基于谱域联合特征的信号调制类型识别

为了在低信噪比条件下识别信号的调制类型,提出了一种由信号瞬时包络的高阶组合矩、零中频信号的高阶组合累积量和谱相关函数共同构建的联合特征,用于数字调制信号调制类型识别的方法。新方法分析了多种常规数字调制信号的瞬时包络的幅度分布特性,利用了高斯过程的高阶累积量当阶次大于2时恒等于零的特性和谱相关函数突出周期平稳信号的特性,对高斯噪声具有较好的抑制作用。仿真试验结果表明,该方法运算简单,适应性强,在低信噪比下正确识别率高。

Modulation Recognition Based on Joint Characteristics in Spectrum Domain

In order to recognizing the modulation type of signals in low SNR,a new method for modulation recognition of digital modulated signals is put forward,using the joint statistics built by the high order combinatorial moment of the signals,the high order combinatorial cumulant of the zero-IF signals and the spectrum correlation function（SCF）together.The new method analyses the amplitude characteristics of the instantaneous envelope for common digital modulated signals,utilizing the properties that high order cumulant is identically null when the order is larger than two and that SCF gives prominence to the periodic stationary signals,which can better restrain the gauss noises.Simulation results show that this method is simple in calculation and has good adaptability,and the recognition probability is high even in low SNR.