最小二乘支持向量机在蒸发波导预测中的应用  

Application of least square support vector machine to predict the evaporation duct

作　　者：杨超[1] 郭立新[1] 吴振森[1] 

机构地区：[1]西安电子科技大学理学院,陕西西安710071

出　　处：《电波科学学报》2010年第4期632-637,共6页Chinese Journal of Radio Science

基　　金：国家自然科学基金(60971067)

摘　　要：针对蒸发波导中折射率剖面与雷达海杂波功率之间的非线性关系,利用最小二乘支持向量机预测了蒸发波导的折射率剖面。由于最小二乘支持向量机是基于数据库的一种学习算法,通过正向传播模型产生的训练数据库来训练最小二乘支持向量机。其中正向传播模型采用抛物方程方法,通过将抛物方程的传播损耗结果和实测数据进行对比,验证了正向传播模型的准确性。基于实测数据来验证最小二乘支持向量机,预测结果表明最小二乘支持向量机在蒸发波导预测中的准确性。According to the nonlinear relation between the refractivity profile of evaporation duct and the radar sea clutter, the least square support vector machine （LSSVM） is used to predict the refractivity profile of evaporation duct. As the LSSVM is a learn algorithm based on training database, the training database obtained by the forward propagation model are utilized to train the LSSVM. The parabolic equation is adopted as the forward propagation model, and the propagation loss calculated by the parabolic equation are compared with the measured data to validate the accuracy of the forward propagation model. The LSSVM is validated by the measured data, and the results show that LSSVM is accurate for the prediction of the evaporation duct.

关 键 词：蒸发波导 最小二乘支持向量机 抛物方程 海杂波 传播损耗 

分 类 号：TN011[电子电信—物理电子学]