Carrier frequency offset estimation based on ESPRIT for OFDM direct conversion receivers

Carrier frequency offset (CFO) can cause significant performance degradation in orthogonal frequency division multiplexing (OFDM) systems. But IQ imbalance has a critical effect on the performance of conventional CFO estimation algorithms in OFDM direct conversion receivers. In this paper, a new estimation algorithm based ESPRIT is proposed to solve the imperfections in time domain. A rotational invariance signal subspace is constructed using a repetitive training preamble. And CFO estimation is obtained by generalized eigenvalue decomposition. Numerical simulations indicate that the proposed algorithm offers better results compared to existing state-of-the-art alternatives at the estimation accuracy and bit error rate (BER) performance.