Secondary spectrum usage in TV white space

Abstract

Currently, the use of TV frequencies is exclusively license based with the area not covered by licensed TV transmitters being known as TV white space. In TV white space, the spectrum can be reused by a secondary user. This thesis studies how the TV white space can be used by a cellular system. The study addresses the problems of how the access to the spectrum is arranged, how the spectrum usage is constrained and how much capacity a secondary system will have. The access to TV white space can be arranged by using spectrum sensing or a geolocation database. This spectrum sensing relies on the performance of the signal detection algorithm. The detector has to operate in a fading environment where it should identify very low signal levels. In this thesis, the detector performance in a slow and fast fading environment is modeled. The model indicates that for a sufficiently long measurement time the impact of the fast fading can be averaged out. Unfortunately, simple single antenna-based detectors are not able to operate at a low enough signal-to-noise level. We propose a novel multi antenna-based detection algorithm that is specially designed to operate in a fading environment. TV white space is characterized by the amount of spectrum available for secondary usage. Because of the signal detection errors, a system using the sensing-based access is not able to use the entire available spectrum. This dissertation provides a method for estimating the spectrum utilization efficiency. The method illustrates how the detection error level affects the amount of available spectrum. One of the central questions studied in this thesis is how to describe the interference generated by the secondary transmitters. In the conventional model, the interference is computed as the sum of the interfering powers from individual transmitters. An alternative approach, pursued here, is to characterize the transmitter by its transmission power density per area. With such a model, the interference computation is done by integrating over the secondary system deployment area. The proposed method simplifies the interference estimation process. In data communication systems the spectrum attractiveness depends on the data rate the system can provide. Within the scope of this work, the achievable data rate is computed for a cellular system. Such computation is described as an optimization problem. The solution to this problem is found by searching for the optimal power allocation among the cochannels and the adjacent channels of a nearby TV transmitter.