Behavioral Modeling of Millimeter Wave GaN Power Amplifiers for 6G Integrated Sensing and Communications Application

The increasing demand for sensing capability has emphasized the importance of integrated sensing and communications (ISAC) in 6G communication systems. In the ISAC scenario, the dynamic effect introduced by rapid switching between sensing and communication modes poses new challenges for PA behavior modeling. In this paper, a novel behavioral model is proposed to effectively characterize this dynamic effect. This model involves integrating the dynamic variable that mainly relates to the amplitude of the long-term signal into a neural network (NN) model, enabling accurate tracking of the nonlinear characteristics of PA. Experimental tests by employing a time-division ISAC signal demonstrate the superior modeling performance of the proposed model compared to existing models.