Frequency Selective Surface Assisted Dynamic Spectrum Access for the Wireless Indoor Environment

This thesis investigates the impact of the use of Frequency Selective Surfaces (FSS) when applied to walls to improve the performance of indoor wireless communications. FSS controlled spectrum sharing is examined using a point-to-point network topology containing two different types of users, intra-room and inter-room, and considers a system with open spectrum access where all users have equal regulatory status. This approach is used together with FSS walls to smartly control resource assignment inside the building.

The FSS filter activation threshold is examined, using a threshold value measured from sensing interference in up to three spectrum bands. It is shown how using this threshold, and different FSS state activation strategies, can significantly improve the way an indoor wireless communications system can control its spectrum resources.

Different FSS activation strategies are explored. It is shown how the model where a specific value of FSS threshold is set and used throughout shows much better performance compared to situations where the FSS is either continually on or continually off. This performance can be further improved if a more deterministic value is used. This is achieved by using a sliding window average assessment of performance which aims to minimize the frequency of instantaneous FSS states changes; this means a statistical value is used to determine when to activate the FSS. The result shows that a longer sliding window tends to give a better performance for inter-room users without significantly decreasing the performance of intra-room users.

An analytical model of system performance using a two-dimensional Markov Chain is developed. Systems with One Available Spectrum (1AS) and Two Available Spectrums (2AS) have been analysed using a state-transition-rate diagram and global equilibrium expressions for both systems are presented.