频率相关的正交调制非平衡的功放线性化技术

考虑系统整体的成本和复杂度的情况下,在5G无线通信系统中装备了大规模的天线和一些低成本的器件,例如功率放大器(PA)和正交调制/解调器(QM/QDM)等。对于这些硬件存在的非平衡影响,即使在考虑到渐近正交性的特性下也不能被忽略。文章针对正交调制/解调器存在非平衡时的数字预失真(DPD)提出了一种精准的模型提取方法。所提方法可以把解调器(QDM)的非平衡从PA的非线性特性中分离出来,因此可以提高求解PA的逆模型的精度。与现有的联合模型先提取然后分解的方法相比,该方法DPD系数的提取中避免了分解操作,更加直接简单。实验测试验证了该方法的线性化性能,并且将所提出的方法进一步推广应用在了多入多出(MIMO)系统中,同样验证了其具有很好的线性化能力。

Linearization of Power Amplifiers with Frequency Dependent Quadrature Modem Imperfections

In the 5th generation (5G) wireless communication systems equipped with large scale antennas, lowcost components, such as power amplifiers (PAs) and quadrature modulators/ demodulators (QM/ QDM), are preferred in consideration of the total system cost and complexity. The impact of these hardware impairments, however, cannot be fully neglected even in the presence of asymptotic orthogonality property. This paper proposes an accurate model extraction approach for digital predistortion (DPD) of non-ideal PA with both the frequency dependent QM and QDM imperfections. The proposed method is able to distinguish the QDM impairments from the PA distortion, and thus improves the accuracy of the inverse model of PA. Contrary to the existing method of joint model extraction and then decomposition, the extraction of the DPD coefficients in the proposed method is straightforward and thus the decomposition operation is avoided. The linearization performance of the proposed method is validated in the experimental test. The proposed method is further extended to MIMO (multiple-input multiple-output) systems, and its good linearization ability is also verified.