Blind and efficient sub-space based carrier offset estimator for multi-antenna OFDM communications in correlated noise

Orthogonal frequency division multiplexing (OFDM) communications transforms frequency selective channels into multiple low rate subchannels robust against frequency selective fading. Carrier frequency offset between transmitter and receiver local oscillators must be estimated and compensated at the receiver to maintain orthogonality of these subchannels. In many practical applications, this problem is complicated by signals corrupted by colored noise due to interferer and diversity combining. Multi-antenna receiver diversity may be susceptible to correlation across receiver branches. The nonlinear least squares (NLS) estimator proposed by the authors in an earlier work exploits diversity for carrier frequency synchronization and is shown here robust to colored noise. Furthermore, the estimator mean square error (MSE) is asymptotically efficient attaining performance corresponding to the Cramer-Rao bound (CRB) for large enough sample. Numerical studies are presented to verify our results.