ANN Modeling of Microstrip Hairpin-Line Bandpass Filter

In this paper a design technique for a Hairpin-Line narrow band Microstrip Bandpass filters is presented by using the artificial neural network (ANN) modeling method at mid-band frequency 2.2 GHz for S-band applications which give minimum insertion loss (S21) of 0.2954 dB and maximum return loss of 27.4 dB in the passband. Consequently an artificial neural network model is developed to observe the Magnitude variation of scattering parameters (S-parameters) of Microstrip Band-pass filters at 2.2 GHz for different dimensions. The developed ANN model of microstrip band-pass filter is computationally more efficient in the design and the results are more accurate as compared to an Electromagnetic simulator. Essential dimensions of the microstrip filter layout are used to obtain the input-output relationships in the ANN model. The simulation and ANN training is performed using the commercial electromagnetic simulation software Zeland IE3D 14.1 and MATLAB programming language respectively. KeywordsMicrostrip Hairpin-Line Bandpass Filters,Coupling Coefficient, ANN model, MATLAB, IE3D EM Simulation, S-parameters, Training Algorithm.