作者: Zlatica Marinković , Nenad Ivković , Olivera Pronić-Rančić , Vera Marković , Alina Caddemi
DOI: 10.1002/JNM.2083
关键词:
摘要: A novel approach for parameter extraction of microwave transistor noise models based on artificial neural networks is proposed in this work. Neural are applied to determine parameters the model directly from measured and small-signal scattering without any optimization procedure. Moreover, unlike similar existing procedures, development procedure does not require data or optimizations a circuit simulator, making more efficient, as described detail paper. The has been Pospieszalski's specific pseudomorphic high-electron-mobility (pHEMT) device working under different temperatures. Copyright © 2015 John Wiley & Sons, Ltd.