论文信息 - Creating neural models using an adaptive algorithm for optimal size of neural network and training set

Creating neural models using an adaptive algorithm for optimal size of neural network and training set

An adaptive algorithm generating neural models of linear microwave components, capable of determining an optimal size of training set and network size is presented. Several waveguide and microstrip models were created and verified by comparing neural calculations to mode-matching and method of moments results of analysis of bandpass filters.

L. Balewski | M. Mrozowski | M. Mrozowski | L. Balewski

[1] P. Kozakowski,et al. Efficient implementation of the Cauchy method for automated CAD-model construction , 2003, IEEE Microwave and Wireless Components Letters.

[2] G. L. Creech,et al. Artificial neural networks for fast and accurate EM-CAD of microwave circuits , 1997 .

[3] M.C.E. Yagoub,et al. A robust algorithm for automatic development of neural network models for microwave applications , 2001, 2001 IEEE MTT-S International Microwave Sympsoium Digest (Cat. No.01CH37157).

[4] P. Meyer,et al. Creating accurate multivariate rational interpolation models of microwave circuits by using efficient adaptive sampling to minimize the number of computational electromagnetic analyses , 2001 .