We propose the use of compressed sensing for an underwater acoustic (UWA) communication system with orthogonal signal division multiplexing (OSDM) to achieve reliable communication in UWA channels with large delay and Doppler spreads. However, OSDM does not fully exploit the characteristics of UWA channels, which have long but sparse reverberation tails. Hence, we propose an OSDM channel estimator using compressed sensing and evaluate its performance in simulations and experiments. The simulations revealed that the optimal gain determining the performance of the channel estimator depends on standard deviation of noise σ, and OSDM using compressed sensing outperforms normal OSDM. We also evaluate OSDM performance using compressed sensing in field testing in the coastal area of Suruga Bay, Japan. The experimental results suggest that compressed sensing can boost OSDM performance; the number of OSDM error blocks using compressed sensing is half that in normal OSDM. Thus, OSDM using compressed sensing provides reliable communication.

Underwater acoustic (UWA) communication is an essential technology that supports underwater surveys by transmitting images and movies wirelessly. However, UWA communication is still challenging due to the large delay and Doppler spreads in UWA channels. Doppler-resilient orthogonal signal division multiplexing (D-OSDM) under water has been found effective in such channels, but Doppler modeling errors prevent further improvement in communication quality. In this paper, we clarify the effect of Doppler modeling errors on the communication quality of D-OSDM and propose the use of a window function in D-OSDM to address the issues that such errors raise. We also evaluate the communication quality of D-OSDM using the window function through simulations and experiments. The experimental results show a 56% increase in the number of error-free data blocks, indicating that windowing can improve communication quality.

Underwater acoustic (UWA) communication is a key technology providing wireless communication, but using UWA channels is challenging due to the existence of large Doppler and delay spreads. UWA channels also have long but sparse reverberation tails. To address these issues simultaneously, we have proposed Doppler-resilient orthogonal signal division multiplexing using compressed sensing (D-OSDM-CS). The optimization of compressed sensing (CS) under a large Doppler spread remains difficult since noise induced by inter-carrier interference (ICI) is present in the received signal. Herein, we mathematically clarify the effect of ICI-induced noise on the received signal, and we propose an advanced D-OSDM-CS method that can measure ICI-induced noise and optimize the regularization parameter of CS. We also evaluate the performance of the advanced D-OSDM-CS in simulations and experiments. The obtained results suggest that the advanced D-OSDM-CS achieves better communication quality than the existing D-OSDM-CS under a large-Doppler-spread environment.