A Study on the Spectrum Sensing Techniques in Cognitive Radio System

• Main Authors: Shiu,Jr Jen, 徐志禎
• Other Authors: Wu, Wei Chiang
• Format: Others
• Language: en_US
• Published: 2011
• Online Access: http://ndltd.ncl.edu.tw/handle/26262011718326670817

碩士 === 大葉大學 === 電機工程學系 === 99 === This thesis emphasizes on the issue of spectrum sensing or awareness in cognitive radio systems. We compare the performances of Matched Filter (MF) detection (non-blind), Non-coherent detection (semi-blind) and energy detection (blind). Besides, exploiting the fact that cumulants of order higher than two suppresses Gaussian noise of arbitrary covariance, we propose a fourth-order cumulants-based blind (without exploiting the primary user’s signal waveform) spectrum sensing algorithm. Furthermore, we designing energy spectrum sensing and diversity spectrum sensing by Ultra-wideband(UWB) signals. We formulate the spectrum sensing problem to transform as a two-sided hypothesis testing problem and employs Neyman-Pearson decision rule to determine the optimum threshold. Simulations under different scenarios confirm that the proposed method is reliable (with low false-alarm probability and high probability of detection) even at very low SNR values.