Cross-correlation quasi-gradient Doppler estimation for underwater acoustic OFDM mobile communications

Abstract

Orthogonal Frequency Division Multiplexing (OFDM) has drawn comprehensive attention due to its capability of achieving high data rate and high spectrum efficiency in underwater acoustic communications. However, the widespread Doppler effect in underwater acoustic channel caused by mobile vehicles or time varying medium can lead to significant distortion and deteriorate the demodulation performance, thus Doppler estimation and compensation are needed for OFDM mobile communications. Generally Doppler effect can be accurately estimated by calculating the dominant correlation-peak or cross-ambiguity function (CAF) between received and transmitted signals. However, this type of method is computationally intensive because of the exhausting search on the two-dimensional (2D) delay and Doppler compression factor space, thus unsuitable for small underwater vehicles that equipped with limited computational payload. In this paper, quasi-gradient of cross-correlation is defined to derive the cross-correlation quasi-gradient (CCQG) iterative estimation algorithm to achieve low complexity Doppler estimation. Moreover, the smoothed quasi-gradient and variable Doppler interval are employed to accelerate the convergence rate as well as to avoid being trapped at a local maximum. Numerical simulations and mobile communication sea trial experiments demonstrate the effectiveness and robustness of the proposed algorithm by comparing to the conventional cross-correlation estimation, auto-correlation estimation, block estimation and BER search estimation strategies.