LTE（long term evolution）無線通訊系統下行採用OFDMA 系統，在此系統中使用了主要同步訊號(primary synchronization signal)與次要同步訊號(secondary synchronization signal)來處理同步與蜂巢確認的問題。然而，OFDM/ OFDMA 系統對於時間和頻率的同步誤差很敏感。此外，當用戶設備（user equipment）在蜂巢系統(cellular system)下運作，由於LTE 的蜂巢身份(cell-ID)有504 個，因此它需要快速的與所服務的基地台建立連接，且要確定所工作的區段(sector)與蜂巢身份(cell-ID)。本篇論文提出三個步驟：第一步，使用延遲自相關找出訊框時序(frame timing)與主要同步訊號，再以此時序開始做小數點時間與頻率估測，在小數點頻率估測中我們利用帄均法來改善小數點頻率的準確度。第二步，我們利用訊號相位差的方式以同步訊號來偵測區段與整數頻率偏移，實驗結果顯示可以有效抵抗通道效應與符號時序的誤差。第三步，蜂巢身份確認。此研究所提出的演算法，是利用相位差的方式來確認區段與蜂巢身份，此方法可以有效的抵抗通道效應，且與匹配濾波器法比較，此演算法能有較高的穩定度。LTE（long term evolution）wireless communication downlink system uses OFDMA. This system uses the primary synchronization signal (P-SCH) and the secondary synchronization signal(S-SCH) for synchronization and cell identities solving problems. However, it is well known that OFDM systems are sensitive to time and frequency synchronization error. Hence we require accurate synchronization techniques. Furthermore, user equipment (UE) operates in a cellular system, and there are 504 cell-ID in LTE, it needs to establish connection as fast as possible with the best serving base station, i.e. to identify the operating sector and cell. The proposed cell search procedure in this paper contains three steps: The first step, using the delay auto-correlation to find frame timing and the primary synchronization signal, and then use this sequence to start estimate the symbol timing and fractional frequency. And in the estimation we use the average method to improve the accuracy of fractional frequency. The second step, we use method of the phase difference between each signal with synchronization to confirm sector and the integer frequency. And the outcome shows that it can effectively withstand the error of channel effect and symbol timing. The third step, confirm cell-ID. The proposed algorithm confirms the sector and cell ID by phase difference, it can resist the channel effect, and it is more stabilize than matched filter.
