本文为奥地利格拉茨技术大学的博士论文，共147页。
数字无线发射机中的射频功率放大器（RF-PA）是关系到能量消耗和信号质量的关键部件。特别是由于当今宽带多载波调制方法产生的信号峰均比很高，很难构建出能量效率高、满足标准严格线性要求的射频功率放大器。因此，射频功率放大器的数字预失真（DPD）已成为实现高效、高数据率无线发射机的关键技术。
本文研究了射频功率放大器的行为建模和DPD的理论基础和实用方法。主要贡献是建立了射频功率放大器联合线性效率特性的半物理模型，详细分析了射频功率放大器的多项式基带模型，重点分析了基带中经常被忽略的偶数阶项，并收集了射频功率放大器双频DPD的实用方法。
在半物理射频功率放大器模型的基础上，研究了射频功率放大器线性度与效率之间的折中权衡。为此，介绍了线性和效率量化的方法，并将其应用于该模型。此外，本文还对射频功率放大器的高效工作模式和效率提升方法进行了综述。
关于多项式基带模型的核心结果是基带中偶数阶项的第一种形式验证。这是通过推导伪无记忆多项式和Volterra级数的显式通带-基带对表达来实现的，这表明基带中的偶数阶项对应于通带中的修正基函数。多项式基带模型分析的另一个核心结果是相位均匀性要求的公式和证明，它表明了时不变通带系统所有复杂基带模型的必要对称性。
射频功率放大器双频段DPD的实用方法包括：通过削波和误差滤波来降低双频段峰值因数的方法；基于矢量切换广义记忆多项式的双频段DPD模型；基于间接和直接学习结构的双频段DPD训练算法。这些方法在MATLAB中实现，并在IEEE国际微波研讨会的2017年DPD竞赛中得到检验，获得了竞赛第一名。
The radio frequency power amplifier (RF-PA) within a digital wirelesstransmitter is a critical component regarding both the energy consumption andthe signal quality. Especially due to today’s broadband multicarrier modulationmethods that generate signals with high peak-to-average power ratio, it is veryhard to construct RF-PAs that achieve good energy efficiency and fulfill thestrict linearity requirements imposed by the standard. Because of this, thedigital predistortion (DPD) of RF-PAs has become a key technique forimplementing energy efficient, high data rate wireless transmitters. Thisthesis investigates theoretical foundations and practical methods for thebehavioral modeling and DPD of RF-PAs. The main contributions are asemi-physical model of the joint linearity-efficiency characteristics ofRF-PAs, a detailed analysis of polynomial baseband models of RF-PAs focusing onthe often neglected even-order terms in baseband, and a collection of practicalmethods for the dual-band DPD of RF-PAs. The trade-off between the linearityand efficiency of RF-PAs is investigated based on the semi-physical RF-PAmodel. For this purpose, linearity and efficiency quantification methods areintroduced and applied to the model. Furthermore, an overview on highly efficientRF-PA operation modes and efficiency enhancement methods is given. A centralresult regarding polynomial baseband models is the first formal justificationof
even-order terms in baseband. This is achieved byderiving explicit passband-baseband pairs for the quasi-memoryless polynomialand the Volterra series, which show that even-order terms in basebandcorrespond to modified basis functionals in passband. Another central result ofthe presented analysis of polynomial baseband models is the formulation andproof of the phase homogeneity requirement, which represents a necessarysymmetry of all complex baseband models of time-invariant passband systems. Thepractical methods for the dual-band DPD of RF-PAs include a method for dualbandcrest factor reduction by clipping and error-filtering, a dual-band DPD modelbased on a vector-switched generalized memory polynomial, and a dual-band DPDtraining algorithm based on the indirect and direct learning architectures.These methods were implemented in MATLAB and validated at the 2017 DPDcompetition at the IEEE International Microwave Symposium, reaching the firstplace within the competition.
1 引言
1.1 无线发射机的数字预失真
1.2 研究动机与研究范围
1.3 本文概述及研究贡献
1.4 与本文相关的博士论文
2 线性度与效率之间的折中权衡
2.1 引言
2.2 线性度与效率的联合模型
2.3 线性度与效率的评估
2.4 高效率的射频功率放大器
2.5 小结与结论
3 多项式基带模型的偶数阶项
3.1 引言
3.2 无需记忆的多项式模型
3.3 需要记忆的多项式模型
3.4 小结与结论
4 包络跟踪功率放大器的双频带数字预失真
4.1 引言
4.2 降低峰值因数
4.3 数字预失真
4.4 测量结果
4.5 小结与结论
5 总结与结论
5.1 结果与贡献总结
5.2 对比相关的博士论文
附录A 复信号
附录B 调制与解调
附录C Wirtinger微积分
附录D 线性最小二乘法
附录E 同步
附录F 组合数学
附录G 不定积分
附录H 定积分
下载英文原文地址：
http://page2.dfpan.com/fs/7l2c7j42028152e9166/
更多精彩文章请关注微信号：
