Blind and Semi-Blind Channel Estimation and Receiver Diversity Techniques for Multi-User OFDM Transmission Systems

正交分頻多工(Orthogonal Frequency Division Multiplexing, OFDM)為多載波調變技術之一種。此信號調變方法可以抵抗多路徑延遲擴散(multipath delay-spread)效應，因此可增加資料傳輸速率。雖然OFDM可有效對抗通道(channel)延遲擴散效應，但它仍會遭遇多路徑傳播所造成的信號衰落(signal fading)問題，以及在多用戶傳輸系統下所形成的干擾(interference)效應。針對上述兩個問題，我們可用接收器分集(receiver diversity)技術來加以解決。就接收器分集技術而言，其權重(weighting)或設計法則(metric)的適當決定，將會增加系統性能。在決定適當權重或設計法則時，發射及接收天線之間通道之估測(channel estimation)將會影響設計結果。 本論文主要針對多用戶 OFDM 系統之通道估測及接收分集技術提出新的演算法並以電腦模擬分析其效能。在多用戶通道估測方面，我們提出了子空間-最小平均平方誤差盲蔽通道估測器(subspace-MMSE blind channel estimator)，以及考慮含有領航信號(pilot)的多用戶半盲蔽通道估測演算法(multi-user semi-blind channel estimation algorithm)。前者係將前人提出之子空間法加以延伸及改善，後者結合了二種方式：一是利用領航信號及簡易多用戶通道估測法則，以求得部份通道估測值；另一是利用子空間法(subspace approach)求出初步通道估測解。結合以上結果及相關公式，我們可求出正確的多用戶之通道增益(gain)。 利用上述的通道估測法所得之結果，我們可實現分集信號組合(diversity combining)以及多用戶信號估測技術(multi-user detection)。本論文中，我們利用了MMSE線性分集信號組合方式，求出主用戶(desired user)之傳輸信號。另一方面，針對選擇式分集(selection diversity)信號組合技術，我們提出了適用於多用戶OFDM系統的最大事後機率選擇式分集(maximum a posteriori probability selection diversity)技術。在多用戶信號估測演算法設計方面，我們結合了序列式干擾源消除法(successive interference cancellation)或平行式干擾源消除法(parallel interference cancellation)及最小平方投影法(least-squares projection)來做多用戶信號估測。從電腦模擬可知，我們所提出的通道估測演算法，分集信號組合技術，以及多用戶信號估測法，均具有良好的系統效能。Orthogonal Frequency Division Multiplexing (OFDM) is a multicarrier modulation technique, which can mitigate the effect of multipath delay-spread and increase data rate. Although OFDM is very effective in combating channel delay-spread, it stills suffers from signal fading caused by multipath propagation and interference caused by other users in a multi-user scenario. Receiver diversity technique can mitigate both effects. In order to achieve the best performance, for diversity combining, the weightings or metrics that require correct channel estimates between all users and receiver antennas must be carefully chosen. In this dissertation, the receiver algorithms for multi-user OFDM with receiver diversity are proposed. For channel estimation, we propose a new two-step blind channel estimation algorithm referred to as subspace-MMSE channel estimation. For pilot-based multi-user OFDM systems, a novel semi-blind channel estimation algorithm is proposed to combine simple multi-user channel estimates obtained from pilot subcarriers with subspace-based channel estimates. Very accurate estimates for the multi-user channel gains can be obtained. Based on the estimated channel, diversity combining and multi-user detection are performed. In this dissertation, minimum mean square error linear diversity combining and selection diversity combining are both investigated. We propose a new multi-user generalization of the maximum a posteriori probability (MAP) selection diversity combining followed by successive interference cancellation with least-square projection (SIC-LSP) for multi-user detection. Instead of SIC-LSP, parallel interference cancellation with least-squares projection (PIC-LSP) is also investigated. Simulation results show that when all the proposed techniques are combined, very good performance is achievable in a multi-user OFDM system.Abstract ………………………………………………………..I Contents ………………………………………………………II List of Figures ………………………………………………...V Chapter 1 Introduction ……………………………………….1 1.1 Motivation …………………………………………………………...1 1.2 Dissertation Organization ……………………………………………2 Chapter 2 Background………………………………………..3 2.1 System Model ………………………………………………………..3 2.2 Channel Estimation for OFDM System ……………………………...4 2.3 Conventional Receiver Diversity Combining ………………………..5 2.4 Interference Cancellation for Multi-User Detection …………………8 Chapter 3 Subspace-MMSE Blind Channel Estimation for Multiuser OFDM with Receiver Diversity ….....10 3.1 Motivation ………………………………………………………….10 3.2 System Model ………………………………………………………11 3.3 Proposed Receiver Algorithm ………………………………………14 3.3.1 Two-Step Blind Channel Estimation ………………………………………14 3.3.2 Frequency-Domain Linear Diversity Combining ………………………….19 3.4 Simulation Results ………………………………………………….20 3.5 Summary ……………………………………………………………27 Appendix: Sufficient conditions for channel identifiability in multi-user generalization of the subspace approach……………………28 Chapter 4 Semi-Blind Channel Estimation for Multi-user OFDM with Receiver Diversity ………………...30 4.1 Motivation ………………………………………………………….30 4.2 System Model ………………………………………………………32 4.3 Proposed Semi-Blind Channel Estimation …………………………34 4.4 Simulation Results ………………………………………………….39 4.5 Summary ……………………………………………………………48 4.6 Appendix ……………………………………………………………49 4.6.1 Derivation of the channel estimation MSE ………………………………..49 4.6.2 Alternative weighting matrix for the proposed algorithm …………………52 4.6.3 The multi-user least-squares channel estimation with linear and quadratic interpolation ………………………………………………………………54 Chapter 5 Blind Per-Subcarrier Per-Phase Near-MAP Selection Diversity with SIC-LSP Multi-User Detection for Multi-User OFDM ……………….56 5.1 Motivation ………………………………………………………….56 5.2 Proposed Receiver Algorithm ………………………………………58 5.2.1 Approximate MAP SDC Metric …………………………………………...59 5.2.2 Successive Interference Cancellation with Least-Squares Projection (SIC-LSP) ………………………………………………………………….63 5.3 Simulation Results ………………………………………………….65 5.4 Summary ……………………………………………………………72 Appendix: The MRC/SIC-LSP scheme ………………………………...73 Chapter 6 Blind Per-Subcarrier Near-MAP Selection Diversity with PIC-LSP Multi-User Detection for multi-User OFDM ……………………………….75 6.1 Motivation ...………………………………………………………..75 6.2 Proposed Receiver Algorithm ………………………………………76 6.2.1 Approximate MAP SDC Metric …………………………………………...76 6.2.2 Parallel Interference Cancellation with Least-Squares Projection (PIC-LSP) …………………………………………………………………..80 6.3 Simulation Results ………………………………………………….82 6.4 Summary ……………………………………………………………88 Appendix: The complexity analysis for MRC and PSSDC …………….89 Chapter 7 Conclusions ………………………………………90 References ……………………………………………………9