Adaptive Beamforming for an Underwater MIMO-OFDM Acoustic Communication System Using Basis Expansion Model

In recent years, Underwater Acoustic Communications (UAC) has been a great matter of consideration because of its importance in different areas such as commercial and military applications. Underwater acoustic communications channel is known as a time-varying and doubly selective channel in both time and frequency domains. The orthogonal frequency division multiplexing (OFDM) modulation is an effective technique to communicate over challenging acoustic channels. In addition, using multiple-input multiple-output (MIMO) systems increases channel capacity which results in high data rate communications. Recently, basis expansion models (BEMs) have been widely used to estimate an underwater acoustic channel. In particular, when the channel is time-varying, the BEM model can effectively estimates the channel with a reduced number of coefficients and low computational complexity. To improve the performance of a MIMO communication channel, various beamforming techniques have been proposed in different areas. Inspired by the basis expansion modeling of an underwater acoustic channel, in this paper we develop a BEM based adaptive space-time beamforming for both the transmitter and receiver of an UAC. The Laguerre basis expansion model is employed in the linearly constrained minimum variance (LCMV) beamformer to obtain an adaptive scheme for updating the beamforming weights at the transmitter and receiver and to optimize the system performance in real-time. Our Simulation results show that the proposed BEM based beamformer method improves the Bit-Error-Rate (BER) and Minimum-Square-Error (MSE) performance substantially for a Rayleigh fading underwater acoustic channel. In particular, our method improves the BER and MSE about 10dB and 4dB compared to the discrete prolate spheroidal sequence (DPSS) method.