Noise estimation plays an essential role in enhancing the performance of non-coherent spectrum sensors such as energy detectors. If the noise energy is misestimated, detector performance may deteriorate. In this paper, we present an energy detector based on the behavior that Empirical Mode Decomposition (EMD) has towards vacant channels (noise-only). EMD decomposes time-series signals into a finite set of components called Intrinsic Mode Functions (IMFs). The energy trend of these IMFs (modes) is used to determine the occupancy of a given channel of interest. The performance of the proposed EMD-based detector is evaluated for different noise levels and sample sizes. Further, a comparison is carried out with conventional spectrum sensing techniques to validate the efficacy of proposed detector and the results revealed that it outperforms the other sensing methods.
M. H. Al-Badrawi, A. Nasr, B. Z. Al-Jewad, and N. J. Kirsch, “An Adaptive Energy Detection Scheme Using EMD for Spectrum Sensing,” in Proc. of IEEE Consumer Communications & Networking Conference (CCNC), Las Vegas, NV, pp.1-6, January 2017.