一種用於分碼複接無線通訊系統之改良式擬似亂碼同步方法研究

Abstract

在直接序列/分碼複接系統中,唯有在辨識用擬似亂碼已完成同步之後才能 正確無誤地檢測傳輸資料。這項辨識碼同步的工作通常分成二個步驟來進 行：第一個步驟稱為擷取 (Acquisition) 負責達成本地產生的與接收到 隱含在雜訊中的辨識碼粗略同步；當擷取完成後，第二個步驟稱為追蹤 (Tracking),它緊接著利用一些迴授電路來完成更精確的同步。本論文主 要由搜尋的觀點提出一種二級式利用序列機率測試 (SPRT)的方法改良擷 取的速度。在可加成性白色高斯雜訊及多個其他使用者干擾環境下，測試 平均擷取時間和錯誤機率的關係。並且由硬體和工作效能的觀點比較用於 追蹤的差時鎖定迴路(DLL),分別討論個別的優缺點。 In DS/CDMA systems, the extraction of information from the received signal passing through noisy environments requires the synchronization of a pseudonoise (PN) code at the receiver. The process of synchronization is usually accomplished in two steps. The first step, called acquisition, is to bring the local replic of PN code in the receiver into coarse alignment with the PN code buried in the incoming waveform. Once the acquisition is achieved, the second step, called tracking, takes over and continuously maintains fine alignment of PN code using a feed- back loop. In this thesis, we propose a two-stage SPRT-based acqui- sition scheme which is more rapid than the traditional simple SPRT-based acquisiiton scheme from the search viewpoint. The relation between mean acquisition time and error (false alarm and missing) probability is examined on the AWGN channel where other nine equal-power users are treated as interferences. In addition, we study the tracking loops and compare two digital DLL schemes from the viewpoint of trade-off in the performance and hardware complexity.