Parameter Estimation For OFDM System In Impulsive Noise Environment

As a special multicarrier transmission scheme, Orthogonal Frequency Division Multiplexing(OFDM) has been widely used in military and civilian communication for its high spectrum efficiency and strong anti-multipath-fading ability. Parameter estimation of OFDM system is a popular research focus in communication field, and research on parameter estimation methods is of great academic significance and application value.In conventional OFDM signal parameter estimation methods, the Gaussian ditstribution is used as background noise model. However, since the widespread natural and man-made noise sources in real wireless communication environment such as lighting, thunder, ignition of the car and signals outside the station, the channel noise usually appears with notable impulses, and this kind of noise can be described accurately by Alpha stable distribution model. In such a non-Gaussian noise background, the performance of the conventional methods can be reduced or even disabled. Therefore, this dissertation is focused on the parameters estimation of OFDM system in Alpha stable noise environment.In OFDM system, frequency offset is often caused by the Doppler effect and the oscillator frequency mismatch between receiver and transmitter. Since OFDM is very sensitive to frequency offset, frequency offset estimation is one of the key technologies in OFDM system. To address the problem that the conventional methods degrade severely in impulsive noise environment, a frequency offset estimation method based on Myriad filter is proposed for OFDM system in Alpha stable distribution noise environment. Myriad filter, which can suppress the impulsive noise effectively, is a nonlinear filtering method in Alpha stable distribution random process. The received signal is processed by a Myriad filter to suppress the impulsive noise, then the fractional frequency offset and the integral frequency offset are estimated by combining the idea of SC algorithm, respectively. Simulation results show that the proposed method can achieve good performance in Alpha stable distribution noise.OFDM time-domain parameters estimation is of significant value in areas such as communication countermeasures, non-cooperative communication and radio spectrum monitoring, to address the problem that the conventional algorithms degrade severely in Alpha stable noise environment, a new time-domain parameter estimation method based on correntropy is proposed for OFDM system. As a generalized correlation function, correntropy is defined as a local similarity measure of a random variable. Taking advantage of the feature that the time domain structure of OFDM signals has local similarity, along with the fact that correntropy can effectively suppress the impulsive noise, the proposed method estimates the time-domain parameters of OFDM signals in Alpha-stable noise. To further improve the estimation performance in strong impulsive noise environment, a cumulative algorithm is used in this paper.