Spectrum sensing in Cognitive Radio based on compressive measurements

Cognitive Radio has attracted a lot of attention in the recent past due to the promise of a better utilization of the available spectrum. However, it faces many constraints in its implementation. Current spectrum sensing techniques are either computationally expensive or are not accurate enough. We propose a compressive signal processing (CSP) based approach for spectrum sensing that provides good accuracy at lower complexity. Since spectrum sensing involves sampling wideband signals, the sampling rates mandated by the Nyquist-Shannon sampling criterion tend be very high. This results in a heavy burden on the hardware devices (ADCs et al) and adds to the computational woes. Compressed sensing seems to be a natural solution to this problem. But typical compressed sensing algorithms involve signal reconstruction, and can be computationally expensive. By noting that spectrum sensing is an inference problem, we adopt the CSP approach that avoids reconstruction. Simulation results demonstrate that the proposed CSP based detector provides high accuracy at a reasonable complexity.