| Summary: | 碩士 === 國立臺灣大學 === 電信工程學研究所 === 96 === Recently, OFDM systems have gained its popularity and become widely used wireless communication systems. The WLAN system is the application which appears everywhere and becomes the part of our life. With the increasing demand of video, audio, and data communication, the new standard IEEE 802.11n is developed to obtain higher data rate and increase range performance. It amends the conventional WLAN systems (IEEE 802.11a/g) and involves the MIMO technique. Compared with the SISO-OFDM system, the synchronization and the channel estimation are more crucial for MIMO-OFDM system due to the additional spatial dimension. In this thesis, we introduce some basic conventional methods used for the synchronization and the channel estimation in SISO-OFDM system and discuss the extended methods in the MIMO-OFDM system.
The OFDM system converts the wideband channel to a set of parallel narrowband subchannels. As a result, it is robust against the multipath frequency selective channel. Besides, the one-tap frequency domain equalizer implemented by fast Fourier transform can be applied at the receiver end. To improve the channel estimation, the DFT-based method is used to reduce the noise influence. In practical DFT-based OFDM systems, however, a number of sub-carriers are left unmodulated, these are referred as virtual subcarriers. The leakage phenomenon induced by the virtual subcarriers limits the performance of the DFT-based channel estimation method. To solve the problem, we propose a novel channel estimation method which recovers the channel frequency response at the virtual subcarriers. The approach is slightly suffered from the noise enhancement. The simulation results show that the mean square error is close to the lower bound of the DFT-based method.
On the other hand, we establish a measurement system for MIMO-OFDM system. The setup and configuration of the transmitter (vector signal generator) and the receiver (real-time spectrum analyzer) are described in detail. Unlike the computer simulation, we use the channel emulator to emulate the practical indoor channel and take the measurement in real-time environment. This measurement system could be the basic study of the MIMO-OFDM measurement systems and used to verify an algorithm designed for the OFDM systems.
|
|---|
