Channel Estimation In Filtered Multitone Modulation System Technology Research

The future mobile communication is intended to provide real-time and seamless mobile multimedia and data transmission service with high quality and high speed. But the features such as multi-path can lead to serious inter-symbol interferences and the time-variance make it difficult for the receiver to follow the channel features exactly. Because of its robustness to multi-path delay and ISI, multi-carrier modulation technique is well developed and widely used. OFDM has been adopted in several standards, and it is very likely to be adopted in the 4th generation wide band mobile communication. But OFDM has intrinsic disadvantages. For example, its orthogonality is easily to be destroyed, so it needs a lot of overheads (such as CP and virtual carrier) and strict timing synchronization is needed to ensure the performance of the system. To overcome the shortcomings, we investigate the use of a new orthogonal multi-carrier modulation technique, the method of filter banks multi-tone modulation, in wireless communication, especially in the field of the channel estimation.Filter banks multi-tone (FMT) modulation technique is a multi carrier modulation technique with filter banks. It divides the whole channel into several sub-channels with filter banks, and transmits the data information parallel with several sub-carriers. The frequency of neighboring sub-channel is not overlapped, so ICI is so lower even than other noise that we can ignore it in the FMT system.In this thesis, we first introduce the wireless channel, the research status of the wireless communication and the advantages and disadvantages of OFDM, and then we establish the channel model for wireless FMT system, and interpret the basic principles of FMT system, and make a comparison with OFDM, then concentrate on the study of FMT channel estimation method. The channel estimation will be introduced in time domain. Least Square estimation and the estimation with Spread Sequence are studied simulated with computer and compared in performance first, then the improved channel estimation method withSpread Sequence is provided. The performance comparison of these three channel estimation methods indicates that the channel estimation for FMT system in time domain is simple but effective, and the performance of the channel estimation with Spread Sequence is better than that of LS channel estimation. The improved method reaches a better performance. At last the space-time joint channel estimation method for FMT system is discussed.The aforementioned methods for channel estimation are simulated in Matlab environment, and the results are provided.