A Nonlinear Behavioral Modeling Approach for Voltage-Controlled Oscillators Using Augmented Neural Networks

This paper describes a method to model the nonlinear time-domain steady-state behavior of voltage-controlled oscillators (VCOs) using augmented neural networks. In the proposed method, a feed forward neural network (FFNN) with a periodic unit is used to capture the periodicity of the oscillatory output waveform. Inside the periodic unit, a second FFNN is used to map the control voltage to the instantaneous frequency. As opposed to the state space model which is based on a system of differential equations, the output of the oscillator is generated explicitly using the neural network presented in this paper. The model is trained using the data obtained from the simulation of transistor-level circuit models. The fidelity and speed-up of the model is demonstrated by an example of a transistor-level VCO. The proposed model is compatible with Verilog-A.