在通訊領域當中，正交分頻多工系統 (Orthogonal Fre- quency Division Multiplexing, OFDM) 一 直 被 廣 為 使 用。 然 而， OFDM 系統卻有很嚴重的高峰均功率比 (Peak-to Average Power Ration, PAPR) 的問題，因此如何改善 PAPR 是一個很重要的課題。 如今已有許多技術被開發於改善 PAPR 的問題，其中部分傳輸序 列 (Partial Transmit Sequences, PTS) 也是常被使用的技術之一，由 於 PTS 採用遍佈式搜索最佳角度相位，所以計算複雜度相當高。 利用智慧型演算法 (Intelligent Algorithm, IA) 結合 PTS 是我們近年 來常使用的次佳化演算法，可以有效的降低 PAPR，並且大幅降低 計算複雜度，因此，這類型的次佳化演算法也是一個常被拿來研 究的議題。 因為智慧型演算法結合 PTS 與傳統 PTS 效能還是有些差距，本 論文根據常用的智慧型演算法做些修改，加入檢視相似度過高的 i相位因子並加以修正，可以避免會陷入區域最佳解或提早收斂的 情況，雖然會增加一些複雜度，但卻比傳統的方法更有效率的降 低訊號的 PAPR 值。;Orthogonal Frequency Division Multiplexing (OFDM) has been widely used in communication fields. One of the major drawbacks of OFDM is high peak-to-average power ratio (PAPR) for the transmitted signals. Therefore, the issue about how to remedy the PAPR in an OFDM system is important. Nowadays, many techniques have been developed to reduce the problem of high PAPR. Partial Transmit Sequence (PTS) is one of the attractive techniques to reduce the PAPR, but PTS requires an exhaustive search for all combinations of allowed phase factors such that the computational complexity is quite high. The sub-optimal algo- rithms that combine the PTS technique and intelligent algorithms have been frequently discussed in the past few years. It can effectively reduce PAPR and significantly decrease computational complexity. As a result, iiithis type of sub-optimal algorithms become a subject that draws a lot of interest. However, there is still some performance gap between the sub-optimal algorithms and the traditional PTS solutions. In this thesis, we propose a new method based on the intelligent algorithms which reduces the sim- ilarity during searching for the phase factors for PTS in order to avoid the solutions falling in local minimum. The proposed method can ef- fectively reduce PAPR though it may increase computational complexity more than the original intelligent algorithms.
