Comparative Performance Evaluation of Spectrum Sensing Techniques for Cognitive Radio Networks

Abstract

Cognitive radio is the key technology for future wireless communication. Spectrum sensing is one of the most important functions in cognitive radio (CR) applications. It involves the detection of primary user (PU) transmissions on a preassigned frequency band. PU licensed band can be sensed via appropriate spectrum sensing techniques. In this paper, we consider three basic spectrum sensing techniques of transmitter detection: Matched filter detection, Energy detection, and Cyclostationary feature detection. Using simulations, a comparative analysis of the three techniques has been carried out in terms of probability of false alarm Pf, probability of detection alarm Pd, and probability of miss detection Pm. Finally, Numerical result shows that at low signal to noise ratio (SNR), cyclostationary feature detection outperforms other two techniques, thus have some difficulties like implementation is complex, long observation time, etc. For simulation we used MATLAB software.