IEEE Transactions on Microwave Theory and Techniques · 2011 · 195 citations · 30 references
EngineeringComputer ScienceElectrical EngineeringCircuit TheoryMachine LearningMicrowave EngineeringDevice ModelingNeural Circuit ModelingNeural ModelingNeural NetworksNonlinear Circuit AnalysisPower AmplifiersAccuracyDoherty Power AmplifiersEquivalent Circuit ModelingNeural Network ModelingNeural Network PredictionPower Amplifier NonlinearityRadio Frequency Power AmplifiersWideband Amplifiers
The paper proposes a novel two‑hidden‑layer artificial neural network (2HLANN) to predict the dynamic nonlinear behavior of wideband RF power amplifiers. The authors build a real‑valued, feed‑forward 2HLANN by transforming a generic low‑pass equivalent circuit, selecting network parameters and back‑propagation settings, and validate it on a 250‑W Doherty amplifier driven with a 20‑MHz signal, achieving accurate predictions of output spectrum, AM/AM and AM/PM characteristics, and low normalized mean‑square error. The 2HLANN‑based linearization of the Doherty PA attains up to 46.5 dBm output power, 40 % average efficiency, and an adjacent channel power ratio exceeding 50 dBc.
In this paper, a novel two hidden layers artificial neural network (2HLANN) model is proposed to predict the dynamic nonlinear behavior of wideband RF power amplifiers (PAs). Starting with a generic low-pass equivalent circuit of the PA, several circuit transformations are applied in order to build an appropriate artificial neural network structure and improve the modeling accuracy. This approach culminates in the development of a real-valued and feed-forward 2HLANN-based model. The parameters (number of neurons, memory depth, etc.) of the proposed model and the back propagation learning algorithm (learning rate, momentum term, etc.) used for its training were carefully studied and thoughtfully chosen to ensure the generality of the constructed model. The validation of the proposed models in mimicking the behavior of a 250-W Doherty amplifier driven with a 20-MHz bandwidth signal is carried out in terms of its accuracy in predicting its output spectrum, dynamic AM/AM and AM/PM characteristics, and in minimizing the normalized mean square error. In addition, the linearization of the Doherty PA using the 2HLANN enabled attaining an output power of up to 46.5 dBm and an average efficiency of up to 40% coupled with an adjacent channel power ratio higher than 50 dBc.
30
Neural Networks and the Bias/Variance Dilemma
Stuart Geman, Elie Bienenstock, René Doursat · Neural Computation · 1992 · 3.6K citations
Progress in supervised neural networks
D. Hush, B.G. Horne · IEEE Signal Processing Magazine · 1993 · 1.3K citations