A Neural Network Modeling Approach to Power amplifiers Taking into Account Temperature Effects
In order to analyze the behavior of PAs at different temperature, this paper propose a neural network modeling approach to predict the temperature dependence of PAs. Three different neural network modeling methods are employed to develop the behavioral models. A 0.1-1.2 GHz CMOS PA is designed and fabricated to verify the effectiveness of the modeling method. Model results agree well with the measurement results. Three different models for the fabricated PA are compared in terms of accuracy. As far as author's knowledge, this is the first time of a temperature dependent behavioral model of CMOS PA is proposed to investigate the temperature effect on the performance of a CMOS PA.
|behavioral model, CMOS PA, extreme learning machine, neural network, temperature effect|
|2018 IEEE/MTT-S International Microwave Symposium, IMS 2018|
|Organisation||Department of Electronics|
Zhoul, S.-H. (Shao-Hua), Fu, H.-P. (Hai-Peng), Ma, J.-G. (Jian-Guo), & Zhang, Q.J. (2018). A Neural Network Modeling Approach to Power amplifiers Taking into Account Temperature Effects. In IEEE MTT-S International Microwave Symposium Digest (pp. 1028–1031). doi:10.1109/MWSYM.2018.8439468