Efficient Nonlinear Behavior Modeling Method for Voltage-Variable Capacitors

A voltage-variable capacitor is an essential part of the matching element in radio frequency (RF) circuits, including the transmitting and receiving channel of mobile devices. However, the nonlinear characteristic of the voltage-variable capacitor produces unwanted harmonics when driven by a large exciting signal, which causes electromagnetic radiation spurious emission (RSE) complications. This article presents an efficient nonlinear capacitance polynomial model method coupling with harmonics test systems to predict the nonlinear behavior of the voltage-variable capacitor. In addition, a simple neural network model is trained to reduce the need to manually obtain the nonlinear parameters. Moreover, for devices exhibiting strong nonlinearity, an improved long short-term memory (LSTM) network is proposed to predict harmonics. Ultimately, the nonlinear parameter prediction for such devices is obtained quickly by implementing our developed method, which provides intuitive guidance in the design phase.