For Full-Text PDF, please login, if you are a member of IEICE,|
or go to Pay Per View on menu list, if you are a nonmember of IEICE.
Vector Evaluated GA-ICT for Novel Optimum Design Method of Arbitrarily Arranged Wire Grid Model Antenna and Application of GA-ICT to Sector-Antenna Downsizing Problem
Tamami MARUYAMA Toshikazu HORI
IEICE TRANSACTIONS on Communications
Publication Date: 2001/11/01
Print ISSN: 0916-8516
Type of Manuscript: PAPER
Category: Antenna and Propagation
genetic algorithm, optimization, vector evaluation method, ICT, sector antenna, small antenna, Yagi-Uda antenna,
Full Text: PDF>>
This paper proposes the Vector Evaluated GA-ICT (VEGA-ICT), a novel design method that employs the Genetic Algorithm (GA) to obtain the optimum antenna design. GA-ICT incorporates an arbitrary wire-grid model antenna to derive the optimum solution without any basic structure or limitation on the number of elements by merely optimizing an objective function. GA-ICT comprises the GA and an analysis method, the Improved Circuit Theory (ICT), with the following characteristics. (1) To achieve optimization of an arbitrary wire-grid model antenna without a basic antenna structure, the unknowns of the ICT are directly assigned to variables of the GA in the GA-ICT. (2) To achieve a variable number of elements, duplicate elements generated by using the same feasible region are deleted in the ICT. (3) To satisfy all complex design conditions, the GA-ICT generates an objective function using a weighting function generated based on electrical characteristics, antenna configuration, and size. (4) To overcome the difficulty of convergence caused by the nonlinearity of each term in the objective function, GA-ICT adopts a vector evaluation method. In this paper, the novel GA-ICT method is applied to downsize sector antennas. The calculation region in GA-ICT is reduced by adopting cylindrical coordinates and a periodic imaging structure. The GA-ICT achieves a 30% reduction in size compared to the previously reported small sector antenna, MS-MPYA, while retaining almost the same characteristics.