Undersampling is an effective way to reduce the cost of digital predistortion (DPD), but it poses challenges for signal alignment, especially for fractional delay alignment. In this letter, a Taylor expansion-based method for fractional delay approximation solution is proposed. To verify the proposed alignment algorithm, the basis manifold regularization (BMR) is introduced to ensure that the DPD can converge to a satisfactory performance even at an ultralow sampling rate. Simulations and experimental tests demonstrate that the proposed method not only has high alignment accuracy but also low complexity in comparison with traditional methods.

中文翻译：

---

欠采样数字预失真的高效信号对齐算法

欠采样是降低数字预失真 (DPD) 成本的有效方法，但它对信号对齐提出了挑战，尤其是对于分数延迟对齐。在这封信中，提出了一种基于泰勒展开的分数阶延迟近似求解方法。为了验证所提出的对齐算法，引入了基本流形正则化（BMR）以确保即使在超低采样率下 DPD 也能收敛到令人满意的性能。仿真和实验测试表明，与传统方法相比，所提方法不仅具有较高的对齐精度，而且复杂度较低。