2019 IEEE 8th Joint International Information Technology and Artificial Intelligence Conference (ITAIC) | 2019
Energy Detection using Savitzky-Golay Smoothing Method for Spectrum Sensing in Cognitive Radio
Abstract
Reliable spectrum sensing is one of the key technologies for enabling spectrum coexistence. To overcome the challenges of the real-time spectrum sensing requirement in the dynamical noise level environment, a Savitzky-Golay smoothing method based on Welch periodogram (WS-G) is proposed for spectrum sensing in cognitive radio. The mathematical expressions of false alarm and detection probabilities are derived in accordance with asymptotic normality and independence of the Fourier transformation and validity and unbiasedness of the Savitzky-Golay (S-G) smoothing. The simulation results show that the proposed WS-G algorithm is robust to noise uncertainty. The energy detector can meet the requirement of real-time detection with low computational complexity. Meanwhile, the results has demonstrated the importance of the optimization of the S-G parameters during the conversion of spectrum into matrix form, specifically segment size and polynomial frame length of the fitting curve.