Research On Channel Estimation Based On Training Sequence In TD-SCDMA System

TD-SCDMA proposed by China is one of the 3G standards with independent intellectual property rights. It adopts time division duplex mode. It can be allocated bandwidth flexible and can adjust time slot base on different service types for uplink and downlink which increases bandwidth efficiency. It exploits many key technologies such as smart antenna, joint detection and uplink synchronization to combat interference in the channel and improves the system performance effectively. However, these key technologies need exact channel impulse response. Therefore, an exact and quick algorithm for channel estimation is very important for performance of TD-SCDMA system.This dissertation mainly researches on algorithm for channel estimation based on training sequence in TD-SCDMA system. Firstly, the dissertation introduces the history of mobile communication, the third generation mobile communication system, TD-SCDMA system, significance of channel estimation, the progress of research for channel estimation in recent years, the characteristic and model of mobile communication channel in research background. Then, the dissertation introduces the channel and physical layer structure of TD-SCDMA system, especially training sequence and its correlativity. And then, according to theoretical derivation, the dissertation introduces the system model of channel estimation based on training sequence in TD-SCDMA system. On this basis, the dissertation introduces the basic channel estimation algorithm known as Steiner channel estimation and several other common algorithms for additional processing with threshold. Advantage and disadvantage for each algorithm are also shown. Because the conventional channel estimation algorithms can't solve the problem of amplitude increased cased by noise, the dissertation propose an improved algorithms for additional processing with threshold to adjust the amplitude of channel impulse response. And then, each algorithm is simulated on the platform of link-level simulation for TD-SCDMA. The performance for each algorithm is compared and analyzed by the simulation results. Simulation results show that compared other algorithm, the improved algorithm proposed by this dissertation can improve the system performance to a certain extent. In the end, a summary to the whole dissertation and prospects for future work is shown.