Institutional Repository
Technical University of Crete
Technical University of Crete
EN
|
EL
| URI: | http://purl.tuc.gr/dl/dias/F0C1568C-CCD9-48AB-8F70-AD696E5DF5CD | ||
| Year | 2005 | ||
| Type of Item | Conference Full Paper | ||
| License |
|
||
| Bibliographic Citation | K. Karamichalis, I.O. Vardiambasis, and G. Liodakis, "Computational investigation of asymmetric coplanar waveguides using neural networks: A microwave engineering exercise," in Proc. of the 2005 WSEAS Inter. Con. on Engin. Edu. (EE'05), July, pp. 8-10. | ||
| Appears in Collections |
In order to compute the characteristic impedance and the relative effective dielectric constant of an asymmetric coplanar waveguide with infinite or finite dielectric thickness, the use of artificial neural networks is valuable. The method of neural computing presented in this paper uses only one neural model for both parameters, for this specific waveguide type. The BFGS quasi-Newton back-propagation algorithm was used to train the developed neural network. Numerical results are given for several configurations along with comparisons with previously published data
Copyright © DIAS 2013
