BP神经网络在载波调制度定标中的应用

介绍了运用MATLAB神经网络工具箱进行BP神经网络设计的基本方法与过程。在调制度定标过程中,将BP神经网络应用于调制度数值与副/主载波功率差值特性曲线拟合,并通过训练和仿真。验证了应用BP神经网络进行测试数据预测的优越性。仿真结果表明,BP神经网络应用于调制度定标具有较高的预测精度和良好的泛化能力,它为调制度定标工作提供了一种全新的思路和方法。

Application of BP Neural Network in Calibration of the Carrier Modulation Degree

Introduces the application of MATLAB toolbox neural network for BP neural network is the basic design method and process. In the process of system calibration, the BP neural network is applied in the system numerical and carrier/subcarrier power difference characteristic curve fitting, and through the training and the simulation verified the BP neural network application test data to predict superiority. The simulation results show that the BP neural network is applied in the system calibration is of high precision and good generalization ability, it provides a new thinking and method to calibration of modulation degree.