Please use this identifier to cite or link to this item:
標題: Low Complexity PTS Techniques Using Non-Disjoint Sub-Block Partition and Minimal Trellis in OFDM Systems
作者: 陳後守
關鍵字: 正交分頻多工系統
trellis structure
non-disjoint partition
turbo codes
摘要: The proposal of this project investigates the reduction of peak-to-average power ratio (PAPR) in orthogonal frequency division multiplexing (OFDM) systems, and also in OFDM variations such as multiple-input and multiple-output OFDM (MIMO-OFDM) systems. Interests of research and application in OFDM have grown exponentially in recent years because of its many advantages, such as high spectral efficiency and resistance to intersymbol interference and multipath fading. However, one drawback of OFDM is that the transmitted signal has a high PAPR. The proposal is a three-year long term project, using the technique of partial transmit sequences (PTS) to resolve this problem.In the first year, we will present a modified PTS algorithm, called MPTS, for PAPR reduction of M-QAM OFDM signals by partitioning an OFDM block into non-disjoint OFDM subblocks. Since an M-QAM constellation can be written as sum of several QPSK constellations, we can then apply different disjoint partitions on the QPSK OFDM blocks, which result in an equivalent non-disjoint partition on the M-QAM OFDM block. Some numerical results show that the PAPR of MPTS is much better than that of conventional PTS. The explanation of better PAPR performance, the receiver structure of MPTS, and the BER of MPTS will be our main study.In the second year, we will present a novel algorithm for PAPR reduction of OFDM signals by executing PTS in minimal trellis of block codes. The proposed method uses a linear code with good minimal trellis to select the transmitted OFDM signal with minimum PAPR and to transmit side information with error correction. The goals of combining strategy of trellis and PTS are twofold: to provide a low complexity PTS for weighting vector search and to highlight error protection capability for side information. The trade-off between the PAPR performance and the PTS complexity in trellis-PTS will be our main focus.Finally, in the third year, we will apply and modify these results, MPTS and trellis-PTS, to the MIMO-OFDM and OFDM-CDMA systems to reduce the PAPR values. Moreover, the overall PAPR and BER performance of coded OFDM systems, such as turbo-coded OFDM and LDPC-coded OFDM will also be examined.
本計劃將探討如何降低正交分頻多工系統(OFDM)、多重輸入輸出正交分頻多工系統(MIMO-OFDM)之過高峰均功率比值( PAPR)。近幾年由於正交分頻多工系統具有許多優點,例如高頻寬效益(high spectral efficiency)且可抵抗符號間干擾(intersymbol interference)及多路徑衰減(multipath fading),因此吸引很多學者在此方面之研究與應用。然而當同相位的子載波相加時會使正交分頻多工訊號有過高的峰均功率比值的缺點。因此降低正交分頻多工系統的峰均功率比值是一項熱門的研究領域。本計畫預計為三年期長程計畫,我們將探討如何改善部分傳輸序列(PTS)方法並且降低部分傳輸序列之複雜度。第一年我們將提出一種藉由將一個正交分頻多工區塊分割為數個非互斥子區塊(non disjoint subblock)來降低M-QAM正交分頻多工訊號峰均功率比值的改良型部分傳輸序列演算法,稱為MPTS。因M-QAM星座圖可寫成數個不同大小QPSK的星座圖相加。我們將兩個QPSK正交分頻多工區塊分別使用不同的互斥分割(disjoint partition)方式,即會等同於一個M-QAM正交分頻多工區塊使用非互斥的分割。結果顯示此種非互斥子區塊的分割方式比傳統部分傳輸序列有較佳的PAPR性能。第一年我們將尋求理論基礎解釋非互斥子區塊部分傳輸序列有較佳峰均功率比值,另外MPTS方法之接收端設計與位元錯誤機率的計算也是我們研究之重點。第二年我們將研究一種結合部分傳輸序列與線性區塊碼最小籬笆結構(minimal trellis)來降低正交分頻多工訊號峰均比值的新型演算法,稱為trellis-PTS。此方法利用一個擁有極佳最小籬笆結構的線性碼來選擇出具有最小峰均比值的正交分頻多工訊號,並且使此額外訊息(side information ),亦即最佳權重向量(optimum weighting vector),具有錯誤更正的能力。結合最小籬笆結構與部分傳輸序列有兩個目的:降低搜尋權重向量時所需的複雜度及提供額外訊息具有錯誤保護能力。因此第二年的研究重心將是如何達到有較低複雜度且有不錯PAPR之trellis-PTS。最後於第三年我們將改良使用於正交分頻多工系統之MPTS與trellis-PTS方法並將之應用於MIMO-OFDM與OFDM-CDMA系統藉以降低此兩系統之峰均功率比值。另外我們也將探討渦輪編碼(turbo codes)或低密度查核碼(LDPC codes)結合正交分頻多工系統後之整體系統之峰均功率比值與位元錯誤機率(BER)之效能。
其他識別: NSC98-2221-E005-037
Appears in Collections:電機工程學系所



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.