|摘要: ||本論文探討藉由直交表(Orthogonal Array, OA)的使用以進行天線的創新迭代設計，目標設定在設計一種平面式的雙頻天線，針對一以同軸線直接饋入的π型天線結構進行優化，期待能在有限且少量次數的迭代次數內，使天線結構達到我們所要求之優化設計。|
令上述的π型天線結構駐在一本研究係針對厚度為0.8mm的FR4基板(相對介電係數為4.4、Dielectric Loss Tangent為0.02)，且其頻寬所涵蓋的應用範圍包括ISM/Bluetooth、WiMAX、WLAN(802.11a/b/g/n)，針對此進行模擬分析與如下的優化設計，並和實作結果相驗證，其涵蓋應用範圍包括ISM/Bluetooth、WiMAX、WLAN(802.11a/b/g/n)。
In this thesis, the application of Orthogonal Array (OA) to design antennas through a novel iterative procedure is investigated. The objective is to design a planar dual-band antenna. A coaxial-fed π-shape antenna structure is optimized through the proposed novel procedure. It is expected that within a certain limited small iterations the S11 characteristics of the antenna can satisfy the required specificaion.
The antenna concerned is assumed to reside on an FR4 substrate (of thickness of 0.8mm, and with relative dielectric constant of 4.4 and loss tangent~0.02), and with frequency range covering the application of ISM / Bluetooth, and WiMAX, WLAN (802.11a/b/g/n). For the above conditions, the antenna simulation and the iterativelly optimized design are carried out, and finally the novel antennas are fabricated and verified by measurement.
For the novel iterative design procedure, the parameters of the π-shaped antenna are first chosen, and their ranges are suitably set, then a corresponding OA can be estabilished. Next, the fractional factorial experiments of the OA are carried out by using a commertial electromagnetic simulation software. Based on the simulation results, the fitnesses of the fractional factorial experiments are calculated such that the limited small number of iterations are conducted to optimize the antenna parameters. The objective is to test that whether the proposed novel antenna design procedure work to have the convergent antenna meet the pre-set frequency requirement or not,
As compared to other design method, the most important feature of the proposed novel method is the utilization of the OA, which exhibits the uniformity for the occurrence of each level of the parameters. The inclusion of OA is expected to accellerate the design process, reduce the search time. When the OA is combined with the range reduction technique, it is found quite suitable to serve as an alternative for antenna design.