Use of the FFT in the Sampling Time Offset Estimation for the Interpolator of IEEE 802.11a WLAN

碩士 === 中華大學 === 電機工程學系(所) === 94 === In recent years, Orthogonal Frequency Division Multiplexing (OFDM) techniques have frequently been used for wireless transmissions. It is well known that accurate timing and frequency synchronization is important for an OFDM receiver to operate properly. In this...

Full description

Bibliographic Details
Main Authors: Chung-Yi Chiang, 蔣忠易
Other Authors: In-Hang Chung
Format: Others
Language:en_US
Published: 2005
Online Access:http://ndltd.ncl.edu.tw/handle/94426820950718337992
Description
Summary:碩士 === 中華大學 === 電機工程學系(所) === 94 === In recent years, Orthogonal Frequency Division Multiplexing (OFDM) techniques have frequently been used for wireless transmissions. It is well known that accurate timing and frequency synchronization is important for an OFDM receiver to operate properly. In this paper we propose a scheme to estimate the symbol time offset and the sampling time offset. This scheme employs correlations on the short preamble, which is specified in the IEEE 802.11a WLAN standard, to estimate the sampling time offset. The estimated sampling time offset is used in the interpolator to recover the originally transmitted data. Since this scheme uses feed forward estimating approach, it is suitable for burst-mode transmissions that necessitate rapid synchronization. An important feature of this scheme is that the FFT processor in the OFDM receiver has been utilized to carry out sampling time offset estimation during its idle period. This significantly simplifies the hardware implementation complexity of the interpolator. Important performance measures such as the bit error rate and sampling time offset jitter have been simulated for both AWGN channels and a multipath channel. Numerical results show that the new scheme outperforms a referenced scheme proposed by other researchers. In addition, to further enhance the performance of this new scheme, we also study the case in which curve-fitting functions are applied. Three curve-fitting functions, linear, quadratic, and cubic functions are considered. Performance improvements resulted from applications of these functions are also illustrated via numerical examples.