Multi-User CFOs Estimation for SC-FDMA System Over Frequency Selective Fading Channels

Frequency synchronization in single-carrier frequency division multiple access (SC-FDMA) uplink system is a challenging task due to the presence of different carrier frequency offsets (CFOs) for different users. In this paper, we propose a blind CFOs estimation algorithm for SC-FDMA uplink system by oversampling method in the presence of frequency selective fading channel. A cost function is derived which minimizes the power of the off-diagonal elements of a signal covariance matrix while estimating the correct CFOs. The off-diagonal elements are nothing but inter-carrier interference and multiple-access interference introduced in the presence of multiple CFOs. A complete mathematical model has been presented for multiple CFOs estimation under multiple access scenarios. The proposed CFOs estimation method does not require channel-state information which results in lower computational complexity. The higher iteration complexity of grid search algorithm has been further reduced by a deterministic approach. We also derive the Cramer-Rao bound for CFO estimation. The simulation results show that the proposed CFOs estimation method outperforms the existing subspace theory-based methods, specifically in the low signal-to-noise ratio region.