Collaborative Research: Efficient and Robust Underwater Acoustic Sensor Networks: An Integrated Coding Approach

University Of Connecticut, Storrs CT

Investigators

Abstract

The objective of this research is to address one of the greatest challenges that hinder the development of underwater sensor networks into practical systems: How to ensure data reliability and delivery efficiency over unreliable acoustic links with large error rates, low available bandwidth, and long propagation delays. The project proposes an integrated coding approach, seamlessly combining digital fountain, and network and channel coding schemes with complementary strengths, across the application, network, and physical layers. The intellectual merit of this project lies in the rigorous study of the underwater network limits and a systematic design of powerful fountain and network coding schemes to address the challenging issues facing underwater sensor networks. The project outcomes will include physical layer network coding theoretical results, novel non-binary fountain codes, sparse linear network codes, and physical-layer network coding schemes for underwater channels, as well as networking protocols for underwater unicast scenarios. The theoretical results, coding, and protocol designs will be evaluated through real world experiments. The broader impact of this project is reflected in two aspects. First, this project is expected to accelerate the development of emerging underwater sensor networks, which will have major impact on scientific communities, national security, and the society as a whole. Second, this project will expose graduate students to new challenging research problems, fostering them to become leading experts in the fast-growing field of underwater acoustic networks. It will also support undergraduates, women and other under-represented groups, allowing them to participate in advanced research and its practical applications.