Determining resonant frequencies of various microstrip antennas within a single neural model trained using parallel tabu search algorithm

Sagiroglu S., KALINLI A.

ELECTROMAGNETICS, vol.25, no.6, pp.551-565, 2005 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 25 Issue: 6
  • Publication Date: 2005
  • Doi Number: 10.1080/02726340591007013
  • Journal Name: ELECTROMAGNETICS
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Page Numbers: pp.551-565
  • Keywords: microstrip antenna, resonant frequency, neural networks, parallel tabu search, GENETIC ALGORITHM, GLOBAL OPTIMIZATION, ELECTRICALLY THIN, NETWORKS, ELEMENTS, PROP
  • Middle East Technical University Affiliated: No

Abstract

Artificial neural networks (ANNs) are one of the popular intelligent techniques in solving engineering problems. In this paper, an intelligent new approach based on ANN trained with a parallel tabu search (PTS) algorithm to determine the resonant frequencies of microstrip antennas of regular geometries is presented. A single ANN model was used to determine the resonant frequencies of the rectangular, circular, and triangular microstrip antennas. The determination performance of a single neural model was improved with the help of PTS. The results obtained from the single neural model for the resonant frequencies of the rectangular, circular, and triangular microstrip antennas are in very good agreement with the experimental and other methods presented in the literature.