摘要
针对基于决策判决理论的传统算法对通信信号调制识别在低信噪比下正确识别率较低的问题,提出了一种基于多种特征参数判决的改进算法。利用高阶矩取代了利用零中心归一化瞬时幅值绝对值的标准偏差来区分ASK信号,对FSK信号采用功率谱谱峰数进行识别。仿真使用了包括模拟调制信号在内的多达11种不同类型的调制信号,仿真结果表明,在信噪比不低于8 dB下,该方法具有较高的正确识别率。
For the low identification accuracy of communication signal modulation based on traditional decision theory algorithm at low （SNR） signal noise ratio,an improved method based on multi-characteristic parameters is proposed.The ASK signals can be classi-fied by high-order moment instead of using the standard deviation of the absolute value of the normalized centered instantaneous ampli-tude.The FSK signals can be distinguished by the number of signal power spectrum peaks.The simulation uses as many as 11 types of modulation signals including some analog signals,its results demonstrate that the algorithm has higher identification accuracy when the SNR is not lower than 8 dB.
出处
《无线电工程》
2014年第5期30-33,共4页
Radio Engineering
关键词
决策理论
调制识别
高阶矩
功率谱峰数
decision theory
modulation identification
high-order moment
number of power spectrum peaks
