一种基于非高斯性测度的认知无线电频谱感知新方法

Novel Spectrum Sensing Approach for Cognitive Radio Systems Based on Non-gaussianty Measuring

针对认知无线网络中小尺度授权用户的频谱检测问题,提出一种新的基于非高斯性测度的频谱感知新方法.由于在某一频段内,授权用户信号的出现会使得认知用户接收信号功率谱密度的统计分布特性发生显著改变,因此,利用峭度和偏度设计一个非高斯性测度统计量,进而提出利用非高斯性测度统计量来检测认知用户接收信号功率谱密度统计分布特性的变化,从而可以实现对认知无线网络中具有低发射功率的小尺度授权用户的准确检测.仿真实验结果表明,本文所提方法具有较好的频谱感知性能,并且对噪声的不确定性具有较好的鲁棒性.

For the small-scale primary users （PU） in the cognitive wireless network, a novel spectrum sensing method is proposed in this paper. In the cognitive wireless networks, since the power spectral density （PSD） of a secondary user （ SU ） ＇s received signal has the different statistical distribution when the PU is present or not. Based on this fact,an non-Gaussianity measuring statistic is designed utilizing the skewness and kurtosis to measure the varying of the statistical distribution of the PSD of the SU＇s received signal,which can be further used to detect the presence of the small-scale PUs. Simulation results show that the proposed method obtains a good spectrum sensing performance and is robust to the uncertain noise.