Jayanta Ghosh

Design of slot loaded proximity coupled microstrip antennas using knowledge based neural networks

In this paper, a knowledge based hybrid neural network (KBHNN) is utilized for designing of different slotted proximity coupled microstrip antennas. The slot loaded antennas can be designed from 1 to 6 GHz frequency ranges. By using this model, accuracy is found to be really beneficial, even if the required number of training data has been brought down to half. This method requires less time and scales down the complexities of the design processes. The solutions obtained by this neural approach are compared with the CST simulation results. The results of the KBHNN method are in good accord with the simulated values.

Hybrid method to predict the resonant frequencies and to characterise dual band proximity coupled microstrip antennas

ABSTRACT A new hybrid technique, which is a combination of neural network (NN) and support vector machine, is proposed for designing of different slotted dual band proximity coupled microstrip antennas. Slots on the patch are employed to produce the second resonance along with size reduction. The proposed hybrid model provides flexibility to design the dual band antennas in the frequency range from 1 to 6 GHz. This includes DCS (1.71–1.88 GHz), PCS (1.88–1.99 GHz), UMTS (1.92–2.17 GHz), LTE2300 (2.3–2.4 GHz), Bluetooth (2.4–2.485 GHz), WiMAX (3.3–3.7 GHz), and WLAN (5.15–5.35 GHz, 5.725–5.825 GHz) bands applications. Also, the comparative study of this proposed technique is done with the existing methods like knowledge based NN and support vector machine. The proposed method is found to be more accurate in terms of % error and root mean square % error and the results are in good accord with the measured values.

Design of electromagnetically-coupled microstrip antennas using neural networks

In this paper, back propagation neural network and radial basis function have been used for the design of electromagnetically coupled microstrip antennas. The antennas can be designed for different frequencies from 1 to 6 GHz using these two ANN models. A comparison is made between these two methods and radial basis function has been found to be better. So, further radial basis function has been used for the design of rectangular slot loaded ECMA. RBF takes less time and reduces the complexities of the design procedures. The results obtained by the RBF are compared with the simulation results obtained from the CST software and also with the measured results. The results of the RBF are in good agreement with the simulated and measured values.

Modeling of a compact dual band PIFA using hybrid neural network

In this paper, a compact dual band planar inverted-F antenna (PIFA) has been proposed. Dual band is achieved by inserting slots on the top radiating patch. The dimension of the patch is 15 × 12 mm2 and finite ground plane size is 44×40mm2 which can easily be integrated inside the mobile phone. Further, a hybrid neural network (HNN) is used for the design of dual band PIFA. This method is more accurate and requires less time. The HNN results are compared with the CST simulation results and are found to be in good accord.

Modeling of a compact triple band PIFA using knowledge based neural network

In this paper, a compact triple band planar inverted-F antenna (PIFA) has been proposed. Triple band is achieved by inserting slots on the top radiating patch. The dimension of the patch is 12 × 15 mm2 and finite ground plane size is 44 × 40mm2 which can easily be integrated inside the mobile phone. Further, a knowledge based neural network (KBNN) is used for designing of triple band PIFA. By using this method, accuracy is found to be really beneficial with the least amount of training data. This method requires less time and scales down the complexities of the design processes. The solutions obtained by this approach are compared with the CST simulation results. The results of the KBNN method are in good accord with the simulated values.

Feed Point Optimization using Neural Network

Microstrip antenna is gathering a lot of interest in communication systems. In this paper neural network approach has been used for calculation of feed position of microstrip antenna for maximum power transfer. This paper demonstrates the validity of neural network for the estimation of feed point of patch antenna varying with input impedance. Accuracy of the results encourages the use of Neural network.. Further simulations are done using CST software.

Design and Optimization of Proximity Coupled Antenna Using GA

Microstrip antenna is gathering a lot of interest in communication systems. Genetic algorithm is a popular optimization technique and has been introduced for design optimization of proximity coupled antenna. Patch length, patch width are taken as optimization parameters. Return loss and radiation pattern for the optimized antenna are verified using IE3D software. Accuracy of the results encourages the use of genetic algorithm. Further a parasitic patch coupled to electromagnetically coupled microstrip antenna is designed for broadband operations using IE3D software

A Modified Crown Square Fractal Patch Antenna for Dual and Multiband Operation

A simple and efficient method of achieving multiband operation on a microstrip patch antenna is presented on this paper. The patch is designed by simply modifying the crown square geometry, which is fractal geometry. Multiband characteristic is obtained by the self similar property of the designed patch. Method to vary the operational frequencies for one antenna is also suggested. The physical dimensional parameters of the designed antenna can be adjusted so as to obtain the frequencies to some useful frequency ranges. The proposed antennas are giving broadside radiation on the working frequencies. The antennas are designed using finite element method based HFSS and its performance is demonstrated in terms of return loss, radiation pattern and VSWR.

Application of genetic algorithm to the optimization of resonant frequency of coaxially fed rectangular microstrip antenna

Microstrip antenna is gathering a lot of interest in communication systems. Genetic algorithm is a popular optimization technique and has been introduced for design optimization of microstrip patch antenna. In this paper, genetic algorithm has been used for optimization of resonant frequency of coaxially fed rectangular microstrip antenna. The investigation is made at 3 different frequencies 3GHz, 5GHz and 10GHz respectively. Patch length, patch width & feed position are taken as optimization parameters. Return loss and radiation pattern for the optimized antenna are verified using IE3D software. Accuracy of the results encourages the use of genetic algorithm. Keywords - Rectangular microstrip antenna, genetic algorithm, resonant frequency, IE3D.