Wideband compressed spectrum sensing has the problems of unknown signal sparsity and high overhead of secondary users sensing. Therefore

this paper proposed an efficient cooperative scheme of wideband compressed spectrum sensing. Firstly

based on learning

a sparsity adaptive learning prediction model was derived. Secondly

a wideband spectrum filtering algorithm is designed. Finally

a cooperative wideband compressed spectrum sensing scheme was proposed. The simulation results show that the fitting effect of the adaptive prediction model are better than the existing prediction model

and the proposed sensing scheme effectively reduces the sampling rate and spectrum reconstruction delay of secondary users.