摘要
高阶多进制正交振幅调制(QAM)有较高的频谱利用率和容错能力。针对QAM信号高阶调制方式,提出了一种M-QAM信号的识别算法。首先经CZT、HOS估计频偏和相偏,然后从信号的四次方谱和分形维数中,提取一组鲁棒性强的特征参数,同时运用信号子空间估计信噪比,实现16QAM、32QAM、64QAM和128QAM信号的调制识别。在高斯白噪声背景下仿真表明,算法能适应较大的信噪比动态范围,整体信噪比大于13 dB时可以达到90%以上的识别率,且盒维数抗噪声性能较好。
The high-order M-QAM has higher spectral efficiency and fault-tolerance capability. Aiming at high-order QAM modu- lation,a modulation recognition algorithm for M-QAM signals is proposed. Firstly,the frequency offset and phase offset estimation with CZT and HOS is performed,then a set of feature parameters are extracted from signal quartic spectrum and fractal box dimension,mean- while,the signal subspace is used to estimate SNR,and all that signals modulation recognition of 16QAM,32QAM ,64QAM, 128QAM signals can be achieved. The simulation results show that nnder the background of Gauss white noise,this algorithm can be suitable to a lager SNR dynamic range,and achieve more than 90% recognition rate when SNR is above 13 dB,and well anti-noise performance of box dimension method.
出处
《无线电工程》
2014年第1期32-35,共4页
Radio Engineering
关键词
四次方谱
盒维数
子空间
盲信噪比估计
quartic spectrum
box dimension
subspace
blind SNR estimation
