正交分頻多工行動通訊系統之同頻干擾消除

Abstract

本論文探討行動通訊系統的細胞邊緣使用者之同頻干擾消除。我們首先提出一基於領航訊號之同時估計所需以及干擾通道的方法。假設領航訊號為已知，且從所需和干擾基地台發送出來時是被設置在不同的子載波位置上，在這種情況之下我們所提的方法即使在有強烈的干擾的情況下仍然能夠同時估計出所需和干擾的通道，我們主要是應用壓縮取樣的方法，如匹配搜尋(matching pursuit，MP)、正交匹配搜尋(orthogonal matching pursuit, OMP) 以及子空間搜尋(subspace pursuit，SP)等演算法作通道估計。我們接者結合決策迴提出一個遞迴的的方法，以消除在領航子載波上的干擾訊號。此方法不但可以提升通道估計的精準度，且能減少在領航子載波的額外增加的功率。通道估計完成號，我們便能應用標準的多輸入單輸出(multi-input-single-output, MISO)或多輸入-多輸出(multi-input-multi-output, MIMO)偵測器來解出所傳送的訊號。模擬結果顯示我們所提出的方法可以很精準的估計通道，干擾問題也可以有效的消除。

This thesis considers co-channel interference mitigation for cell edge users in reuse-1 cellular OFDM systems. We first propose a pilot-based channel estimation method for the desired and interfering channels. Assuming that the pilots from the desired and interfering BSs are known and do not allocate in the same subcarriers, the proposed methods can conduct the channel estimation for both desired and interfering channels, even in a strong co-channel interference (CCI) environment. The main idea is to apply compressive sampling (CS) methods such as matching pursuit (MP), orthogonal matching pursuit (OMP) and subspace pursuit (SP) algorithms to do the job. We then propose an iteration method in which decision-feedback is taken into play to suppress the data symbols interfering the pilot symbols. With the method, the estimation performance can be further enhanced and the power boosting requirement for the pilot subcarriers can be reduced. Once the channels are estimated, standard multi-input-single-output (MISO) or multi-input-multi-output (MIMO) detections can be applied to detect the desired signal. Simulation results show that the proposed algorithm can estimate the channels accurately and the interference problem can be mitigated effectively.