水声通信是建设海洋信息网等“智慧海洋”国家重大战略工程的关键技术。然而，当前500-2000米的近中程水声通信理论可达速率仅在40kbps左右，而商用设备的可达速率更低，远不能满足日益增长的高速传输需求。本项目围绕波束对准难、信息交互慢、可用带宽窄等制约水声通信速率提升的瓶颈而展开研究，拟结合能灵活配置空间复用增益的MIMO技术和高频谱效率的OFDM调制技术，并创新性地融合声呐探测与人工智能构建探测协同的智能MIMO-OFDM水声通信系统，以期突破上述瓶颈。具体地，首先设计探测-通信协同机制感知空间方位信息以辅助波束对准；其次建立信道状态自主学习与预测模型以解决交互慢的难题；进而基于获知的方位和信道状态开发高谱效的波束成形算法以及智能编码与调制器以缓解带宽窄的制约；最后搭建链路级仿真系统与原型验证平台评估提出的关键理论与技术。通过上述研究，拟实现倍增近中程水声通信传输速率的研究目标。

Underwater acoustic communications are one of the most crucial techniques for carrying out the “Smart Ocean” National Major Strategic Projects such as establishing marine information networks. However, for short- and medium-scale underwater acoustic communications that spread from 1000-2000 m, the achievable data rate is theoretically only about 40 kbps currently and is much lower for commercial devices, and thus could hardly meet the ever-increasing demands in high-data-rate underwater transmission. Among the key bottlenecks that restrict the data rate improvement, three most significant are difficult beam alignment, slow information exchange, and narrow available bandwidth. Aiming at overcoming these three bottlenecks, this project intends to combine the multi-input multi-output (MIMO) technology with flexible spatial multiplexing gain and the orthogonal frequency-division multiplexing (OFDM) modulation technology with high spectral efficiency, and then innovatively integrates sonar detection and artificial intelligence to construct a detection-cooperative and intelligent MIMO-OFDM underwater acoustic communication system. Specifically, we first develop detection-communication-cooperative positioning schemes to assist the alignment of acoustic beams for directional data transmission. We then construct intelligent models for self-learning and predicting channel state information (CSI) to address the challenge of slow information exchange. Based on the obtained position and CSI, we further develop spectral-efficient and intelligent beamforming and coding-modulation schemes to mitigate the constraint from narrow bandwidth. Finally, we establish a link-level simulation system and prototype verification platform to evaluate the proposed key theories and techniques. In general, by this project, we devote to reaching the goal of multiplying the data rate achieved by current short- and medium-scale underwater acoustic communications.