Murat Simsek

The recent developments in microwave design

Artificial neural networks have been used as an important technique in microwave modelling and optimisation. This paper gives an overview and recent developments on the knowledge-based neural modelling techniques in microwave modelling and design. The knowledge-based artificial neural networks are constructed by incorporating the existing knowledge such as empirical formulas, equivalent circuit models and semi-analytical equations in neural network structures. When one of the knowledge-based methods can not provide sufficient accuracy, two of them can be used in the same modelling process. This combination of methods is named hybrid technique. Using knowledge-based techniques requires less training data and has better extrapolation performance than classical neural networks. The advantages of using knowledge-based neural network modelling are demonstrated with microwave device modelling applications.

An efficient inverse ANN modeling approach using prior knowledge input with difference method

Artificial Neural Networks (ANN) have emerged as a powerful technique for modeling. Since the embedding knowledge in ANN models is possible by the Knowledge Based ANN (KBANN) methods, more accurate results than classical ANN approach can be obtained with KBANN. Source Difference (SD), Prior Knowledge Input (PKI) and Prior Knowledge Input with Difference (PKI-D) are several methods to be mentioned which combines existing knowledge with ANN methods. The existing knowledge is obtained either by mathematical formulations, ANN modeling or measured data. The Prior Knowledge Input with Difference, which is the latest method amongst KBANN approaches is discussed in this work. We compared the response efficiency and time consumption performances of PKI-D and classical ANN methods to obtain model for Inverse Scattering Problem.

Knowledge Based Three-Step Modeling Strategy Exploiting Artificial Neural Network

Artificial Neural Network (ANN) is an important technique for modeling and optimization in engineering design. It is very suitable in modeling as it needs only the information based on relationship between the input and the output related to the problem. For further improvement in modeling, a priori knowledge about the problem such as an empirical formula, an equivalent circuit model, and a semi-analytical equation is directly embedded in ANN structure through a knowledge based modeling strategy. Three-step modeling strategy that exploits knowledge based techniques is developed to improve some properties of conventional ANN modeling such as accuracy and data requirement. All these improvements ensure better accuracy with less time consumption compared to conventional ANN modeling. The necessary knowledge in this strategy is generated in the first step through conventional ANN. Then this knowledge is embedded in the new ANN model for the second step. Final model is constructed by incorporating the existing knowledge obtained by the second step. Therefore each model generates better accuracy than previous model. Conventional ANN, prior knowledge input, and prior knowledge input with difference techniques are used to improve accuracy, time consumption, and data requirement of the modeling in three-step modeling strategy. The efficiency of three-step modeling strategy is demonstrated on the nonlinear function modeling and the high dimensional shape reconstruction problem.

The Efficiency of Difference Mapping in Space Mapping-Based Optimization

In space mapping, a time-consuming but accurate fine model is used along with a less accurate but fast coarse model to reduce the overall computational effort. In this work, techniques using the difference mapping concept will be introduced. These techniques are efficient in reducing the computational effort while improving convergence. Difference mapping is constructed similarly to the mechanism used in space mapping, but, unlike space mapping, it facilitates the use of terminating conditions based on the simultaneous use of input and output values. Rigorous mathematical expressions related to difference mapping techniques will be given, and the improvement provided by these techniques will be discussed. Furthermore, to expose the efficiency of using the difference in input and output, simulation results obtained for high-dimensional applications will be given.

The recent developments in knowledge based neural modeling

Abstract Artificial neural networks have been recognized as an important technique in microwave modeling and optimization. This paper gives an overview and recent developments on the knowledge based neural modeling techniques in microwave modeling and design. The knowledge based artificial neural networks are constructed by incorporating the existing knowledge such as empirical formulas, equivalent circuit models and semi-analytical equations in neural network structures. The existing knowledge reduces the complexity of neural network model. This combination requires less training data and has better extrapolation performance than classical neural networks. The advantages of using knowledge based neural network modeling are demonstrated with two microwave modeling applications such as characteristic impedance modeling of thin-film microstrip line and parametric modeling of the differential via holes.

Developing 3-step modeling strategy exploiting knowledge based techniques

Artificial neural networks have been used as an important technique in modeling and optimization for engineering design. In this work, 3-step modeling strategy based on knowledge based techniques is proposed to develop new efficient modeling instead of conventional artificial neural network (ANN) modeling. The knowledge based artificial neural networks are constructed by incorporating the existing knowledge such as empirical formulas, equivalent circuit models and semi-analytical equations in neural network structures. In this new technique, required knowledge is created in the first step and used in the second step as a coarse model. Therefore each model shows better performance than former. In this strategy, conventional ANN, prior knowledge input and prior knowledge input with difference techniques are utilized not only to improve modeling accuracy but also to reduce time consumption during modeling. The advantages of using 3-step modeling are demonstrated on Branin function modeling application.

Development of knowledge based response correction for a reconfigurable N-shaped microstrip antenna design

This study presents the use of prior knowledge of inverse artificial neural network (ANN) to model and optimize a reconfigurable N-shaped microstrip antenna. Three accurate prior knowledge inverse ANNs with large amount training data are proposed where the frequency information is incorporated into the structure of ANN. The complexity of the input/output relationship is reduced by using prior knowledge. Three separate methods of incorporating knowledge in the second step of the training process with a multilayer perceptron (MLP) in the first step are demonstrated and their results are compared to EM simulation.

Space mapping with inverse difference technique for reconfigurable antenna design problem

In surrogate based optimization techniques, time consuming but accurate fine model is used along with less accurate but fast coarse model to reduce the overall computational effort. In this work, space mapping with inverse difference technique that is one of the space mapping based optimization techniques based on surrogate optimization is applied to antenna design problem. This technique provides efficient strategy to reduce computational effort while improving the convergence. Inverse difference mapping is constituted in terms of difference knowledge obtained by input and output of problem space. In addition inverse coarse model that is obtained by feed forward multi layer perceptron can generate necessary knowledge to form inverse mapping from coarse model input space to fine model input space instead of parameter extraction process for same knowledge. The efficiency of space mapping with inverse difference technique will be demonstrated by reconfigurable antenna design example in terms of its convergence and accuracy.

Efficient Reconfigurable Microstrip Patch Antenna Modeling Exploiting Knowledge Based Artificial Neural Networks

Artificial neural network (ANN) is widely used for modeling and optimization in antenna design problems. It is a very convenient alternative for using computationally intensive 3D-Electromagnetic (EM) simulation in design. The reconfigurable microstrip patch antennas have been considered to ensure operational frequencies for different kind of purposes. ANN is used for modeling of antenna design problems to obtain a surrogate based model instead of a computationally intensive 3D-EM simulation. Further improvement in modeling, a prior knowledge about the problem such as an empirical formula, an equivalent circuit model, and a semi-analytical equation is directly embedded in ANN structure through a knowledge based modeling technique. Knowledge based techniques are developed to improve some properties of conventional ANN modeling such as accuracy and data requirement. All these improvements ensure better accuracy compared to conventional ANN modeling. The necessary knowledge can be obtained by the coarse model which is a complex 3D-EM simulation in terms of grid size selection. Knowledge based techniques can improve the performance of conventional ANN through the guidance of the coarse model. As long as the coarse model approximates to the computationally intensive 3D-EM simulation, the performance of the knowledge based surrogate model can converge to the design targets. The efficiency of modeling strategies is demonstrated by a reconfigurable 5-fingers microstrip patch antenna. The antenna has four modes of operation, which are controlled by two PIN diode switches with ON/OFF states, and it resonates at multiple frequencies between 1 and 7 GHz. The number of training data is changed in terms of selected parameters from the design space. Three different sets are used to show modeling performance according to the size of training data. The simulation results show that knowledge based neural networks ensure considerable savings in computational costs as compared to the computationally intensive 3D-EM simulation while maintaining the accuracy of the fine model.

Convergence improvement of surrogate based optimization for reconfigurable antenna design using knowledge based inverse 3-step modeling

Engineering design process requires modeling and optimization to find optimum design parameters. While direct optimization only exploits time consuming but accurate fine model, surrogate based optimization exploits less accurate but fast coarse model to reduce the overall computational effort. In this work, space mapping with inverse difference technique is applied to antenna design problem together with efficient 3-step modeling. The combination of two techniques provides less computational effort and better convergence through the accuracy improvement based on the new inverse 3-step modeling strategy. The inverse coarse model which is used for the parameter extraction process during the optimization is realized by knowledge based inverse 3-step modeling. Inverse 3-step coarse model is obtained by multi layer perceptron in MATLAB ANN toolbox. The efficiency of the combination of space mapping with inverse difference technique and 3-step modeling strategy will be demonstrated by reconfigurable antenna design example in terms of their convergence and accuracy through its multiple operating frequency characteristic.