Temperature Compensation-Based Behavioral Modeling for Digital Predistortion of RF Power Amplifiers

The adaptive digital predistortion is currently widely used to compensate for the non-linearities. Such systems need not only sufficient computational power but also feedback. However, there are great demands on feedback paths to create such-and-such systems. Ideally, the digital predistortion should follow changes of behavior for example as an ambient temperature alternates. To compensate the non-linearities for different temperatures, one can use Look-Up tables [1] which consume considerable resources. In this paper we demonstrate how digital predistortion using polynomial series, with simple real valued function (parameter), can be extended in order to include thermal compensation trait. That means, we can eliminate recalculation of coefficients or more precisely no need for feedback path to compensate temperature depended behavior of power amplifiers.