Research On Channel Estimation Echnologies Based On ANFIS

Mobile communication technology has the broadest application and the fastest development as one of the modern communication technologies. The third generation communication system will be applied into commercial use soon. Next generation system which can provide higher speed larger capacity and wide band multimedia service has generated.Future communication system will achieve multimedia transmission with high quality and high speed. To achieve this goal, communication system must overcome the unfavorable effects brought by wireless channel, for instance, time-variant, frequency-selective and noisy wideband radio etc. Fast and exact data transmitting by wireless communication system need more knowledge about wireless channel. So that, channel estimation technology has become a major researching area as one of the modern communication related technologies.There are several channel estimation technologies. And adaptive channel estimation is a branch of them. Least Mean Square filtering and Kalman filtering have the most broadest application among adaptive channel estimation algorithms.In this thesis, these two adaptive channel estimation schemes for wireless communication system have been investigated, and a novel channel estimation algorithm is proposed which is based on Adaptive Neuro-Fuzzy Interference System– ANFIS. First, wireless channel's input and output data are simulated by ANFIS. After that, a new channel estimation algorithm focusing on multipath channel is proposed and simulated.Theoretical analysis and computer simulations are used in the comparisons of these algorithms, and the result shows that the proposed channel estimation algorithms have simpler implement, better performance and wider application than traditional adaptive algorithms.