| Recent years have witnessed the rapid increase of demand for new wire-less services and applications as well as the number of wireless subscribers. Therefore, the total amount of mobile data traffic is also skyrocketing. Given the scarcity of the natural frequency spectrum, it becomes quite apparent that the current static frequency allocation schemes can not meet the demands of an increasing number of higher data rate devices. As a result, innovative tech-niques that can offer new approaches of exploiting the available spectrum are in need. Cognitive radio is currently regarded by regulators as the technology that would enable dynamic reuse of already licensed but spatially or temporarily unutilized spectrum, thereby enhancing availability for new applications.On the other hand, recently there is a trend worldwide characterized by the notable transition from analogue to digital terrestrial television. This dig-ital switch is expected to generate huge amount of free spectrum. These free spectrum, often referred to as TV White Spaces(TVWS), can be reused by low power devices on a non-interfering basis with regard to primary services. Cognitive Radio is considered to be the most feasible technology to reuse the TVWS.Successful operation in TV White Spaces hinges on the ability of cogni-tive radios to reliably detect and use secondary spectrum opportunities in TV White Spaces. Several approaches have been proposed and investigated that aims at achieving these manifold objectives. So far the utilization of a geoloca-tion database, which is capable of eliminating possible interference to primary services, is considered to be the most feasible solution.Being the focus of this thesis, geolocation database in cognitive network stores records of all licensed services in the TV frequency bands and can pro-vide a list of available channels and power levels for each geographical pixel based on the interference protection requirements.To put in a nutshell, this thesis conducts thorough research on following topics:Firstly, we make a brief description of the background of cognitive wire-less network. Next, we mainly introduces the three methods for obtaining in-formation, spectrum sensing, bacon and geolocation database.Focusing on the reuse of the TV white spaces, the architecture and the deployment scenarios of the cognitive wireless network in the TV white spaces and is overviewed. The database service providers would use the raw data together with other required regulatory inputs (algorithms, parameters and ex-ceptions) in the process of converting the information provided to the database into a list of allowed channels and associated transmit powers. We analyzed the raw data and the required algorithms, parameters, moreover, the detailed calculations which a geolocation database would need to perform in order to derive the available channel list and the location specific maximum permitted emission levels is presented.Afterwards, we analyzed the construction mechanism of the geolocation database and a novel method to reduce the complexity and cost is proposed and explained, which compress some compressible geographic pixels and use nonuniform dimensions of pixels in populating the database. With simulation and performance evaluation, we concluded that this method can significant-ly reduce the computational complexity and implement cost of geolocation database.Hereafter, this work was sponsored by National Basic Research Program of P.R.China (No.2009CB320400), we introduced the design and development of the demonstration platform of geolocation database in this program.In the last part of the thesis, conclusions are made about this research and the concentration of future research is indicated. |
PDF Full Text Request