Enhancing the Performance and Stability of a Dual Adaptive Biased Doherty Power Amplifier for 5G Mobile Handsets

A dual adaptive bias (DAB) control technique has been proposed for Doherty power amplifiers (DPAs) in mobile handsets to improve robustness against load variations in various operating environments. However, the practical deployment of this method has thus far been hindered by the operational complexity of establishing bias settings across varying power levels. In this study, we address these challenges by improving the feedback loop gain to enhance overall performance. Increasing loop gain introduces two main challenges, including amplified errors within the feedback loop and potential stability degradation. To mitigate these issues, we identify the drive level detector (DLD) as the primary source of errors and propose techniques to substantially reduce these errors. Furthermore, we propose a design methodology that ensures system stability without excessively compromising loop bandwidth. Our approach achieved a stable loop gain exceeding +5 dB at 100 MHz. The experimental results demonstrate that the proposed improvements effectively suppressed distortion at low power under varying load conditions and partially replaced manual bias tuning with fully automatic tuning. This method also improved DPA performance under nominal load. We achieved a power-added efficiency of 43.6% and an adjacent channel leakage ratio (ACLR) of −35 dBc using a 100 MHz bandwidth 5G signal without digital predistortion (DPD). The proposed methodology includes a quantitative evaluation of loop gain and provides clear guidelines to improve performance and enhance robustness. This approach shows promise for accelerating the commercialization of next-generation DPAs.