Design and optimization of SIW patch antenna for Ku band applications using ANN algorithms

Substrate Integrate Waveguides (SIW) present very compatible components to planar technologies since recent years that have has been widely implemented in microwave antennas as a class of effective integrated transmission lines to provide high quality factor capacities and incomparable self-consistent shielding. Artificial Neural Networks (ANN) present ones of the fundamental electromagnetic (EM) design automations through numerical optimizations which become nowadays ubiquitous in various modeling fields such as microwave engineering. Accordingly, this paper provides for the Ku microwave band (12–18 GHz), a new design of a patch antenna based on SIW technology using a tree-dimensional electromagnetic (EM) simulation based on structured supervised learning alternative to neural networks to provide accurate geometric dimensions for the target requirements. The SIW patch antenna is designed to operate in Ku frequency band and resonate at 16.10 GHz. The optimized antenna shows very low return losses of less than −10dB to −19dB for the selective band resulting in good performance. ANN algorithms implemented for the training process present than a reliable tool of estimating the antenna performance to provide precise geometrical dimensions with the specific requirements.

[1]  M. Chetioui,et al.  Design and Analysis of Ku/K-band Circular SIW Patch Antenna using 3D EM-based Artificial Neural Networks , 2018 .

[2]  Ke Wu,et al.  Accurate modeling, wave mechanisms, and design considerations of a substrate integrated waveguide , 2006, IEEE Transactions on Microwave Theory and Techniques.

[3]  K. C. Gupta,et al.  Emerging trends in millimeter-wave CAD , 1998 .

[4]  Wenquan Che,et al.  Equivalence between waveguides with side walls of cylinders (SIRW) and of regular solid sheets , 2005, 2005 Asia-Pacific Microwave Conference Proceedings.

[5]  Ke Wu,et al.  Review of substrate-integrated waveguide circuits and antennas , 2011 .

[6]  James T. Kwok,et al.  Constructive algorithms for structure learning in feedforward neural networks for regression problems , 1997, IEEE Trans. Neural Networks.

[7]  K. Wu,et al.  Integrated microstrip and rectangular waveguide in planar form , 2001, IEEE Microwave and Wireless Components Letters.

[8]  Qi-Jun Zhang,et al.  Neural Networks for RF and Microwave Design , 2000 .

[9]  M. Chetioui,et al.  A Novel Dualband Coaxial-fed SIW Cavity Resonator Antenna using ANN Modeling , 2018 .

[10]  Kurt Hornik,et al.  Multilayer feedforward networks are universal approximators , 1989, Neural Networks.

[11]  Ke Wu,et al.  Guided-wave and leakage characteristics of substrate integrated waveguide , 2005, IMS 2005.

[12]  Qi-Jun Zhang,et al.  Neural Networks for Microwave Modeling: Model Development Issues and Nonlinear Modeling Techniques , 2001 .