This paper presents a new piecewise two-dimensional (2D) digital predistortion (DPD) method for dual-band power amplifiers (PAs). A self-organizing map network is used to divide dual-band signals into different regions. Then, a principal component analysis algorithm is used to reduce the number of predistortion coefficients in each region. The proposed method is stable in finding the optimum region ranges and can achieve a good balance between the system complexity and model accuracy. To validate proposed method, a dual-band WCDMA signal and a dual-band OFDM signal are employed. The device under test is a GaN PA with a peak output power of 6 W. The adjacent channel power ratio and normalized mean square error measurement results for both cases demonstrate that the new piecewise DPD method can obtain better linearization performance with fewer coefficients.

中文翻译：

---

基于自组织映射的双频功率放大器二维分段数字预失真

本文提出了一种新的用于双频功率放大器 (PA) 的分段二维 (2D) 数字预失真 (DPD) 方法。自组织地图网络用于将双频信号划分为不同的区域。然后，使用主成分分析算法减少每个区域的预失真系数的数量。所提出的方法在寻找最佳区域范围方面是稳定的，并且可以在系统复杂度和模型精度之间取得良好的平衡。为了验证所提出的方法，采用了双频 WCDMA 信号和双频 OFDM 信号。被测器件是峰值输出功率为 6 W 的 GaN PA。