论文部分内容阅读
针对当前调制识别算法在低信噪比下识别率不高的问题,提出一种结合高阶累积量和小波变换的混合调制识别算法。该算法利用了小波变换提取的两个特征参数,以及基于四阶和六阶累积量构造出一个新的特征参数,并应用反向传播(back propagation,BP)神经网络分类器对调制信号进行识别。仿真结果证明,该算法能够在信噪比低至2d B时,识别率仍可达到98%以上,由此证明了该方法的有效性和稳健性。
Aiming at the problem that the current modulation identification algorithm has low recognition rate at low signal-to-noise ratio, a hybrid modulation recognition algorithm combining high-order cumulant and wavelet transform is proposed. The algorithm uses two characteristic parameters extracted by wavelet transform and constructs a new characteristic parameter based on the fourth and sixth cumulants. The back propagation (BP) neural network classifier is used to identify the modulated signal . The simulation results show that the proposed algorithm can still achieve a recognition rate of more than 98% when the signal-to-noise ratio is as low as 2dB, which proves the effectiveness and robustness of the proposed method.