Modeling and simulation of time-correlation properties of spectrum use in cognitive radio

The development of the cognitive radio technology can enormously benefit from the availability of realistic and accurate spectrum occupancy models. Time-dimension models of spectrum use proposed in the literature so far are able to capture and reproduce essential temporal characteristics such as the mean occupancy level (duty cycle) and the statistical distribution of busy/idle-period lengths. However, more sophisticated aspects of spectrum use, such as time-correlation properties, have not been properly addressed yet. In this context, this paper studies and analyzes, based on field measurements of various radio technologies, the time-correlation properties of spectrum use. Appropriate models and simulation techniques are proposed in order to accurately capture and reproduce such features.