Snr-based Adaptive Spectrum Sensing for Cognitive Radio Networks

Cognitive radios are indispensable to shift paradigm from conventional exclusive spectrum assignment to dynamic spectrum access. They can boost up spectrum utilization significantly, by dynamically accessing unused primary spectrum while bringing no harm to primary users. The envisioned radio calls for fast and accurate spectrum sensing. Researchers are focusing on cooperative spectrum sensing to improve reliability but still there is room for improvement in local spectrum sensing. In cooperative spectrum sensing, it will be hard to cooperate with local network nodes in a short time as cognitive radio has to operate in heterogeneous wireless networks. Most of the renowned local spectrum sensing technique in the literature up to now is cyclostationary feature detection, although it is reliable but computationally complex. Other well-known local sensing techniques are energy detection and matched filter detection. This paper proposes an adaptive local spectrum sensing scheme, in which cognitive radio can adopt one-order cyclostationary or energy detector for spectrum sensing on the basis of estimated SNR, which is calculated in advance for available channels. Simulation results indicate that we can achieve reliable results equal to one-order cyclostationary detection with less mean detection time.