摘要
提出了一种新的基于最小二乘支持向量机的天线指向模型(LSSVM-PM)。不同于线性指向模型(Linear-PM)对天线指向偏差的分项进行代数相加,LSSVM-PM是对影响天线指向的因素进行考察,将因素变量作为支持向量机的输入,指向偏差作为输出,模型的求解是将原空间非线性地映射到一个高维的特征空间,然后在此空间中引入最小二乘法进行优化求解。因此,LSSVM-PM模型可解决天线指向偏差的非线性问题,同时将影响指向的外部因素考虑周全。实验数据模拟结果表明,该模型与线性模型相比,指向预测精度提高了17.36%,能更有效地提高天线的指向精度。最后,对LSSVM-PM模型的参数选择、学习样本数量及分布对模型精度的影响进行了分析。
A new antenna pointing model （LSSVM-PM） based on Least-square support vector machines （LS-SVM） is proposed in this paper. Different from the linear pointing model （Linear-PM） carring out algebraic addition to the sub-item of antenna pointing offset, the LSSVM-PM analyzs the factors affect antenna pointing, and it takes these factor variables as the LS-SVM inputs and antenna pointing offset as the outputs. The LSSVM-PM is established by the input space mapped non-linearly to a higher dimensions feature space, and in the feature space a optimization problem can be solved by the least squares method. So nonlinear problem of pointing error can be handled and all environment variables can be considered. Simulation result for testing datum shows that the LSSVM-PM can improve pointing precision by 17.36% by contrast with the Linear-PM. Finally, the LSSVM-PM accuracy influenced by model parameter, the number and distributing of learning samples is analyzed.
出处
《电波科学学报》
EI
CSCD
北大核心
2007年第5期804-809,共6页
Chinese Journal of Radio Science
关键词
天线
指向误差
最小二乘支持向量机
建模
预测分析
antenna, pointing error, least-square support vector machines, model construction, predict analysis
