Open AccessBlind continuous hidden Markov model‐based spectrum sensing and recognition for primary user with multiple power levels
Author(s) -
Liu Boyang,
Li Zan,
Si Jiangbo,
Zhou Fuhui
Publication year - 2015
Publication title -
iet communications
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.355
H-Index - 62
eISSN - 1751-8636
pISSN - 1751-8628
DOI - 10.1049/iet-com.2015.0090
Subject(s) - hidden markov model , computer science , markov model , spectrum (functional analysis) , power (physics) , speech recognition , markov chain , pattern recognition (psychology) , artificial intelligence , machine learning , physics , quantum mechanics
Spectrum sensing has been well studied because of its significance in cognitive radio. Different from the existing works which a primary user (PU) is assumed to have only one constant transmit power, a more practical scenario that the PU transmitting with multiple power levels is considered. A continuous hidden Markov model (CHMM)‐based blind algorithm for not only detecting the presence of PU but also recognising the transmit power level of the PU is proposed. The training problem of CHMM is solved by combining the wavelet singularity detection with k ‐means clustering algorithm. An effective method for estimation of the number of power levels is proposed. Two different strategies are designed to perform spectrum sensing. Simulation results show the efficiency of the proposed algorithm.