功率放大器為通訊傳送端不可缺少的元件，然而功率放大器的非線性效應，會造成頻譜增生(spectral regrowth)，導致其他頻帶的訊號受到干擾，因此本論文透過數位預失真演算法，補償功率放大器之非線性效應。預失真演算法可以分為直接學習架構與間接學習架構，本論文根據文獻中的間接學習架構，來發展預失真演算法。然而文獻中並沒有對架構中的演算法參數設定有太多著墨，因此我們發展了一套簡易設定參數的流程，使得輸出功率可以達到預設目標的值。實測結果顯示此方法對寬頻或窄頻的訊號，都能有效補償功率放大器之非線性效應。
傳送訊號頻寬愈大，代表基頻操作的取樣頻率愈高，訊號處理的運算複雜度變隨之增加。因此本論文提出兩個方法來降低運算複雜度。第一個方法是將峰值因數削減方塊的位置由預失真器之前(較高取樣頻率)，轉移至較低操作頻率之調變器之後，以降低峰值因數削減方塊之操作頻率，再透過內插後輸入預失真器方塊。另外一個方法則是查表法，現有文獻中的方法仍然需要複數乘法器，本論文提出一個不需任何乘法器的方法，只是所需記憶體較大。實測結果顯示採用此二種方法之數位預失真器皆能有效線性化功率放大器之非線性現象。

Summary
The power amplifier (PA) is an essential element in the mobile communication systems. However, nonlinear characteristics of the PA results in spectral regrowth, which causes interference to other users in the adjacent frequency bands. In the thesis, we develop the digital predistortion (DPD) algorithm to compensate for the PA nonlinearity. DPD algorithms can be classified into the direct learning architecture (DLA) and indirect learning architecture (ILA). In the thesis, we study the ILA-based DPD algorithm. We first develop a simple procedure to set appropriate values for the algorithm parameters of the ILA such that the PA output power can reach the given targeted value. Experimental results on the narrowband and wideband orthogonal frequency division multiplexing (OFDM) signals show that, with the developed procedure, the DPD algorithm effectively compensates for the PA nonlinearity while reaching the specified PA output power.
Transmitted signals with a lager bandwidth imply a higher sampling frequency in baseband, which in turn implies a higher computational complexity. In the thesis, we propose two methods to reduce the computational complexity of the DPD algorithm. The first one is to lower the operating sampling frequency of the crest factor reduction (CFR) block by moving it from the DPD input (which operates at a higher sampling rate) to the output of the baseband modulator (which operates at a lower sampling rate). The output of the CFR block is then up-sampled and interpolated to a higher sampling rate before entering the DPD. The second method is to use the look-up table (LUT) to efficiently generate the DPD output. The LUT approach in the literature still needs real-valued and complex-valued multiplications. In the thesis, we propose a new LUT method that does not need any multiplications. However, the tradeoff is that the required table size is much larger. Experimental results on 5G OFDM signals demonstrate that, using these two methods, the DPD algorithm can effectively linearize the PA nonlinearity.

Keyword: power amplifier, spectral regrowth, digital predistortion, crest factor reduction, indirect learning architecture, look-up table.

[1] L. Ding, G. Tong Zhou, D. T. Morgan, Z. Ma, J. S. Kenney, J. Kim, C. R. Giardina, "A robust digital baseband predistorter constructed using memory polynomials. " IEEE Transactions on Communications, 52(1):159–165, 2004.
[2] LUT-based Digital Predistorter Implementation for FPGAs using LTE signals with 60 MHz Bandwidth, IET Science, Measurement and Technology, 2012.
[3] S. Benedetto and E. Biglieri, Principles of Digital Transmission With Wireless Applications. New York: Kluwer Academic/Plenum, 1999.
[4] D.R. Morgan, Z Ma, J Kim, M.G. Zierdt, J. Pastalan, "A Generalized Memory Polynomial Model for Digital Predistortion of RF Power Amplifiers." IEEE Transactions on Signal Processing, 54(10):3852–3860, Oct 2006.
[5] S. H. Han and J. H. Lee, “An overview of peak-to-average power ratio reduction techniques for multicarrier transmission,” IEEE Wireless Comm., vol. 12, no. 2, pp. 56–65, April 2005.
[6] “Behavioral Modeling and Linearization of RF Power Amplifiers,” John Wood, Artech House, 2014.
[7] “Behavioral Modeling and Predistortion of Wideband Wireless Transmitters”, Fadhel M. Ghannouchi, Oualid Hammi, Mohamed Helaoui, 2015
[8] S. Pipilos, Y. Papananos, N. Naskas, M. Zervakis, J. Jongsma, T. Gschier, N. Wilson, J. Gibbins, B. Carter, G. Dann, "A transmitter IC for TETRA systems based on a Cartesian feedback loop linearization technique", IEEE J. Solid-State Circuits, vol. 40, no. 3, pp. 707-718, 2005.
[9] A. Gokceoglu, A. Ghadam, M. Valkama, "Steady-state performance analysis and step-size selection for LMS-adaptive wideband feedforward power amplifier linearizer", IEEE Trans. Signal Processing, vol. 60, no. 1, pp. 82-99, 2012.
[10] J. Kim and K. Konstdntinou, "Digital predistortion of wideband signals based on power amplifier model with memory." Electronics Letters, 37(23):1417–1418, 2001.
[11] S. H. Han and J. H. Lee, “An overview of peak-to-average power ratio reduction techniques for multicarrier transmission,” IEEE Wireless Comm., vol. 12, no. 2, pp. 56–65, April 2005.
[12] N. Srirattana, A. Raghavan, D. Heo, P. E. Allen, and J. Laskar, “Analysis and design of a high-efficiency multistage Doherty power amplifier for wireless communications,” IEEE Transactions on Microwave Theory and Techniques, 53(3):852–860, March 2005.
[13] K. Moon, Y. Cho, J. Kim, S. Jin, B. Park, D. Kim, and B. Kim, “Investigation of Intermodulation Distortion of Envelope Tracking Power Amplifier for Linearity Improvement,” IEEE Transactions on Microwave Theory and Techniques, PP(99):1–10, 2015.
[14] “Digital Front-End in Wireless Communications and Broadcasting: Circuits and Signal Processing,” Fa-Long Luo, Ed., Cambridge University Press, 2011.
[15] Xiaowen Feng, “Efficient baseband digital predistortion techniques for linearizing power amplifier by taking into account nonlinear memory effect”