Research On Carrier Frequency Offset Estimation Technology In OFDM System

OFDM (Orthogonal Frequency Division Multiplexing) technology has been widely used in wireless communication areas and defined not only the core of the fourth generation mobile communication technology but also one of the fundamental techniques of wireless broadband access system in future. OFDM system is very sensitive to carrier frequency offset and the system performance can be seriously degraded even by slight frequency offset. Carrier frequency offset is one of the key problems to be solved in OFDM technology application and has become a hot research topic recently. The paper focuses on research of accurate and efficient estimate algorithms of carrier frequency offset.The paper studies several typical estimate algorithms of carrier frequency offset currently and compares their performance based on simulation results. The paper mainly involves works as follow:(1) Introduction of fundamental principles, structure and system model, characteristics of OFDM system. Detailed analysis of DFT implement methods, series-parallel transformation, guard intervals and cyclic prefix function.(2) Analysis of symbol timing offset, sampling timing offset and carrier frequency offset of OFDM system, with the focal point of their influence on OFDM system. There are two main jeopardizes of carrier frequency offset. One is degradation of signal amplitude and the other is ICI (Inter-Carrier Interference) due to adjacent subcarriers.(3)Analysis of one frequency offset estimate algorithm that based on cyclic prefix and another two based on virtual subcarriers. Similar MUSIC and similar ESPRIT algorithms are analyzed and compared after illustration of principal thoughts of the algorithms. Simulation result shows that both of the two methods can realize blind estimate of carrier frequency offset. Similar MUSIC algorithm can achieve relative high estimate accuracy and its performance exceeds that of similar ESPRIT algorithm. The performance of similar MUSIC algorithm is largely influenced by step-length due to its adoption of ergodic method. Besides, similar MUSIC algorithm has relatively large calculation volume and is not suitable for mobile environment requirement in future.