Research of the Robust Optimum Design of Surface Acoustic Wave (SAW) Devices for Mobile and Wireless Communications

近年来，声表面波(SAW)器件被广泛应用于各种移动和无线通信系统中，如个人数字助理和移动电话。声表面波器件的频率响应特性主要由其几何结构决定：在压电衬底上制作的叉指换能器(IDT)和光栅反射器的结构。为了在计算机模拟的基础上确定合适的声表面波器件结构，通常采用包含多个不确定常数的叉指换能器的等效电路模型。为了更好地处理这些常量参数不可避免的离散性带来的设计缺陷，提出了一种稳健优化设计方法。首先，将声表面波器件的稳健优化设计问题描述为一个约束优化问题，即在满足期望函数的约束条件下，使声表面波器件的稳健性最大化。在此基础上，提出了一种结合变邻域搜索的罚函数方法，并将其应用于约束优化问题。此外，为了解决声表面波器件的大规模稳健优化设计问题，还提出了一种进化算法与遗传局部搜索算法相结合的遗传局部搜索方法。计算实验表明，所提出的稳健优化设计方法在不影响声表面波器件性能的前提下，提高了器件的稳健性。

In recent years, surface acoustic wave (SAW) devices have been used wildly in various mobile and wireless communication systems, such as personal data assistants and mobile phones. The frequency response characteristics of SAW devices are governed primarily by their geometrical structures : the configurations of interdigital transducers (IDTs) and grating reflectors fabricated on piezoelectric substrates. For deciding desirable structures of SAW devices based on the computer simulation, the equivalent circuit model of IDT, which includes several uncertain constant parameters, is usually used. In order to cope well with the designing imperfections caused by the inevitable dispersion of these constant parameters, a robust optimum design technique is proposed. First of all, the robust optimum design of SAW devices is formulated as a constrained optimization problem, in which the robustness of SAW devices is maximized under the constraints of their desirable functions specified by frequency response characteristics. Then, a new penalty function method combined with a variable neighborhood search (VNS) is proposed and applied to the constrained optimization problem. Furthermore, in order to solve the large-sized robust optimum design problem of SAW devices, a genetic local search method combining an evolutionary algorithm with the VNS is also proposed. Computational experiments show that the proposed robust optimum design technique improves the robustness of SAW devices without losing their specified functions.