李學智臺灣大學：電信工程學研究所陳楨明Chen, Cheng-MingCheng-MingChen2007-11-272018-07-052007-11-272018-07-052006http://ntur.lib.ntu.edu.tw//handle/246246/58727在這本論文中，我們提出了一個結合傳統球狀解調演算法(Sphere Decoding algorithm)與排序的漸續式干擾刪除(Ordered Successive Interference Cancellation)演算法來減少傳統球狀解調演算法的運算時間。我們將此演算法命名為SD-OSIC演算法。此演算法的錯誤率曲線(BER)能夠逼近最大相似(near Maximum Likelihood)準則，並且其所需要檢驗落在預設半徑內的晶格數目會大大的減少，因此整個解調的問題將降為較低維度的問題。 我們接著推導出SD-OSIC與傳統的球狀解調演算法其解調所需要的乘法與加法數目。模擬數據結果顯示我們所提出的演算法相較於傳統的球狀解調演算法需要較少的運算迴圈數目，並且在某些情況下相較於傳統的球狀解調驗算法SD-OSIC需要較少的運算量。藉由此兩種比較結果的支持，我們總結出我們的演算法相較於傳統的球狀解調演算法需要較少的運算時間。In this thesis we propose an algorithm which combines the Sphere Decoding algorithm (SDA) and the Ordered Successive Interference Cancellation algorithm (OSIC) to reduce the computation time of the SDA. We name it SD-OSIC algorithm. This algorithm has the near-ML (Maximum Likelihood) Performance and needs to search fewer lattice points inside the sphere radius, hence the searching problem are reduced to a lower dimension ones. We also derive the number of multiplication and addition operations of SD-OSIC and conventional SDA. Numerical simulation reveals our proposed algorithm requires less loop iterations than conventional SDA and in some scenario SD-OSIC requires less computation power than the SDA. With the supports of these two comparisons, we conclude our algorithm requires less computation time than the conventional SDA.Abstract I Contents III List of Figures V List of Tables XI Chapter 1 Introduction 1 1.1 Motivation 1 1.2 Organization of Thesis of 2 Chapter 2 Spatial Multiplexing Technique 3 2.1 System Description 4 2.2 Zero-Forcing Detection 6 2.2.1 Linear Zero-Forcing Detector 6 2.2.2 Zero-Forcing BLAST 7 2.3 MMSE Detection 9 2.3.1 Linear MMSE Detector 9 2.3.2 MMSE-BLAST 11 2.4 Maximum Likelihood Detector 11 2.5 Sphere Decoding Algorithm 12 2.5.1 Sorting Enhanced SDA 16 2.5.2 modified SDA 20 2.6 Simulation Results 21 2.7 Summary 32 Chapter 3 Effects of Channel Matrix 35 3.1Indoor Channel Measurement and Analysis 36 3.1.1 Rank and Condition Number 36 3.1.2 Experiment Setups 38 3.1.3 Measurement Results 42 3.1.4 Brief Conclusion 55 3.2 Lattice Reduction aided Detectors 56 3.2.1 The LLL-reduced Algorithm 56 3.2.2 LR-aided LD and OSIC 60 3.2.3 LR-aided SDA 64 3.3 MMSE extended SDA 65 3.4 Simulation Results 66 3.5 Summary 68 Chapter 4 Proposed Modified SDA and Complexity Analysis 75 4.1 SE-SDA-inner-LR-MMSE-OSIC algorithm 75 4.2 Complexity Analysis 82 4.3 Numerical Results 87 4.4 Summary 103 Chapter 5 Conclusion 105 Reference 1071011350 bytesapplication/pdfen-US球狀解調晶格正交化排序後的漸序式干擾消除Sphere DecodingLattice ReductionOrdered Successive Interference Cancellationthesishttp://ntur.lib.ntu.edu.tw/bitstream/246246/58727/1/ntu-95-R93942042-1.pdf