分碼多重進接碼擷取及適應性解展頻演算法之研究

Abstract

人為干擾與窄頻干擾為影響多重進接碼擷取和解展頻性能的兩個主要因素。碼擷取是基頻多重進接接受器的前端，對於資料偵測至關重要。因此如何設計一個在人為或窄頻干擾的環境中也能有效運作之時序估計器便成為重要的課題。在本論文中，吾人將介紹幾種傳統的時序估計器及其原理，並將介紹戈德碼的特性。吾人將引入一種能夠在人為干擾的環境中達到碼擷取的適應性時序估計器。經由電腦模擬，吾人比較了傳統時序估計方法和適應性方法的效能。由結果得知，適應性方法在人為干擾的環境下能有較好的效能。為了在實際的狀況下能有效運作，吾人提出了一種碼擷取重新的方法，在沒有訓練訊號的情況下仍能有效提供時序估計。最後，吾人考慮使用部分適應性相關器來達到資料偵測的目的。經過電腦模擬可證實，部分適應性相關器在適當的複雜度之下，可以成功的壓抑少數的強人為干擾。

Jamming and narrowband interference (NBI) are the factors which adversely affect the performance of code acquisition and data detection in a CDMA system. Code acquisition is the front-end of a baseband CDMA receiver, and is crucial for data detection. The design of a timing estimator that can work in an environment contaminated with jamming or interference is thus a subject of importance. In this thesis, we first review some conventional timing estimators and their working principles. The properties of the Gold codes are investigated. We then introduce an adaptive timing estimator for reliable code acquisition in the presence of jamming. Through a set of computer simulations, we compare the performance of the correlator-based estimator and the adaptive estimator. It is shown that the adaptive estimator significantly outperforms the correlator-based estimator in dealing with the jammers. For practical operations, a re-acquisition method is proposed which provides effective timing estimate without the aid of a training signal. Finally, we consider the realization of a partially adaptive (PA) correlator for data detection. Through computer simulations, it is shown that the PA correlator is successful in suppressing a few strong jammers with a moderate complexity.