Iterative LMMSE-SIC Detector for DSE-Aware Underwater Acoustic OTFS Systems

To cope with the extremely double-dispersive channels in underwater acoustic (UWA) communication, orthogonal time frequency space (OTFS) modulation has been regarded as a potential candidate since delay-Doppler symbols can fully utilize the time-frequency diversity. So far, limited work has been devoted to UWA-OTFS systems, and the Doppler squint effect (DSE) in UWA channels is always ignored in these researches, which causes severe performance degradation and even transmission failure. Though the DSE-aware modeling has been investigated for OTFS-based mobile communications in our prior work, the assumption of system constraints cannot be satisfied in UWA-OTFS scenarios due to the significantly smaller velocity of sound, which forces us to reconsider the input-output evaluation and the corresponding system design. In this paper, UWA-OTFS systems are carefully investigated considering DSE. The intuitive characterization of the inputoutput relation for ideal waveforms is derived at first, where we also offer the design principle of system parameters. The precise closed-form representation with practical waveforms is then provided as the guideline for realistic UWA-OTFS systems. Based on the establishment of the input-output analysis, the linear minimum mean squared error (LMMSE) successive interference cancellation (SIC) detector is applied to carry out the equalization and data recovery iteratively. Simulation results demonstrate the significance of considering DSE in UWA-OTFS systems and the excellent performance of our proposed detection scheme