OFDM系統之低複雜度時序與載波頻率偏移估計

Abstract

眾所週知，正交分頻多工(Orthogonal Frequency Division Multiplexing，OFDM)對同步誤差非常敏感，同步誤差包括符元時序偏移(symbol timing offset，STO) 和載波頻率偏移 (carrier frequency offset，CFO)。這些誤差會破壞子載波間的正交性，造成子載波間的干擾(Inter Carrier Interference，ICI)和符元間的干擾(Inter Symbol Interference，ISI)。因此對接收端來說STO和CFO估計是相當重要的議題。傳統ML方法假設通道是加成性白高斯雜訊(additive white Gaussian noise，AWGN)，因此在多路徑通道中效果不佳。最近，一個基於新時序函數(new timing function，NTF)的方法被提出來解決同步誤差的問題，NTF方法具有良好的性能，但是其計算複雜度非常高。在本論文中，我們提出了一種新的方法來克服這個困難，我們的方法是先估計通道長度，使得在NTF方法中所進行的二維搜索可以被減少到一維，因此可以顯著降低其計算複雜度。由模擬結果可知，我們的STO和CFO估計的性能幾乎與NTF方法的性能相同，然而，計算複雜度可以降低到百倍的等級。

It has been known that OFDM is sensitive to synchronization errors, i.e., symbol timing offset (STO) and carrier frequency offset (CFO). These errors will destroy orthogonality between subcarriers causing inter carrier interference (ICI) and inter symbol interference (ISI). Hence, the estimation of STO and CFO is critical for OFDM receivers. Traditional ML method, which assume that the channel is an additive white Gaussian noise, perform poorly in the multi-path channels. Recently, a method based on a new timing function (NTF) is proposed to solve the problem. The NTF method is shown to have good performance, but its computational complexity can be very high. In this thesis, we propose a new method to overcome this difficulty. Our method estimates the channel length such that the two-dimensional search conducted in the NTF method can be reduced to one-dimension, significantly reducing its computational complexity. Simulation results show that the performance of our STO and CFO estimation is almost the same as that of the NTF method. However, the computational complexity can be reduced by two order.