A Wideband Spectrum Sensing Method Based on Compressed Sensing by Using Matching Pursuits
Academic Article
Overview
Identity
Additional Document Info
Other
View All
Overview
abstract
Secondary users are considered for using the licensed spectrum without causing harmful interference to the primary users in cognitive radios, which results in a challenge for spectrum sensing. Spectrum sensing is an ability of secondary users to independently detect spectral opportunities without any assistance from primary users. The methods based on standard analog-to-digital converters could lead to unaffordable high sampling rate or implementations for wideband spectrum sensing. Based on the compressed sensing theory, a wideband spectrum sensing method is presented. Gabor functions are selected to build the atom dictionary for exploring the sparse representation of the wideband signal. Matching pursuit algorithm is introduced to select the optimal atoms that can result in the most sparsity of representation. Finally, Wigner-distribution is used to reconstruct the spectrum of the wideband spectrum on the sparse representation. Simulation results show that this method can greatly decrease the sampling rate of the wideband signal and sense the primary users existence successfully.
