高速的數位通訊因為它可以改善頻帶上暫時性的強烈突波和窄頻干擾，所以運用是非常廣泛，而其中又以正交分頻多工(OFDM)技術更為。 正交分頻多工(OFDM)技術因為可以改善頻帶上暫時性的強烈突波和窄頻干擾，所以在高速的數位通訊上運用是非常廣泛。而OFDM最主要的優點是可以增強基本信號的能量進而克服通道上的衰減。 在數位通訊中，通道上的信號是由調變過的連續二進制信號所發送出去。而接收端則是經由取樣和量化之後，所解調變過的離散衰減信號。所以在接收端中最主要的關鍵就是讓時間同步，這樣才會使得接收端的接收效果達到最好。而時間同步所指的就是，在進來的資料信號取樣時間必須同步。 非同步的取樣信號是這個研究的主要範本。一般來說，接收端是使用固定的取樣頻率，但是由傳送端和接收端的資料取樣時間是不一樣的,這樣會導致非同步的情形發生。所以，在取樣後就必須做內插技術的補償。而內插技術通常是使用Lagrange的內插法，它的方式是改變有限脈衝頻率響應濾波器的係數，進而可以自動調整取樣後的資料。 這篇論文是以 Farrow 的架構來實現 cubic 和 quintic 兩種內插法的濾波器。在從Farrow 的架構來發展一個新的架構，它和傳統的 cubic Farrow 架構來做比較，在硬體方面是減少23%。 OFDM technique has been widely implemented in high-speed digital communications to increase the robustness against frequency selective fading or narrowband interface. The major advantage of OFDM is the ability to enhance the basic signal using approaches that can overcome channel impairments. In digital communication, binary information is converted by means of a modulator into a continuous-time signal which is sent over the transmission channel. A digital receiver is to extract the information sequence from a discrete signal obtained after sampling and quantizing the distorted signal presented to the demodulator. At the receiver, accurate timing recovery is critical to obtained performance close to that of the optimal receiver. Timing in a data receiver must be synchronized to the symbols of the incoming data signals. This study considers a non-synchronized sampling scheme. The received signal is performed by a fixed sampling clock; the samples are not synchronized to the incoming data symbols. Timing adjustment is done after sampling using interpolation. Farrow structure has been commonly used to efficiently implement the Lagrange interpolation for timing adjustment. This thesis presents the efficient implementation of the Farrow structure for cubic and quintic interpolations. Results show that the developed cubic Farrow structure achieves a hardware cost reduction by 23% from the conventional one. The design concept can be readily extended to the Farrow structures for higher order interpolations.
