论文部分内容阅读
以四阶和六阶累积量作为特征,研究了IQ失衡状态下发射机或接收机中数字调制样式分类过程。此外,还提出了多种监督学习方法来降低接收机中IQ失衡带来的影响,包括k-最近邻(k-NN)、支持向量机(SVM)和决策树学习。同时,还研究了发射机中IQ失衡带来的影响,以及每种调制样式中IQ失衡对理论累积量的影响。仿真表明,监督学习法可有效补偿接收机中的IQ失衡。
Taking the fourth-order and the sixth-order cumulants as features, the classification process of digital modulation pattern in transmitter or receiver under IQ imbalance is studied. In addition, a variety of supervised learning methods are proposed to reduce the impact of IQ imbalance in receivers, including k-nearest neighbors (k-NNs), support vector machines (SVMs), and decision tree learning. At the same time, the influence of IQ imbalance in the transmitter and the effect of IQ imbalance in each modulation style on the theoretical cumulant are also studied. Simulation shows that supervised learning method can effectively compensate IQ imbalance in the receiver.