基于循环平稳特征的频谱感知技术分析.docx

(4)本文提出了一种在正交频分复用信号中采用跳频子载波的循环平稳特征签
名方法,在频率选择性衰落信道中具有高可检测性,并具备很低的开销。
关键词:认知无线电 频谱感知 循环平稳 广义似然比检验多天线频谱感知
Abstract
Over the last decade, munications have grown rapidly, and more spectrum resources are needed to support numerous emerging wireless services. As the issue of spectrum scarcity es more obvious, cognitive radio technology was recently proposed. It can improve the spectrum utilization by allowing works (users) to borrow unused radio spectrum from primary works (users) or to share the spectrum with the works (users). One of the most ponents of cognitive radio technology is spectrum sensing. This thesis deals with the problem of spectrum sensing by exploiting the cyclostationarity of primary signals. The main contributions are as follows:
Many existing cyclostationarity-based spectrum sensing methods have employed the Dandawaté-Giannakis’ algorithm due to its constant false alarm rate (CFAR) property and robustness in the low signal-to-noise ratio (SNR) regime. The Dandawaté-Giannakis algorithm has assumed that the distribution of cyclic auto-correlation estimations has the same asymptotic covariance under the null hypothesis and the alternative hypothesis. It was derived in this thesis that this does not hold true for the spectrum sensing application and a generalized likelihood ratio test based on a different hypotheses formulation has been derived too. The asymptotic distributions of the proposed test statistics, which proves the CFAR property of the proposed algorithm, is also be given during the derivation. The simulation results show that the proposed GLRT algorithm is superior to the Dandawaté-Giannakis algorithm in detection performance.
Moreover, the proposed GLRT algorithm for multi-antenna cyclostationary spectrum sensing has been extended in the following study. The proposed multi-antenna spectrum sensing algorithm achieves a significant performance gain over its single-antenna counterpart by taking into account the cyclic auto-correlations ob