Skip to Main Content
Majority of the existing works on cooperative spectrum sensing (CSS) in cognitive radio networks (CRNs) explicitly or implicitly assume that all secondary users (SUs) share the same occupancy of the licensed channel and the problem of CSS is always modeled as a common binary hypothesis test, i.e., the presence or absence of a primary user (PU), among all SUs. However, in practice, SUs at different locations may experience different (or heterogeneous) spectrum opportunities, in which case the common hypothesis does not hold any more. In this paper, we extend the traditional detection of the presence of the PU to the detection of spectrum opportunity in spatial and temporal domains jointly. Under a stochastic geometry model of the SU network, we drive the close-form expressions of spatial-temporal false alarm, detection and access probabilities for energy-based non-cooperative case. The simulated results match the theoretical results very well.