Performance analysis of different neural network models for parameters estimation of coaxial fed 2.4 GHz E-shaped Microstrip patch antenna

In this paper, a design and implementation of single-band E-shaped Microstrip patch antenna for IEEE 802.11b (2.38GHz ~ 2.455 GHz) frequency band represented. An E-shaped patch antenna with substrate thickness of h=2mm, relative permittivity of dielectric substrate is 2.55 and resonate at resonance frequency of fr = 2.4 GHz for Bluetooth applications is designed and simulated successfully [2-3]. The design and simulation of antenna is performed by IE3D (Method of Moment) based full wave electromagnetic simulator. After that, now the same antenna is modelled and analysed using different Artificial Neural Network (ANN) models for parameters estimation having `Translm' function based Multilayer Perceptron (MLP) and Radial Basis Function (RBF) model. An ANN used for analysis is firstly trained by using data set obtained from electromagnetic simulator of antenna using IE3D software. This model has two input parameters: x-coordinate and y- coordinate of probe feed point and has four output parameters: Resonant Frequency(fr) Return Loss (S11 parameter), VSWR (dB) and Input Impedance (Rin in ohm).