The Dynamic Spectrum Access (DSA) paradigm based on the Cognitive Radio (CR) technology has emerged as a promising solution to conciliate the existing conflicts between spectrum demand growth and current spectrum underutilization without changes to the existing legacy wireless systems. The basic underlying idea of DSA/CR is to allow unlicensed users to access in an opportunistic and non-interfering manner some licensed bands temporarily unused by the licensed users. Due to the opportunistic nature of the DSA/CR paradigm, a realistic and accurate modeling of spectrum occupancy patterns becomes essential in the domain of DSA/CR research. In this context, this paper provides an overview of the existing spectrum occupancy models recently proposed in the literature to characterize the spectrum usage patterns of licensed systems in the time, frequency and space dimensions.