A Novel Method to Compute Resonant Frequency of Metamaterial Based Patch Antennas Using Neural Networks
Journal Title: International Journal for Research in Applied Science and Engineering Technology (IJRASET) - Year 2016, Vol 4, Issue 1
Abstract
This paper presents a novel metamaterial based miniaturised patch antenna for wireless application, whose resonant frequency is computed using an artificial neural network approach. The conventional patch resonates at 2.5 GHz; but when loaded with Complementary Split Ring Resonator (CSRR) structures; it achieves miniaturisation and is then found to resonate at 1.77 GHz. The extent, to which metamaterial creates miniaturisation, is so far not addressed analytically and is conventionally computed by a trial and error method in some commercial antenna simulators. In this paper, an attempt is made to predict the resonant frequency of the patch after it is loaded with meta-structures using neural networks for which the neural tool box in MATLAB 10 is utilized. Of all neural algorithms, the feed forward Levenberg’s back propagation algorithm based is found to provide accurate results with minimum error when trained with sufficient number of inputs. The inputs and outputs for training the network are generated from rigorous parametric analysis carried out in ANSYS HFSS® an FEM based commercial software. The analysis results are presented along with the measured performance of an experimental prototype that is found in close agreement.
Authors and Affiliations
Gayathri Rajaraman, Khagindra Sood, S. Anbazhagan
Keywords
Related Articles
- EP ID EP21566
- DOI -
- Views 281
- Downloads 3
Design and Stress Analysis of AL-SIC MMC and Cast Iron Single Cylinder 4 Stroke Engine Crankshaft
the crankshaft is the important component in the automobiles. In this paper static analysis is conducted on the $ stroke engine crankshaft. The part modelling of the 2 crankshaft is made in the modelling tool UNIGRAPHIC...
The Faces of Engagement: A Method to Infer Emotions from Facial Action Units
We present a robust method to map detected facial Action Units (AUs) to six basic emotions. Automatic AU recognition is prone to errors due to illumination, tracking failures and occlusions. Hence, traditional rule base...
Topical Advancements in Various Spline Techniques for Boundary Value Problems
the present paper surveys and reviews papers on numerical solution of different types of boundary value problems by Spline based Technique. Among a number of numerical methods used to solve boundary value problems such...
Two Level Authentication System Based on Pair Based Authentication and Image Selection
This paper presents a two level authentication using pair based authentication and image selection. The most common method for authentication is textual passwords. Though textual passwords are easy to remember, they are...
slugDetermination of Atrial Diastole and Systole from 2-D Echocardiographic Images Using Artificial Neural Network
This project proposes a new approach to estimate the cardiac cycle phases in 2-D echocardiographic images as a first step in cardiac volume estimation. We focused on analyzing the atrial systole and diastole events. The...