A Blind CFO Estimation Algorithm for OFDMA Based on ESPRIT Algorithm

This paper proposes a blind carrier frequency offset (CFO) estimation algorithm based on estimating signal parameters via rotational invariance techniques (ESPRIT) algorithm for the interleaved orthogonal frequency division multiple access (OFDMA) uplinks. This method utilizes the special periodic structure of the received signal and the subspace rotational invariance techniques to estimate the CFO of all users. The main advantages of this method are that it obtains the CFOs of all users simultaneously within only one OFDMA block and has high accuracy due to the application of data reconstruction. Moreover, it saves frequency band effectively and degrades the computational complexity greatly. Because of estimation quickly it is suitable for the rapid change channel especially. Simulation results show that this algorithm has good performance even under low SNR condition and improves the variance performance a lot.