Study on Periodogram and Correlogram-Based Spectrum Sensing Implementation on GNU Radio

In cognitive radio network, spectrum sensing process plays a major role in the detection of unused licensed spectrum, which can later be used by other users that do not possess any license. To the extent of our knowledge, the existing implementations of energy-detection-based spectrum sensing (EDSS) method in GNU Radio and software defined radio are mostly based on the periodogram power spectral density (PSD) estimation instead of the correlogram PSD estimation. In this work, we implement ED-SS method based on both periodogram and correlogram in GNU Radio. We exploit GNU Radio embedded Python block to develop some functionalities that are required to implement the aforementioned method and that currently have not existed in the GNU Radio platform. The developed ED-SS based on both periodogram and correlogram are evaluated both in term of the resulting PSD estimate and detection performance. We found that the PSD estimate based on correlogram has a smaller variation in the frequency domain compared to the one based on periodogram. The detection performance is examined by evaluating the amount of miss detection and false alarm events and we found that the ED-SS based on correlogram has a better detection performance. This can be observed in the receiver operating characteristic plot illustrating the probability of false alarm (P
fa
) and the probability of detection (P
d
) produced by different threshold value.