Abstract: To solve the problem of wideband spectrum sensing in cognitive radio networks， a spectrum sensing algorithm based on the spectrum structure of primary user signals was proposed， which is called DGS-SS algorithm. Firstly， compressed sensing theory was applied to signal acquisition to achieve a sub-Nyquist rate. Secondly， the group sparsity structure of primary user spectrum was used to modify the spectrum and residual support set during the reconstruction process， which can speed up the convergence and improve the accuracy of the reconstruction of primary user spectrum. Finally， effective detection of spectrum holes was given by the reconstructed signal spectrum. Simulation results show that the proposed algorithm can accurately reconstruct the spectrum at low compression ratio and have stronger robustness to noise variation， which makes the spectrum sensing performance significantly improved.

Key words: cognitive radio, spectrum sensing, compressed sensing, dynamic group sparsity, primary user signal reconstruction