Electronic Thesis and Dissertation Repository

Degree

Doctor of Philosophy

Program

Electrical and Computer Engineering

Supervisor

Dr. Abdalllah Shami

Abstract

The ultimate goal of the next generation access networks is to provide all network users, whether they are fixed or mobile, indoor or outdoor, with high data rate connectivity, while ensuring a high quality of service. In order to realize this ambitious goal, delay, jitter, error rate and packet loss should be minimized: a goal that can only be achieved through integrating different technologies, including passive optical networks, 4th generation wireless networks, and femtocells, among others.

This thesis focuses on medium access control and physical layers of future networks. In this regard, the first part of this thesis discusses techniques to improve the end-to-end quality of service in hybrid optical-wireless networks. In these hybrid networks, users are connected to a wireless base station that relays their data to the core network through an optical connection. Hence, by integrating wireless and optical parts of these networks, a smart scheduler can predict the incoming traffic to the optical network. The prediction data generated herein is then used to propose a traffic-aware dynamic bandwidth assignment algorithm for reducing the end-to-end delay.

The second part of this thesis addresses the challenging problem of interference management in a two-tier macrocell/femtocell network. A high quality, high speed connection for indoor users is ensured only if the network has a high signal to noise ratio. A requirement that can be fulfilled with using femtocells in cellular networks. However, since femtocells generate harmful interference to macrocell users in proximity of them, careful analysis and realistic models should be developed to manage the introduced interference. Thus, a realistic model for femtocell interference outside suburban houses is proposed and several performance measures, e.g., signal to interference and noise ratio and outage probability are derived mathematically for further analysis.

The quality of service of cellular networks can be degraded by several factors. For example, in industrial environments, simultaneous fading and strong impulsive noise significantly deteriorate the error rate performance. In the third part of this thesis, a technique to improve the bit error rate of orthogonal frequency division multiplexing systems in industrial environments is presented. This system is the most widely used technology in next-generation networks, and is very susceptible to impulsive noise, especially in fading channels. Mathematical analysis proves that the proposed method can effectively mitigate the degradation caused by impulsive noise and significantly improve signal to interference and noise ratio and bit error rate, even in frequency-selective fading channels.

Share

COinS