Image transmission in UWA channel using CS based OTFS system

Transmission and reception of images and videos are critical for remote real-time underwater monitoring in internet of underwater things (IoUT) applications. Underwater communication is particularly challenging since it relies on highly lossy underwater acoustics (UWA) channel. High- speed mobility is difficult in UWA systems as UWA includes fast-varying channels. For channel estimation and image transmission in UWA, the performance of the conventionally used orthogonal frequency-division multiplexing (OFDM) might deteriorate dramatically if precise Doppler correction is not used. To alleviate this issue, the orthogonal time frequency space (OTFS) modulation is effective. The advantages of OTFS are two-fold, i.e., the orthogonality in time–frequency domain and clustered sparsity of channel information in the delay-Doppler domain. For efficient image transmission in the UWA channel, the integration of compressed sensing (CS) has also been proposed in this work. The two-dimensional structured Robbins Monro (RM) and the greedy orthogonal matching pursuit (OMP) CS algorithms are considered and the better between the two have been established. The proposed approach converts a time–frequency doubly fading UWA channel to a time invariant channel. Here the high-resolution picture that is to be sent is separable and orthogonal in character. All of the delay- Doppler diversity branches are grouped together. OTFS is able to achieve channel capacity with optimum performance-complexity because of the high-resolution delay-Doppler separation. Moreover, OTFS also provides robustness against Doppler and multipath channel conditions. Simulation results prove that the suggested 2D-OMP and 2D-RM algorithms offer higher MSSIM and PSNR values than the OTFS system alone.