A Diversity Enhanced Particle Filter for Carrier Frequency Offset Estimation in Nonlinear OFDM System

A diversity enhanced-particle filtering (DE-PF) approach is proposed in this paper to estimate carrier frequency offset in non linear OFDM system models. The major problem in using existing particle filter (PF) for estimating carrier frequency offset (CFO) is particle impoverishment due to its present sequential importance resampling process. To solve this problem, DE-PF is proposed which uses a novel resampling algorithm to obtain a new set of resampled particles contain more state information of their adjacent particles also. Hence, the output particles can express the posterior PDF of the state better. Also, simulations indicate that the proposed DE-PF can evidently improve CFO estimation accuracy.