Differential Sensing Under Varying Primary Activity Factor in Cognitive PANs

Cognitive radio technology necessitates accurate and timely sensing of the primary users' activity on the chosen set of channels. We assume incomplete sensing, in which the number of sensing nodes is smaller than the number of channels; the results of sensing are cooperatively combined to form a coherent channel map. As the times when spectral opportunities end are more critical for network performance, free channels should be sensed more frequently than the active ones. The paper presents a probabilistic analysis of this sensing policy, referred to as differential sensing, and investigates the range of values in which such incomplete sensing is capable of maintaining an accurate view of the status of the working channel set.