Smoothing the canonical piecewise-linear model: An efficient and derivablelarge-signal model for MESFET/HEMT transistors

In this paper we present the smoothed piecewise-linear (SPWL) model as a us eful tool in the device modeling field. The SPWL model is an extension of t he well-known canonical piecewise-linear model proposed by Chua, which subs titutes the abrupt absolute value function for a smoothing function (the lo garithm of hyperbolic cosine), thus providing the model with several intere sting properties. In particular, this function makes the model derivable, w hich is important to predict the intermodulation distortion behavior. Moreo ver, it allows one to control the smoothness of the global model by means o f a single smoothing parameter. The parameters of the model are adapted to fit the nonlinear function, while the smoothing parameter is selected accor ding to derivative constraints. The applied learning algorithm is a second- order gradient method. The proposed SPWL model is successfully applied to m odel a microwave MERIT transistor under optical illumination using real mea surements. The model receives as input the bias voltages of the transistor, the instantaneous voltages, and the optical power and provides the drain t o source current. The performance and computational burden of the SPWL mode l is compared with an empirical model and with some neural networks-based a lternatives.