Study on channel estimation technology in OFDM system

Abstract

The technique of Orthogonal Frequency Division Multiplexing (OFDM) channel estimation is an important topic being studied in the fourth generation mobile communications. It is significant in increasing the capacity of the system and the efficiency of the frequency spectrum. The traditional channel estimation algorithms have the disadvantages of performance worsening and they are all hard to obtain good performance in the transmission efficiency and operational precision, so a new method based on insertion pilot method and Regression Least Square Support Vector Machine (RLS-SVM) algorithm is proposed. Firstly, the frequency responses of the positions of pilot frequencies and the frequency responses between the carriers are obtained, then the frequency responses of the positions of the subsequent sub-carriers are predicted by applying RLS-SVM algorithm. The simulation result shows that increasing with SNR, performance of this method is better than all the traditional methods when applied to OFDM system. At the same transmission rate, the number of pilot frequencies in RLS-SVM algorithm is less than that in traditional algorithms, so the transmission efficiency of the system is improved. In a word, the performance of RLS-SVM is far and away better than that of traditional methods. © 2009 IEEE.