應用引導通道協助於寬頻分碼多重進接系統

Abstract

在分碼多重進接系統中，具有寬頻特性的展頻碼被用來區別不同使用者的窄頻訊號，當這些展頻碼無法正交就會造成多重存取干擾(MAI) 。在蜂巢式環境中，如果各使用者訊號無法適切的控制，強接收訊號使用者會嚴重弱化弱接收訊號使用者的接收機效能，也就是發生所謂的近遠效應(near-far effect)，多重使用者檢測(MUD)技術可以用來解使用者之間的相互干擾問題。在各種多重使用者檢測法中，由於具有簡單及在衰變通道環境下錯誤率表現優越的特性，序列式干擾消除法被視為一個可行的技術，在這個論文中，我們首先探討影響序列式干擾消除法效能很大的排序問題,接著提出數項改善序列式干擾消除法缺點的技術，再進一步提出增加系統容量的方法，使序列式干擾消除技術能成為寬頻分碼多工進接存取系統之上傳鏈路中的實用技術。 排序法對於序列式干擾消除法的效能有極大影響，在本論文中，我們提出一適用於寬頻分碼多工進接存取系統的排序法，此法在複雜度及效能上有適中的表現，接著比較三種排序法，包含實作的議題如重排頻率處理延遲時間、潛伏時間及計算複雜度等，還有評比與錯誤率表現有關的參數如引導符碼與使用者資訊符碼增益比、區塊檢測間隔、功率分佈比、及通道與時序估測錯誤等。除了單速率系統，我們也延伸所提架構，使之同樣適用於多速率系統。 在改善缺失方面，首先，對於通道資訊估測準確度敏感的問題，由於寬頻分碼多工進接存取系統之上傳鏈路具有與傳輸通道正交的引導通道，可以用來達成較準確到通道估測，但也由於非同步接收這些引導符碼會干擾使用者資訊的檢測，因此我們提出引導符碼消除技術來解決這個問題；對於處理延遲時間較長的問題，則提出一包括通道估測及使用者資訊檢測的管線式架構；最後，關於功率控制比較複雜的問題，我們發現，就算在一般性的功率控制法下，適當選擇排序法的序列式干擾消除法仍能優於並列式部分干擾消除法(PPIC)的效能表現。 除了考慮單純的序列式干擾消除法，我們提出一能針對變動環境及不同需求而調整的可調式架構，根據系統負載及效能需求，可調整處理延遲時間及計算複雜度，此法結合序列式干擾消除法、並列式干擾消除法及再精鍊的通道估測法，經過額外增加的計算量，此法的處理延遲時間比純粹序列式干擾消除法少，並可達到更加的效能。 最後，我們延伸序列式干擾消除的概念至有通道編碼的系統，與渦輪碼解碼同時考慮，首先根據排序資訊，提出一提早停止（early-stopping）準則以減少渦輪碼解碼時不必要的運算，與其他提早停止準則比較，在幾乎相同的檢測能力下，此準則具有快速及低計算複雜度的特性,接著提出一包含前述準則之遞迴式干擾消除法，將序列式干擾消除運用在符碼及跨級之碼塊干擾上,並且只有被檢定為錯誤的編碼區塊訊號需要再進一級的遞迴檢測，此法不僅在效能表現上更為優越，也節省了大量且不必要的計算量。

In CDMA systems, the narrowband message signals of different users are discriminated by multiplying the spreading signals with large bandwidth. The multiple access interference (MAI) is introduced when spreading signals are non-orthogonal. In the cellular environment, if the power of each user within a cell is not controlled appropriately, error performance of the user with small received power can be dramatically decreased by the user with large received power, i.e., the near-far problem occurs. A technique known as multiuser detection (MUD) can be employed to mitigate the MAI. The successive interference cancellation (SIC) is considered a promising technique among the MUDs due to its simplicity and superior error performance in fading environment. To make SIC a practical technique in uplink WCDMA systems, we first analyze the ordering method which has large influence on the performance of SIC. Then we present techniques to alleviate the drawbacks of SIC. Furthermore, techniques to increase the system capacity are proposed. It has been shown that the ordering method has a great effect on the performance of SIC. Three ordering methods are discussed and compared in the aspect of the implementation issues (such as reordering frequency, processing delay, latency, and computational complexity), and error performance related parameters (such as pilot-to-traffic amplitude ratio, cancellation-ordering method, grouping interval, received power distribution ratio and channel estimation as well as timing estimation errors). In addition to considering the single-rate system, a generalized pilot-channel aided SIC scheme is presented to apply to multirate communications. SIC has several drawbacks: sensitive to channel estimation error due to error propagation from stage to stage, longer processing delay than parallel interference cancellation (PIC), and complicated power control. In the uplink of WCDMA systems, the pilot-channel signals can be employed to reduce channel estimation errors. However, the traffic-channel signals are always interfered by other users’ pilot and traffic signals even without any fading. This interference can be alleviated by employing pilot-channel signal removal (PCSR) technique. To shorten processing delay, a pipeline scheme is proposed. It is shown in the thesis that even with the equal power control profile, the SIC with properly chosen ordering method still outperforms multistage partial PIC (PPIC). In addition to considering pure SIC in uplink WCDMA systems, an adaptable scheme with the ability of adapting its structure according to the environment and channel condition is presented. The processing delay and computational complexity can be adjusted based on system loading and required performance. The proposed scheme combines SIC and PPIC for data detection and performs refined channel estimation. The processing delay is shorter than pure SIC with reasonable hardware, and better error performance on both channel parameter estimation and user data detection are achieved. To extend the SIC technique to turbo-coded systems, an iterative IC with ordered SIC at front-end is proposed. To avoid unnecessary computation, the ordering information obtained from SIC front-end is utilized in a low-complexity stopping criterion with high efficiency. In addition to bit-wise interference cancellation, the SIC technique is also applied to code-block-wise interference cancellation. And, only the bits in incorrect blocks should be preceded to the next outer iteration. As a result, huge amount of computational complexity can be saved, and better performance is achieved.