寬帶通信系統之序列設計

Abstract

在寬頻通訊系統中，多輸入多輸出技術已經被證明可以提供高速以及增進頻帶效率的重要技術。寬頻通訊系統標準例如IEEE 802.16e、3GPP LTE和LTE Advanced都採用多天線技術。而這些先進的通訊系統十分依賴適當設計後的訊號，來獲得正確的通道狀態並且執行精確的同步。但因傳送這些非資料訊號，例如像保護位元和指標，使得無線通訊系統會蒙受頻率效率的降低。為了達到多輸入多輸出系統所保證的資料傳輸率增益，準確的空間相關訊息是非常重要的。此外，使用正交分頻多工技術雖然有許多益處像是簡單的等化器設計，但高的峰值因數往往會發生，因而降低了發射機的功率效率。當多輸入多輸出系統和正交分頻多工系統結合時，由於越來越多的傳送端，高峰質因數更容易發生並嚴重地對系統造成影響。

在這篇論文裡頭，利用仿射編碼來設計耐用(robust)疊加訓練序列，即使只有不確定性存在的空間相關訊息，此序列依舊可以幫助預測空間相關的多輸入多輸出通道狀態。序列是直接加到資料訊號上，不會造成頻率效率的損失。此設計不需要準確的空間相關訊息（矩陣），在這篇論文中，我們也證明此耐用設計優於之前被提出用來預估空間相關多通道的方法，例如RMMSE和LS-RMMSE。這個耐用設計可以用投影凸集的迭代算法來解決，只要訓練序列的初始化是滿秩的，此迭代保證會收斂。當通道是不空間相關時，我們證明出耐用設計和RMMSE是漸近相同的。除此之外，我們也提出一個功率分配方法來達到最佳的資料檢測效能。

由於使用多輸入多輸出正交多頻分工系統，我們提出了疊加序列來執行耐用通道預測並且來降低峰值因數。我們提出了載波仿射編碼來降低高峰質因數，而接收端不需要知道任何的額外訊息來解調資料訊號。相對於之前已知的技術，我們提出的方法降低了很多的額外傳輸開銷，也增進資料檢測效能。即使我們提出的設計也要傳送額外的訊號，但此多餘的訊號可以小到每個載波只要一個符號。此設計可以讓設計者很容易的權衡資料檢測和降低峰值因數的效能。在模擬結果中，我們的方法明顯在降低峰值因數和傳輸效率上都優於Tone Reservation。

MIMO technology has proven to be the key enabler of high-speed, bandwidth efficient broadband communication systems such as IEEE 802.16e, 3GPP LTE and LTE-Advanced. Similar to many traditional systems, these advanced communication systems rely heavily on proper signaling in order to obtain correct channel state information and perform precise synchronization. Unfortunately, traditional signaling methods can incur a loss of spectral efficiency due to transmission of overhead data such as preamble, guard bits and pilots. Moreover, accurate spatial correlation information is crucial in achieving the theoretical capacity gain promised by MIMO. Furthermore, with the use of OFDM, high PAPR is often incurred, thus lowering the power efficiency at the transmitter. The problem is worsened when OFDM is combined with MIMO as more RF chains are required for transmission.

In this thesis, a new signaling scheme is proposed for spatially correlated MIMO channels which exploits affine precoding to produce robust superimposed training sequence such that CSI can be accurately obtained even when uncertainty in the spatial correlation matrix exists. The sequence is algebraically added to the data such that there is no loss of spectral efficiency. The proposed scheme does not require accurate knowledge about the spatial correlation matrix and it is shown to outperform previously proposed robust correlated MIMO channel estimators such as relaxed MMSE (RMMSE) and least-squares-RMMSE (LS-RMMSE). A solution for the sequence can be obtained easily by using a projection on convex sets based iterative algorithm which is guaranteed to converge as long as the training sequence matrix is initialized to have full rank. Furthermore, it is shown that the proposed scheme is asymptotically identical to the RMMSE based schemes when the MIMO channel is spatially uncorrelated. A power allocation scheme is also proposed that can maximize the detection performance.

Next, a joint superimposed sequence design is proposed to jointly perform robust channel estimation and lower the PAPR of MIMO-OFDM systems. A per-tone affine precoding technique is proposed to reduce the PAPR such that no side information is required to be transmitted for the removal of the sequence at the receiver. This is in contrast to previous known techniques which incurs a large amount of transmission overhead, or can dramatically increase the BER. Furthermore, some of these techniques are based on heuristics that cannot optimally lower the PAPR. Even though redundant information has to be sent, this can be as small as 1 symbol/subcarrier. Furthermore, the proposed design allows the designer to easily trade off between BER and PAPR reduction performance. Simulation results have shown that the proposed scheme outperforms the tone reservation scheme not only in PAPR reduction but also in transmit efficiency.