卫星常用调相信号自动调制识别算法

针对高阶幅度相移键控（APSK）信号在卫星通信中的应用，提出了一种识别数字调相信号的新方法。在信噪比估计的基础上，利用信号包络的统计特征对MPSK，MQAM，16APSK和32APSK信号进行区分。理论推导和实验仿真验证了该统计特征具有对加性高斯白噪声和滚降系数不敏感的特性。根据鲁棒性较好的四阶循环累积量提出一种新的特征参数Q以实现MPSK和MQAM信号的类内识别。仿真表明，当信噪比达到5dB时，该方法拥有较好的识别率（〉95％）。

Automatic recognition for the satellite phase modulated signals

A new algorithm for automatic classification of digital phase modulated signals is proposed aiming for the application of high-order Amplitude Phase Shift Keying(APSK) signals in satellite communication. The algorithm utilizes signal envelope statistical feature to distinguish MPSK, MQAM, 16APSK and 32APSK based on the SNR estimation. Nevertheless, it is proved that this statistical feature parameter is insensitive to additive Gaussian noise and the roll-off factor of the pulse shaping filter by theoretical deduction and simulation experiments. Therefore a new statistical parameter Q is proposed according to four-order cyclic cumulant which is robust to the presence of carrier frequency offsets, time delay and phase offsets. {QPSK 8PSK} and {16QAM 32QAM} can be classified based on the parameter Q. Simulation results show that the proposed recognition algorithm can obtain high probability of correct identification(>95%) when the SNR is above 5 dB.