Collaborative Research NeTS-NOSS: SEA-Swarm: A Rapidly Deployable Underwater Sensor Network

该项目旨在定义、优化和实现一种名为SEA-Swarm （sensor Equipped Aquatic Swarm）的新型水下传感器网络架构，该架构由大量低成本的水下传感器组成，这些传感器随水流和分散作为一个群体（群）运行和移动。拟议的SEA-Swarm架构将开启一个全新的水环境观测、监测和勘探时代。这种工具就其本质而言将是多学科的，预计将促进网络和通讯方面的研究人员与环境科学、海洋生物学和海岸监测与安全等其他科学界之间的广泛合作。本项目将解决一些技术难题，以及相关的基础研究方面：基于OFDM设计新一代水声通信调制解调器，通过多普勒补偿和MIMO时空信号处理技术实现前所未有的数据速率。在信息论中继信道驱动下，协同通信协议与一种新型节能地理路由算法协同优化。实际上，从源到目的地的虚拟管道中的所有节点都协作以可靠地传递消息。为了实现可靠的数据传输，研究了用于擦除校正的网络编码和用于能效的分组合并。最后，提出了基于水下GPS和具有专用数据采集功能的节点（数据骡子）的高效定位方案，以解决网络群体特性带来的移动性和拓扑随机性问题。通过专用仿真环境（Aqua-Sim）验证了涉及上述各方面的跨层设计。南加州大学将开发一个具有成本效益的水声通信试验台，以提供基本网络算法的概念证明。最后，一个水下测试平台的概念验证计划将与其他现有项目（如康涅狄格大学的MyPond和MySound以及加州大学洛杉矶分校的Marina水上中心）协同进行。所有三个合作团队的理论和实验工作都将涉及研究生，并让本科生接触到在上述测试平台框架下开发的最先进的通信工程项目。

This project aim at defining, optimizing and implementing a novel underwater sensor network architecture, called SEA-Swarm (Sensor Equipped Aquatic Swarm), that consists of a large number of low cost underwater sensors that operate and move as a group (swarm) with water current and dispersion. The proposed SEA-Swarm architecture will enable a whole new era of observations, monitoring and explorations in the aqueous environment. Such tool will be by its very essence multidisciplinary and is expected to foster a broad collaboration between researchers in networking and communications and other scientific communities, such as environmental sciences, marine biology and coastal surveillance and security. A number of technological challenges shall be addressed by this project, along with the related fundamental research aspects: a new generation of underwater acoustic communications modems is to be designed based on OFDM, achieving unprecedented data rates thanks to Doppler compensation and MIMO space-time signal processing techniques. Cooperative communication protocols, driven by the information theoretic relay channel, are synergistically optimized jointly with a new energy efficient geo-routing algorithm. In essence, all the nodes in a virtual pipe from source to destination cooperate to reliably deliver the message. For reliable data transport, network coding for erasure correction and packet combining for energy efficiency is investigated. Finally, efficient localization schemes based on underwater GPS and nodes with dedicated data-collection functions (data mules) are advocated in order to tackle the mobility and the topology randomness problems due to the swarm nature of the network. The cross-layer design involving the above aspects is validated through a dedicated simulation environment (Aqua-Sim). A cost-effective over-the-water acoustic communication testbed shall be developed by USC in order to provide proof of concept of the basic network algorithms. Finally, an underwater testbed proof of concept is planned in synergy with other existing projects such as MyPond and MySound at U-Conn, and at the UCLA Marina Aquatic Center. The theoretical and experimental work at all three collaborating teams shall involve graduate students and also expose undergraduate students to state-of-the art communication engineering projects, developed in the framework of the above mentioned testbeds.