The relationship between choice of spectrum sensing device and secondary-user intrusion in database-driven cognitive radio systems

As radios in future wireless systems become more flexible and reconfigurable whilst available radio spectrum becomes scarce, the possibility of using TV White Space devices (WSD) as secondary users in the TV Broadcast Bands (without causing harmful interference to licensed incumbents) becomes ever more attractive. Cognitive Radio encompasses a number of technologies which enable adaptive self-programming of systems at different levels to provide more effective use of the increasingly congested radio spectrum. Cognitive Radio has the potential to use spectrum allocated to TV services, which is not actually being used by these services, without causing disruptive interference to licensed users by using channel selection aided by use of appropriate propagation modelling in TV White Spaces.The main purpose of this thesis is to explore the potential of the Cognitive Radio concept to provide additional bandwidth and improved efficiency to help accelerate the development and acceptance of Cognitive Radio technology. Specifically, firstly: three main classes of spectrum sensing techniques (Energy Detection, Matched Filtering and Cyclostationary Feature Detection) have compare in terms of time and spectrum resources consumed, required prior knowledge and complexity, ranking the three classes according to accuracy and performance. Secondly, investigate spectrum occupancy of the UHF TV band in the frequency range from 470 to 862 MHz by undertaking spectrum occupancy measurements in different locations around the Hull area in the UK, using two different receiver devices; a low cost Software-Defined Radio device and a laboratory-quality spectrum analyser. Thirdly, investigate the best propagation model among three propagation models (Extended-Hata, Davidson-Hata and Egli) for use in the TV band, whilst also finding the optimum terrain data resolution to use (1000, 100 or 30 m). it compares modelled results with the previously-mentioned practical measurements and then describe how such models can be integrated into a database-driven tool for Cognitive Radio channel selection within the TV White Space environment. Fourthly, create a flexible simulation system for creating a TV White Space database by using different propagation models. Finally, design a flexible system which uses a combination of Geolocation Database and Spectrum Sensing in the TV band, comparing the performance of two spectrum analysers (Agilent E4407B and Agilent EXA N9010A) with that of a low cost Software-Defined Radio in the real radio environment. The results shows that white space devices can be designed using SDRs based on the Realtek RTL2832U chip (RTL-SDR), combined with a geolocation database for identifying the primary user in the specific location in a cost-effective manner. Furthermore it is shown that improving the sensitivity of RTL-SDR will affect the accuracy and performance of the WSD