A framework for multidimensional modelling of spectrum occupancy in the simulation of cognitive radio systems

Abstract

Spectrum models play a key role in the research of Dynamic Spectrum Access/Cognitive Radio (DSA/CR) systems. Models existing to date are able to characterise the statistical properties of spectrum occupancy in only one of its dimensions (time, frequency or space). However, to the best of the authors’ knowledge, there is no model capable to account for the properties of spectrum occupancy in all its dimensions simultaneously. Since the development of a single model to characterise spectrum occupancy in such a holistic manner constitutes an extremely challenging task, this work explores an alternative based on the combination of existing dimension-specific models into a unified modelling approach. In particular, a framework is proposed for the integration of spectrum models that have been designed independently for a specific dimension of spectrum occupancy (time, frequency or space), thus enabling the multidimensional modelling of spectrum occupancy. A sophisticated simulation tool implementing state-of-the-art time, frequency and space dimension models is developed according to the proposed framework and used to generate synthetic spectrum data that simultaneously accounts for the properties of spectrum occupancy in all its dimensions. The use of such modelling framework in the contextof the simulation of DSA/CR systems is illustrated as well.