高移動性OFDM系統中基於多項式模型及最小平均方差等化器之通道估計與資料偵測方法

Abstract

對於應用於高移動通訊環境之正交分頻多工(Orthogonal Frequency Division Multiplexing, OFDM)系統，每一個OFDM符元期間的通道時變破壞了次載波間的正交性，導致載波間干擾(Inter-carrier Interference, ICI)發生且使得效能衰降。當載具速度、載波頻率、OFDM符元期間增加，影響將更形嚴重。此也是目前產學界正努力思考如何降低ICI的研究課題。在這篇論文，我們提出一個基於多項式參數通道模型及最小平均方差(Minimum Mean Square Error, MMSE)等化器做為資料決策的方法。推導此方法是結合通道估計與資料偵測的問題，以最大可能性(Maximum Likelihood, ML)估計架構公式化，以及使用牛頓法最佳化理論，去找到一個搜尋方向向量做為更新二階多項式通道路徑參數，然後再求出通道頻率響應及利用MMSE 法則做資料決策，來達到抑制ICI效應的目的。本文接續著實驗室學長之前的研究，為顧及完整性，也對該研究中兩級通道估計方法及改進做一詳盡之描述。電腦模擬結果驗證了提出方法在高移動性射頻多路徑衰減通道的效能增進。

For an orthogonal frequency division multiplexing (OFDM) systems applied in high mobility communication environment, time-variations of multi-path channel in one OFDM symbol duration destroys the orthogonality among subcarriers, resulting in intercarrier interference (ICI) and performance degradation. The affect becomes more severe when vehicle speed, carrier frequency or OFDM symbol interval increases. In this paper, a joint channel estimation and data detection method based on polynomial parametric channel model and MMSE equalization is proposed. We derive this method by formulating the joint channel estimation and data detection problem in a maximum likelihood estimation framework and adopt Newton’s method to find a search direction vector that is used to update the second order polynomial channel parameter, then utilize the MMSE equalization to achieve the purpose of ICI suppression. Computer simulation results are provided to verify the performance of the proposed method in high-mobility radio multipath fading channels.