Precoded OFDM over underwater acoustic channels

The high peak-to-average power ratio (PAPR) and lack of frequency diversity are two well-known issues with orthogonal frequency-division multiplexing (OFDM) modulations, especially over underwater acoustic (UWA) channels. The precoding (spreading) technique is a viable way to simultaneously address the aforementioned issues, and it has long been researched in the wireless communication field. In the past decade, it has also gained increasing attention in multicarrier UWA communications. However, a comprehensive study on precoded OFDMs over UWA channels is still lacking. In this work, a unified system model accommodating linear precoding schemes is introduced for precoded OFDMs. Based on the model, a linear minimum mean square error (LMMSE) detector is developed for performance evaluation. Several typical precoding schemes are then evaluated and compared, in terms of PAPR and uncoded bit error rate (UBER), over the UWA channels measured in an at-sea experiment. This analysis showed the discrete Fourier transform (DFT) and the inverse DFT (IDFT) precoding schemes achieve the lowest PAPR. In terms of UBER, the DFT, IDFT, and discrete cosine transform (DCT) precodings achieved comparable performance that is significantly better than that of the standard OFDM over highly frequency-selective experimental channels.