Title: | Using an ANN Approach to Estimate Output Power and PAE of A Modified Class-F Power Amplifier |
Authors: | Jamshidi, Mohammad Behdad Roshani, Saeed Talla, Jakub Roshani, Sobhan |
Citation: | 2020 International Conference on Applied Electronics: Pilsen, 8th – 9h September 2020, Czech Republic. |
Issue Date: | 2020 |
Publisher: | Západočeská univerzita v Plzni |
Document type: | conferenceObject konferenční příspěvek |
URI: | http://hdl.handle.net/11025/39907 |
ISBN: | 978-80-261-0891-7 (Print) 978-80-261-0892-4 (Online) |
ISSN: | 1803-7232 (Print) 1805-9597 (Online) |
Keywords: | komponent;umělá neuronová síť;výkonový zesilovač třídy F;přidaná účinnost |
Keywords in different language: | component;artificial neural network;class-F power amplifier;power added efficiency |
Abstract in different language: | In this paper, an efficient Class-F power amplifier (PA) is designed, simulated and modeled. This type of amplifier has nonlinear behaviors and uses tuning and controlling harmonics as the most important mechanism to increase efficiency. Feedforward artificial neural network (ANN) model is proposed to predict and estimate the nonlinear output of the power amplifier. The designed amplifier operates at 900 MHz, with 18 dB gain and 70 %Power-Added Efficiency (PAE). In the design process, the artificial neural network model is used to predict PAE and output power parameters as a function of input power, drain voltage and gate voltage of the applied transistor (DC Biasing voltages). The obtained mean relative errors (MREs) are less than 0.03% and 0.09% for the predicted output power and PAE parameters. |
Rights: | © Západočeská univerzita v Plzni |
Appears in Collections: | Konferenční příspěvky / Conference Papers (KEV) Applied Electronics 2020 Applied Electronics 2020 |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
09232787.pdf | Plný text | 452,47 kB | Adobe PDF | View/Open |
Please use this identifier to cite or link to this item:
http://hdl.handle.net/11025/39907
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.