概率神经网络在高功率微波探测数据处理中的应用

为了更好地处理高功率微波探测过程中产生的样本数据,在深入分析高功率微波特性参数的基础上,建立一个高功率微波器件特征参数库,并结合概率神经网络系统建立了一个高功率微波探测预测模型。通过部分学习样本和非学习样本进行预测,预测结果证明该模型能够基本再现原始数据,同时,对非样本数据有着较好的预测能力。这一数据处理方法在处理复杂样本、模式分类和判别过程中具有较高的实用性和实时性,能够在高功率微波探测数据的数据分类、结果预测等方面得到较好的应用。

Application of probability neural network to high power microwave exploration data processing

Neural network method is applied to processing exploration data of high power microwave(HPM). A HPM exploration prognosticating model based on analyzing HPM characteristic parameters and creating a HPM characteristic parameters data-base is established. Results can be reverted with training samples by this model. The model shows good ability on predicting. This data processing method can be used in classing or prognosticating exploration data, especially in complex data processing.