基于CAM的认知超宽带频谱感知方法

Spectrum Detection for Cognitive Ultra-wideband Based on Cyclic Autocorrelation Matrix

频谱感知是认知超宽带系统的核心部分,针对超宽带频带内授权信号类型确定的特点,为了弥补自相关检测不能够识别信号类型的缺点,提出利用信号的循环谱特征和自相关矩阵差异性来检测授权信号,该方法融合了自相关检测和循环谱检测的优点进行合作判决。仿真表明该方法在高低信噪比环境下均能比循环谱检测和能量检测得到更好地检测效果,因此适合于认知超宽带系统。

Spectrum sensing is one of key parts for CUWB system, since styles of targeted signals in the UWB spectrum band are predefined, in order to offset shortcomings of autocorrelation detection？which could not distinguish signal styles. This paper presents a sensing method employing the differences of cyclic feature and autocorrelation matrix between signals to distinguish whether the frequencies can be used by cognitive users, this method combines advantages of cyclostationary detection and autocorrelation detection to make a decision cooperatively. The simulation shows that performance of the proposed method is much better than the cyclostationary detection and energy detection both in high and low SNR environment, therefore, this method is fit for CUWB system.