Abstract
As wireless communication systems demand for higher and higher transmission rate, there are more and more transmission standards adopt OFDM (Orthogonal Frequency Division Multiplexing) technique as the modulation process because it can improve the efficiency of bandwidth usage. Besides, some recent researches have proved that the channel capacity can be enhanced when transmission systems use multiple antennas, due to the increase of independent transmission paths (or called spatial diversity). It breaks the limit of the performance in the single antenna transmission systems. Recently, many researches exploring the improvements of the MIMO (Multiple Input Multiple Output) system have been conducted, not only on the architecture, but also on the specific coding algorithms in the MIMO system, especially the Space-Time Code (STC) because the STC plays an important role on maximizing the spatial diversity. According to the above description, the system combines OFDM and MIMO techniques would exhibit both high data rate and high performance properties. Hence, there are many new transmission standards which have adopted both techniques. For example, IEEE 802.11n、IEEE 802.16-2004 and IEEE 802.16e. In the MIMO-OFDM system, the channel estimation has played a critical role in determining the performances of either the channel equalizer in an OFDM or the space-time decoder in a MIMO. But, the channel estimation needs a complex circuit operation. Hence, there are some related researches which have proposed new architectures or algorithms to reduce the complexity of channel estimation. In this thesis, we propose a novel MIMO-OFDM architecture for achieving such purpose of high performance and low circuit complexity. The proposed architecture is based on the physical layer in IEEE 802.16-2004 standard and operates in FDD mode transmission. Moreover, we use the spread spectrum technique in the downlink to further reduce the complexity of the processing circuit in Customer Premises Equipment.