A modified spectrum sensing method for wideband cognitive radio based on compressive sensing

In cognitive radio, secondary users require fast and accurate spectrum sensing, so that they can dynamically monitor the spectrum and rapidly tune their parameters to utilize the spectrum available, as well as avoid causing interference to primary users. The traditional spectrum sensing methods in a wideband cognitive radio are challenging to implement since they require very high sampling rates at or above the Nyquist rate. A new technique called compressive sensing (CS) can solve the problem, which exploits the sparsity of signal's frequency response. In this paper, a parallel spectrum sensing structure in cognitive radio is proposed. In the structure, we use compressive sensing and wavelet to process the signal in each branch. Then we get the final renconstruction output from the results of all branches. The modified sensing method based on this special structure is more accurate since it can reduce the effect of noise. Simulation results show the proposed method has performance improvement over traditional methods.