Performance Assessment of Utilizing the Neural Networks and Adaptive Neuro-Fuzzy Inference System in Analysis of Planer Structures
Traditionally, models used to obtain the parameters of planer structures are empirical functions and not unique i.e, have many definitions according to the normalized width (w/h) substrate. In this paper, artificial intelligent techniques are used to analyze planer structures to obtain unique, fast and accurate models. Artificial Neural Network (ANN) and Adaptive Neuro-Fuzzy inference system (ANFIS) are utilized to compute the effective permittivity and characteristic impedance of Single Microstrip Line (SML). The same techniques are used to calculate an even and odd effective permittivity and characteristic impedance of Microstrip Coupled line (MCL). The feasibility study of both techniques in modeling the planer structures under study reveals that ANFIS is more accurate in predicting the characteristic parameters of both structures. The Normalized Root Mean Square Error (NRMSE) in case of ANFIS is, at most, one quarter of its counterpart of ANN model.
Keywords: Microstrip, coupled lines, ANN, ANFIS