Design of modern antenna structures heavily depends on electromagnetic (EM) simulation tools. EM analysis provides reliable evaluation of increasingly complex designs but tends to be CPU intensive. When multiple simulations are needed (e.g., for parameters tuning), the aggregated simulation cost may become a serious bottleneck. As one possible way of mitigating the issue, the recent literature fosters utilization of faster representations, or surrogates, of the system at hand. Notwithstanding, conventional models are severely affected by the curse of dimensionality. In practice, modelling of antenna structures described by no more than a few parameters over narrow parameter ranges is possible. In the context of the structural complexity of modern antennas, this is hardly acceptable. This paper presents a novel technique for cost-efficient design-oriented modelling of multi-band antennas. Our approach integrates a recently reported nested kriging framework and the response feature technology. This combination enables rendering of reliable surrogates valid within broad ranges of geometry parameters while using small training data sets. Benchmarking against conventional modelling methods demonstrates superiority of the proposed framework in terms of both the predictive power of the surrogate and its setup cost. Design applications for antenna optimization are discussed as well.

现代天线结构的设计在很大程度上取决于电磁（EM）仿真工具。EM分析可对日益复杂的设计提供可靠的评估，但往往会占用大量CPU。当需要多次仿真时（例如，用于参数调整），汇总的仿真成本可能会成为严重的瓶颈。作为缓解此问题的一种可能方法，最近的文献促进了对现有系统的更快表示或替代的利用。尽管如此，常规模型仍然受到维度诅咒的严重影响。实际上，可以在狭窄的参数范围内对不超过几个参数描述的天线结构进行建模。在现代天线的结构复杂性的情况下，这几乎是不能接受的。本文提出了一种新颖的技术，用于低成本设计的多频带天线建模。我们的方法集成了最近报告的嵌套克里金框架和响应特征技术。这种组合可以在使用小的训练数据集的情况下，呈现在广泛的几何参数范围内有效的可靠替代变量。相对于传统建模方法的基准测试，在代理的预测能力及其设置成本方面都证明了所提出框架的优越性。还讨论了天线优化的设计应用。这种组合可以在使用小的训练数据集的情况下，呈现在广泛的几何参数范围内有效的可靠替代变量。相对于传统建模方法的基准测试，在代理的预测能力及其设置成本方面都证明了所提出框架的优越性。还讨论了天线优化的设计应用。这种组合可以在使用小的训练数据集的情况下，呈现在广泛的几何参数范围内有效的可靠替代变量。相对于传统建模方法的基准测试，在代理的预测能力及其设置成本方面都证明了所提出框架的优越性。还讨论了天线优化的设计应用。