[[alternative]]結合功率法則之適應性半盲蔽式通道估測器在混合式非補零展頻碼的多輸入-多輸出分碼多工系統

碩士[[abstract]]在傳統盲蔽式通道估測法中，相位模糊 (phase ambiguity)是一個潛在的問題。 目前文獻在探討其採用的盲蔽式通道估測方法時，都忽略此問題或假設通道相位的資訊在接收端為已知，實際上在接收端無法得知真實通道相位。在Chern和Cheng所發表的文獻中，他們結合時間-空間編碼 (ST-BC)與非補零 (None zero-padding)冗餘符碼的概念，設計出一組混合式展頻碼 (Hybrid padding chip-spreading codes)，以解決相位模糊的問題，並抑制符碼區塊間干擾 (Inter-block interference; IBI)。 他們所採用的時間-空間編碼架構，是假設傳送端為兩根天線，而接收端有N根接收天線。他們的目的是希望估測通道向量的第一個參數值以取得相位初始值，因此僅取得其中一根天線之通道相位初始值。本論文延伸先前文獻中提出之方法並做改進，以獲取兩根天線之相關通道向量的完整參數初始值，並利用功率法則 (Power method) 求得相關自相關矩陣 (Autocorrelation matrix)的最小特徵值所對應之特徵向量，使通道參數的估測能快速收斂。 由於是採用時間-空間編碼架構，本論文所提出的改進方法，是將文獻中原本的混合式非補零展頻碼在兩根天線之間做交替(Interleave) 使用。即在第一個時間區塊 (Time block) 採用如同文獻中之作法，以取得通道資訊 (h1) 的初始值，在第二個時間區塊，我們將原本設計插入在第一根天線之混合式非補零展頻碼向量，改置入於第二根天線，以取得另一組通道資訊 (h2) 的初始值。相較於先前文獻的作法，其所得之通道資訊初始值較以前完整。 結合功率法則所得到的通道估測能更快收斂。最後，透過從接收到的訊號移除冗餘符碼，可以進一步提高系統效能並抑制多用戶干擾及符碼間干擾。[[abstract]]The direct-sequence code-division multiple-access (DS-CDMA) is one of the primary modulation techniques, which has been widely studied in the literature. In the downlink, these systems rely on the orthogonality of the spreading codes to separate the different user signals. By exploiting the spatial diversity of multiple transmit-antenna and receive-antenna (MIMO), channel capacity of the DS-CDMA systems be increased effectively. Moreover, MIMO systems with the space-time coding (STC) can be used to counteract the channel fading effect due to multipath propagation. A space-time (ST) code is a bandwidth-efficient method that can improve the reliability of data transmission in MIMO systems. By doing so, more reliable detection can be obtained at the receiver. In additions, the time-varying nature of the channel owing to user mobility is another difficulty in multiuser detection in multipath channel environments, which introduces severed ISI (or IBI) effects. To deal with the above-mentioned problems, recently, adaptive interference suppression techniques based on the linearly constrained constant modulus (LCCM) approaches for multiuser detection have been considered as powerful methods for MIMO-CDMA systems; it outperforms the conventional detector under all power distributions. Since the adaptive LCCM filtering algorithm is a blind adaptive technique used for blind channel estimation and desired symbols recovery. It is well-known that all blind algorithms have inherent phase ambiguity problem. We present new ST block coding (ST-BC) MIMO-CDMA transceiver framework associated with the hybrid non-zero padding assisted chip-spreading codes in the transmitter. It is extended from Chern’s work that allows us to perform adaptive semi-blind channel estimation for block symbol recovery in the receiver associated with power method during channel estimation. It avoid the phase ambiguity problem and suppressing the effects of IBI as well as MAI. In the MIMO-CDMA receiver with two-branch filterbank, we build on the generalized sidelobe canceller (GSC) structure associated with the RLS based on the min/max criterion (or Capon approach) for implementing the adaptive semi-blind LCCM receiver. We will verify the merits and the advantages of the proposed scheme in this thesis and show the robustness in turns of performance improvement against inaccuracies in the acquisition of the desired user’s code/timing, referred to as the mismatch problem.[[tableofcontents]]TABLE OF CONTENTS CHAPTER 1 INTRODUCTION 1 CHAPTER 2 CONVENTIONAL BLIND CAPON RECEIVER ST-BC MIMO-CDMA SYSTEM 6 2.1 Introduction 6 2.2 Review of Space-Time Block Code (ST-BC) 6 2.3 Downlink Space-Time Block Coded MIMO-CDMA in Frequency-Flat Channels 9 2.4 MMSE Receiver and Blind Capon Receiver 13 CHAPTER 3 NEW SEMI-BLIND ADAPTIVE ST-BC MIMO-CDMA TRANSCEIVER DESIGN WITH HYBRID USER SIGNATURES 18 3.1 Introduction 18 3.2 Hybrid Non-zero Pre-coded ST-BC MIMO-CDMA Receiver in Multipath Channels 19 3.3 MIMO CM-GSC-RLS Algorithm with Capon Channel Estimation 32 3.4 Semi-blind Channel Estimation with Power Method 39 3.4.1 Method to resolve the problem of phase ambiguity 40 3.5 System Model in Time-Varying Channels 41 CHAPTER 4 SIMULATION RESULTS 44 4.1 Preliminaries 44 4.2 Simulation Results and Some Observations 45 CHAPTER 5 CONCLUSIONS 55 Appendix A Method of Selecting in MIMO CM-GSC-RLS Algorithm 57 Appendix B Power Method 59 References 60 LIST OF FIGURES Fig.2.1 System configuration of two-branch MRRC. 8 Fig.2.2 System configuration of ST-BC. 8 Fig.2.3 Configuration of MIMO-CDMA systems with ST-BC 10 Fig.3.1 CDMA systems with hybrid non-zero pre-coded (the kth user) 20 Fig.3.2 Hybrid non-zero pre-coded MIMO-CDMA Receiver in multipath channel 23 Fig 3.3 Illustration of ISI in multipath channels in time slot 2t−1 (Transmitted Symbol 26 from Tx1) 26 Fig 3.4 Illustration of ISI in multipath channels in time slot 2t (Transmitted Symbol 26 from Tx1) 26 Fig 3.5 Illustration of ISI in multipath channels in time slot 2t+1 (Transmitted Symbol 27 from Tx2) 27 Fig 3.6 Illustration of ISI in multipath channels in time slot 2t+2 (Transmitted Symbol 27 from Tx2) 27 Fig.3.7 Configuration of the MIMO CM-GSC-RLS algorithm 35 Fig.4.1 SINR comparison of different receivers without mismatch effect, L=3, N=1 49 Fig.4.2 BER comparison of different receivers without mismatch effect, L=3, N=1 49 Fig.4.3 BER comparison of different receivers without mismatch effect, L=3, N=2 50 Fig.4.4 SINR comparison of different receivers with mismatch effect, L=3, N=1 50 Fig.4.5 BER comparison of different receivers with mismatch effect, L=3, N=1 51 Fig.4.6 BER comparison of different receivers with mismatch effect, L=3, N=2 51 Fig.4.7 BER of different receivers without mismatch effect in time-varying channels, L=3, N=1 52 Fig.4.8 BER of different receivers with mismatch effect in time-varying channels, L=3, N=1 52 Fig.4.9 The number of iteration with the initial vector all unity 53 Fig.4.10 The number of iteration with the initial vector is channel information 53 Fig.4.11 Tracking ability of 54 LIST OF TABLES Table 3.1 Transmitted symbols in six time intervals and different Tx 20 Table 3.2 Summary of the CM-GSC-RLS Algorithm with Method of Selecting for MIMO-CDMA receiver 38 Table 3.3 Transmitted symbols and corresponding channels in time-varying case 43[[note]]學號: 600440266, 學年度: 10