HMM based spectrum sensing in the presence of censored data

Spectrum Sensing (SS) techniques play an important role in the Cognitive Radio (CR) systems. In recent years, many spectrum sensing techniques have been proposed in the literature to identify the state of the Primary Users (PUs) in the temporal domain. However, these techniques are usually interested in the current state of channel without consideration to their status in the past. In this paper, we applied Hidden Markov Model (HMM) for SS in Cognitive Radio Network (CRN) and employ an Expectation-Maximization (EM) method to estimate parameters of the HMM in the presence of censored data. Further, we present an optimal likelihood computation for censored data during the online channel status estimation procedure. Simulation results show the effectiveness of the proposed algorithm.