應用於小型基地台之高效率極座標發射機

Abstract

射頻功率放大器為無線基礎設施之基地台應用中的關鍵元件。然而，高峰均功率比、寬帶寬與嚴格的線性度要求，藉由實現高效率之功率放大器的幾個設計挑戰做進一步的探討。另一方面，由於基地台大部分的功率消耗在射頻功率放大器中，因而效率扮演著重要的角色。 本論文的目標為發展適合於寬頻無線通訊系統 (如第四代行動通訊技術標準) 之高效率與高線性度的射頻功率放大器架構。而脈衝調變之極座標發射機架構引起了這類應用的研究興趣，因為此種架構能同時實現高效率與高線性度。 首先，驗證一小型基地台使用寬帶功率放大器於無混疊的脈衝調變之極座標發射機與討論其失真根源為輸入匹配之頻寬。藉由使用寬帶功率放大器，則帶寬增加導致非理想之上升與下降時間減少。在寬頻輸入匹配下，脈衝調變之極座標發射機能通過嚴格的線性度要求而不需要使用數位預失真校正技術。 接著，提出一種新式功率結合方法於多相位之脈衝調變之極座標發射機。藉由多相位無混疊數位脈寬調變技術應用於脈衝調變之極座標發射機，可提高等效之取樣頻率，且消除不需要之帶外雜散發射，亦能放寬帶通濾波器之需求。利用提高取樣頻率之方法，既有之帶通輸出匹配網路可提供顯著的雜散抑制效果。亦可從量測的寬頻輸出頻譜圖中，得知使用雙相位脈衝調變之極座標發射機能通過 -40dBm/100kHz LTE之雜散發射限制。此外，藉由使用所提出之新式功率結合器，衝調變之極座標發射機的效率能顯著的提高。 接著，驗證並討論使用無混疊多準位脈衝調變之架構實現一個線性 Doherty 功率放大器。此種 Doherty 功率放大器為首次實現線性的放大方式，而非連續調幅載波和連續負載調製的方式。從量測結果可知，使用簡單的無記憶數字預失真校正技術可以實現高效率和良好的線性度。無混疊多準位脈衝調變之 Doherty 功率放大器具有離散驅動的線性化優勢，且功率放大器消耗的功率更少，導致熱相關的記憶效應減少。 最後，本論文研究將有助於開發應用在蜂窩無線基地台發射機之高效率且具有良好線性度之脈衝調變極座標發射機。

A key component of the wireless infrastructure is the RF PA for the base station. However, the high PAPRs, wide bandwidth, and stringent linearity requirements have to be studied by several design challenges on implementation of high efficiency PAs. On the other hand, the efficiency plays an important role, since most of the power consumed by a base station is dissipated in the RF PA. The objective of this dissertation is to find the suitable linear high-efficiency RF PA architectures for wideband wireless applications such as LTE. The pulse-modulated polar transmitter (PMPT) architecture has attracted research interest for this type of applications due to this architecture can achieve both high efficiency and good linearity. First, the use of broadband PA for the aliasing-free PMPT for small-cell base stations is identified and discussed that the distortion roots from input matching bandwidth. By using a broadband PA, the bandwidth increases the resulting nonideal rise and fall times would decrease accordingly. With a broadband input matching, spectral requirements can be passed without using digital pre-distortion (DPD) correction. Second, a new method of reactive-combiner structure to be used as part of multi-phase PMPT. The multi-phase AFDPWM technique is used in conjunction with PMPT to increase the effective sampling frequency, which removes the unwanted out-of-band spurious emissions and aims to relax the BPF requirements. By pushing the spurs far away from the center frequency, the inherent band-pass output-matching network can provide significant suppression. It can be seen the measured output wideband spectrum using dual-phase technique, which passes the -40dBm/100kHz LTE spurious emission limit. By using reactive combining, the efficiency of the PMPT can be improved significantly. Third, an aliasing-free multi-level pulse modulation (MLPM) architecture to implement a linear Doherty power amplifier (PA) is identified and discussed. This work is the first reported results of this type of Doherty architecture to achieve linear amplification, rather than the usual situation of a continuous amplitude modulated carrier and continuous load modulation. The results show that both high efficiency and good linearization can be achieved with simple memoryless digital pre-distortion (DPD). Doherty operation with MLPM is believed to have benefits in linearization due to the MLPM discrete drive as less power is dissipated by the RF PA, leading to less thermal-related memory effects. Finally, this work will contribute to the development of high efficiency, good linearity PMPTs for cellular wireless base station transmitter.