Machine learning-based nonlinear pre-distortion system

A method comprising:

• obtaining, at a wireless device in a wireless communications system, a transmit radio signal that is configured for transmitting to one or more radio receivers associated with the wireless communications system;
• accessing a nonlinear pre-distortion machine learning model associated with the wireless device, wherein the nonlinear pre-distortion machine learning model includes model parameters associated with values corresponding to one or more deployment scenarios in the wireless communications system, and at least one nonlinear function to correct radio signal distortion;
• generating a pre-distorted radio signal by processing the transmit radio signal using the nonlinear pre-distortion machine learning model;
• obtaining a transmit output radio signal by processing the pre-distorted radio signal through a transmitting radio signal processing stage of the wireless device;
• and
• transmitting the transmit output radio signal to the one or more radio receivers.