大規模多輸入多輸出 (Massive MIMO) 天線系統是利用基地台配備數量龐大的天線，可以達到提升資料傳輸速度與能量使用效率的效果，預期是未來第五代行動通訊的關鍵技術之一。因其射頻功率放大器數量龐大，實現上必須考量硬體的成本。成本較低的射頻功率放大器線性度不佳，無法直接處理具有高功率峰均值比 (Peak-to-Average-Power-Ratio, PAPR) 的訊號。文獻裡提供的低功率峰均值比預編碼可以改善此問題，並且採用梯度下降法作為求解的工具。然而，梯度下降法的疊代過程導致計算複雜度高pp會大幅影響其求解速度。本論文的貢獻為使用收斂速度極快的概似訊息傳遞 (Approximated Message Passing, AMP) 演算法作為求解的工具。相較於文獻提出的求解方式，此方式所需的計算複雜度極低，相當適合於硬體上的實現。

Massive MIMO antenna system employs a few hundred base station antennas to simultaneously serve many tens of user equipments in the same radio channel. Such system can dramatically improve the data rates and energy-efficient, and thus is widely considered as a future cellular network architecture. Because the amount of RF power amplifier is large, reducing the hardware cost becomes a critical issue in implementation. Low‐cost RF amplifiers have poor linearization property so that they cannot be used to signals with high peak‐to‐average‐power‐ratio (PAPR). To overcome this problem, several works have considered developing low PAPR precoding techniques. Previous works have proposed the gradient descent (GD) method to find the low PAPR precoding. However, the GD method has high computational complexity. Our contributions include the introducing of the approximated message passing (AMP) algorithm in searching the low PAPR precoding. Compared to the GD method, the AMP algorithm is more suitable for the hardware implementation because it enjoys the much lower computational complexity.

論文審定書 i
摘要 ii
Abstract iii
目錄 iv
圖次 vi
表次 vii
1 緒論 1
1.1 背景以及動機 1
1.2 論文組織介紹 2
2 信念傳遞演算法簡介 3
2.1 和積演算法演算法簡介 3
2.2 概似訊息傳遞演算法簡介 9
3 低功率峰均值比預編碼介紹 14
3.1 CE 架構簡介 14
3.2 AC 架構簡介 20
4 使用 AMP 求解預編碼 24
5 模擬討論 27
5.1 常數封包預編碼架構 27
5.2 碟形限制式預編碼架構 30
5.3 連續時間模型下的功率峰均值比 32
6 結論 34
參考文獻 35

[1] F. Rusek, D. Persson, B. K. Lau, E. G. Larsson, O. Edfors, F. Tufvesson, and T.
L. Marzetta, "Scaling up MIMO: opportunities and challenges with very large
arrays," IEEE Signal Process. Mag., vol. 30, no.1, pp. 40-60, Jan. 2013.
[2] V. Mancuso and S. Alouf, "Reducing costs and pollution in cellular networks,"
IEEE Commun. Mag., vol. 49, no.8, pp. 63-71, Aug. 2011.
[3] S. K. Mohammed and E. G. Larsson, "Per-Antenna Constant Envelope Precoding
for Large Multi-User MIMO Systems," IEEE Trans. Commun., vol. 61, no.
3, pp. 1059-1071, Mar. 2013.
[4] C. Moll en, "Low-PAR Precoding for Very-Large Multi-User MIMO Systems,"
M.S. thesis, Dept. Electron. Eng., Linkoping Univ., Linkoping, Sweden, 2013.
[5] E. Hossain, D. I. Kim, and V. K. Bhargava, Cooperative Cellular Wireless Net-
works. New York: Cambridge, 2011, ch. 4.
[6] C. M. Bishop, Pattern Recognition and Machine Learning. Springer, 2006, ch. 8.
[7] F. Krzakala, M. M ezard, F. Sausset, Y. F. Sun, and L. Zdeborov a, "Statistical
physics-based reconstruction in compressed sensing," Physical Review X 2,
021005 (2012).