蘇炫榮臺灣大學:電信工程學研究所徐志寧Hsu, Chih-NingChih-NingHsu2010-07-012018-07-052010-07-012018-07-052009U0001-3007200909432300http://ntur.lib.ntu.edu.tw//handle/246246/188311正交分頻多工存取(OFDM)可以在無線通訊中克服多路徑效應並且提供頻率分集，當使用轉傳通訊的時候，頻率分集可以更充分得被利用。在這篇論文中，假設傳送端知道頻道狀態，我們想要在給定總和功率限制的情況下，藉由共同最佳化兩個接連時間區段的子載波搭配與每個子載波的功率分配來最大化 OFDM轉傳系統下的權重總和速率，我們限制在每個子載波上只有一個節點可以傳送。當轉傳使用的時候，我們考慮解碼前送(DF)策略，除此之外當我們不使用轉傳的時候，會有兩種模式的傳輸，他們的差別是傳送源是否被允許在第二個時間區段傳送。首先我們將問題闡述成一個混和整數規劃問題，這是一個非多項式難題，接著我們利用了連續放寬，並且在對偶域利用次梯度方法解決了放寬過的問題。放寬過的問題對於其中一種模式是凸面的，另一種則不是。雖然有一個原本的問題是凸面的，但是限制評定並沒有滿足，這就是說對偶性缺口在兩種模式中都不是零，所以對偶最佳解對於每種模式提供了一個可達成速率的上界。然而，對偶性缺口在子載波的數量趨近於無限大的時候差不多是零的。因此我們可以達到接近最佳的解。最後我們延伸到使用獨立功率限制的系統上。Orthogonal Frequency Division Multiplexing (OFDM) can overcome multi-path effect and provide frequency diversity in wireless transmission. Frequency diversity can be further utilized when relayed transmission is employed. In this thesis, it is assumed that the channel state information (CSI) is known to the source. We aim to maximizehe weighted sum rate of OFDM relaying system under total power constraint by jointly optimizing subcarrier pairing in two consecutive time slots and power allocation on eachubcarrier in each time slot. We restrict that there is only one node transmitting on each subcarrier. Decode-and-forward (DF) strategy is considered when the relay is used. Besides, there are two types of transmissions when relay is not used, the difference is whether the source is permitted to transmit in the second time slot or not. First we formulate the problem as a mixed integer programming (MIP) problem which is NP-hard. Then we make continuous relaxation and solve the relaxed problem in dual domain by sub-gradient method. The relaxed problem is convex for one type and non-convex for the other type. Although one primal problem is convex, the constraint qualification is not satisfied. It means that the duality gaps for both types are not zero, and the dual optimal solution for each type provides an upper bound of the achievable rate. However, the duality gaps are virtually zero when the number of subcarriers goes to infinity. Thus we can achieve nearptimal solution. Finally we extend the system to have individual power constraint.1 Introduction 1.1 Thesis Organization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Relay 4.1 Introduction to Relayed Transmission . . . . . . . . . . . . . . . . . . . 4.2 Relay System Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.3 Relaying Protocol and Strategy . . . . . . . . . . . . . . . . . . . . . . . 7.3.1 Relaying Protocol . . . . . . . . . . . . . . . . . . . . . . . . . 7.3.2 Relaying strategy . . . . . . . . . . . . . . . . . . . . . . . . . . 8.4 Capacity Theorem on Relayed Transmission . . . . . . . . . . . . . . . . 9.5 Achievable Rate of Different Models . . . . . . . . . . . . . . . . . . . . 10.6 Summary for Chapter 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 Single User Subcarrier Pairing and Power Allocation under Total Power Constraint4.1 System Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14.1.1 Subcarrier Pair Channel Model . . . . . . . . . . . . . . . . . . . 15.1.2 Transmission Mechanism . . . . . . . . . . . . . . . . . . . . . 16.1.3 Achievable Rate for One Subcarrier Pair . . . . . . . . . . . . . . 17.2 Weighted Sum Rate Maximization for Type I Transmission . . . . . . . . 18.2.1 Primal Problem . . . . . . . . . . . . . . . . . . . . . . . . . . . 18.2.2 Dual Problem . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21.2.3 Interpretation of Weighting Factor as Quality of Service . . . . . 23.2.4 Analysis of Duality Gap . . . . . . . . . . . . . . . . . . . . . . 24.2.5 Achievability of the Optimal Value . . . . . . . . . . . . . . . . 25.2.6 Summary of Procedure . . . . . . . . . . . . . . . . . . . . . . . 26.2.7 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26.3 Weighted Sum Rate Maximization for Type II transmission . . . . . . . . 27.4 Simulation Result . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30.5 Summary for Chapter 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 Single User Subcarrier Pairing and Power Allocation under Individual Poweronstraint 34.1 System Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34.1.1 Major Effect on Individual Power Constraint . . . . . . . . . . . 34.1.2 Analysis of the system with fixed pairing . . . . . . . . . . . . . 35.2 Weighted Sum Rate Maximization for type I transmission . . . . . . . . . 38.2.1 Primal Problem . . . . . . . . . . . . . . . . . . . . . . . . . . . 38.2.2 Dual Problem . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39.3 Simulation Result . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41.4 Summary for Chapter 4 . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 Conclusion 45.1 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45.2 Future Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46ibliography 47757610 bytesapplication/pdfen-US正交分頻多工存取功率分配子載波搭配最佳化連續放寬對偶性缺口OFDMpower allocationsubcarrier pairingoptimizationcontinuous relaxationduality gapthesishttp://ntur.lib.ntu.edu.tw/bitstream/246246/188311/1/ntu-98-R96942060-1.pdf