A novel compression ratio allocation method for collaborative wideband spectrum sensing

Spectrum sensing, as a key technology of cognitive radio (CR), needs to reliably and efficiently detect spectrum holes in wireless environments, which challenges the traditional spectral estimation methods typically operating at or above Nyquist rates. This paper develops a novel compression ratio allocation (CRA) method for wideband spectrum sensing in CR networks. In our scheme, each CR terminal performs compressed sensing with sub-Nyquist rate samples to scan a wide spectrum range at practical signal-acquisition complexity. It can greatly reduce the sensing measurements through fewer sample numbers. Meanwhile, the cognitive base station optimizes the compression ratio at each CR terminal according to their local signal-to-noise ratio (SNR), so the total sample number can be further cut down. Simulation results show that the CRA algorithm provides an optimal performance while requiring a relatively low complexity of sensing process.