GHOST: Efficient Goodness-of-fit HOS testing signal detector for cognitive radio networks
Daniel Denkovski, Vladimir M. Atanasovski, Liljana M. Gavrilovska · 2012
Flexible, reliable and robust signal detectors are vital components of the emerging cognitive radio networks. This paper presents a novel Goodness-of-fit Higher-Order-Statistics Testing (GHOST) signal detector, which eliminates the drawbacks of the previous related detectors and offers dominant signal detection performances. The detection is made based on goodness-of-fit testing comparing the skewness and kurtosis of the received power spectrum with empirically estimated system noise statistics. This allows reliable detection of different types of signals with SNR even lower than -32dB. The detector can be easily applied to the existing radio hardware since it adapts to the specific noise characteristics.