802.16e Based OFDMA Physical Layer Simulation Results And Estimation For Slow-Fading Time-Varing Channels

OFDM A broadband wireless mobile system based IEEE802.16e can meet the need of high data-rate and mobile transmission in future wireless communications, which is expected to support high-speed multimedia services applications for mobile stations moving at vehicular speeds. Most work on IEEE802.16e OFDMA physical layer focused on local independent functional modules of the system, not on the performance of the overall system. Moreover, in high-speed mobile environment, the slow fading channels will change slowly over time, and no longer satisfy the quasi-static assumptions. Therefore, the conventional slow fading channel estimation algorithms based on the quasi-static assumptions are not applicable to such channels any longer. Aiming at the above problems mentioned, a systematic simulation and study of IEEE802.16e OFDMA physical layer are presented. Using the Grey Modeling theory, a novel estimation scheme for slow fading time-varying OFDM channels is proposed. The major contributions of this thesis are summarized below:1. The performance of OFDMA physical layer over two categories of independent fading channels is investigated. SUI channel model is utilized to evaluate the performance of OFDMA systems in the fixed wireless channel, including modulation, channel coding and interleaving and so on. On the other hand, M.1225 channel model is also adopted to evaluated the performance of the basic modules. The simulation results show that no matter which channel model is chosen, the low-modulation can always achieve a better performance in high SNR region. Moreover, we conclude that coding and interleaving techniques can also considerably improve the performance of the system.2. A novel grey modeling based estimation for narrow slow fading time-varying OFDM channels in mobile environments is proposed. By establishing the grey model, the channel state information of the previous time-slot is utilized to dynamically predict the unknown channel state information of the subsequent time-slot. The feasibility of grey modeling applied for estimating the channel characterized by 2-D time-frequency variables is verified by both theoretical analysis and simulations. Compared with conventional channel estimation approaches, the proposed algorithm not only performs well in narrowband slow fading time-varying channels, but also has low complexity and ease of realization.