Millimeter-Wave OSDM Underwater Acoustic Communication With Large-Scale Array

Abstract

Millimeter-wave (mmWave) communication is a newly-emerging technique for terrestrial 5G wireless systems. By shifting to the mmWave frequency band, a much larger bandwidth can be achieved. The resulting high-frequency transmission loss can be mitigated with the aid of a large-scale array (LSA) and angular processing. A similar scheme is also promising for underwater acoustic (UWA) communication. Motivated by this, we introduce orthogonal signal-division multiplexing (OSDM), a generalized modulation scheme that improves bandwidth efficiency and flexibility over orthogonal frequency-division multiplexing (OFDM) and single-carrier block transmission (SCBT), and design a mmWave LSA-based OSDM (LSA-OSDM) system in this paper. However, the LSA-OSDM system suffers from inter-vector interference (IVI) over doubly-selective UWA channels. To counteract this effect, zero-vector-based Doppler compensation is performed at the receiver, followed by matrix factorization to diagonalize the composite channel matrix, enabling linear-complexity per-vector equalization. Moreover, because of the angular sparsity of mmWave scenarios, a structure-aware orthogonal matching pursuit (OMP)-based channel estimation algorithm is proposed, which can ease the computational burden by jointly exploiting the Kronecker structure of the dictionary matrix and vectorization-Kronecker identity. Simulation results indicate that the proposed LSA-OSDM system outperforms its OFDM counterpart, and the proposed sparse channel estimation algorithm achieves higher accuracy than the traditional approach.