基于一种新形核函数高斯过程的双频微带天线设计

针对采用高斯过程进行建模时，不同核函数形式有着不同学习效果的问题。提出了一种自定义的平方指数形式的核函数，并基于多项式函数拟合对这种新形的核函数进行了数值仿真。仿真结果表明，该核函数不但可以提高模型的精确度和有效性，而且可以提高模型的学习能力和泛化能力。最后，将基于该核函数的高斯过程建模方法用于矩形双频微带天线优化设计和WLAN双频单极子天线优化设计，进一步证明了这种方法是可行的和有效的。

Design of dual band microstrip antenna based on Gauss process with a new kernel function

In the process of modeling with Gauss process, different kernel functions have different learning effects. This paper presents a new kernel function which is defined as the square exponential form, and carries out the numerical simulation based on the polynomial functions fitting. The simulation results show that the proposed kernel function can improve not only the accuracy and efficiency of the model, but also the learning ability and generalization ability of the model. Finally, this paper applies the GP based on the kernel function to the design of rectangular dual band microstrip antenna and WLAN dual band monopole antenna, and proves the method to be feasible and effective.