elsawy, Hesham Mahmoud Medhat Mahmoud |
| Contributors | Hossain, Ekram (Electrical and Computer Engineering), Irani, Pourang (Computer Science) Yahampath, Pradeepa (Electrical and Computer Engineering) Alouini, Mohamed-Slim (Physical Science & Engineering, King Abdullah University of Science and Technology) |
| Source Sets | University of Manitoba Canada |
| Detected Language | English |
Page generated in 0.0018 seconds