Reliable Communication in Underwater Acoustic Sensor Networks

As a novel networking paradigm for aqueous environmental monitoring and exploration, large-scale mobile underwater acoustic sensor network has attracted significant attention recently from both academia and industry. However, the adverse environmental conditions feature specific characteristics of underwater sensor networks, such as low communication bandwidth, large propagation delay, floating node mobility, high error probability, and intermittent connectivity, and pose a range of challenges to underwater communication and networking. In this dissertation, we will tackle the fundamental issue of reliable communication in underwater acoustic sensor networks. ^ Underwater network communication involves multiple sensors that are powered by battery and hard to be replaced or recharged, so energy efficiency is one paramount consideration. Based on the aforementioned observations, we explore from the following two aspects to provide reliable communication: (1) protect information transmitted through the acoustic channels; and (2) use appropriate routing schemes to enable reliable communication in intermittently connected underwater sensor networks. ^ Firstly, we propose an efficient error-recovery scheme that carefully couples network coding and multi-path routing. Through an analytical study, we provide guidance on how to choose parameters in our scheme and how to extend it to large-scale networks. Through simulation, we demonstrate that the scheme is efficient in both error recovery and energy consumption.^ Secondly, we propose a practical scheme, called non-binary joint network channel coding (NB-JNCC) for reliable communication. Through channel-network codes integration and iterative two-tier joint decoding, NB-JNCC seamlessly couples channel coding and network coding and effectively combats the detrimental effect of fading conditions in wireless channels. This protocol can fully exploit the spatial diversity and redundancy residing in both physical and network layers.^ Finally, we present a generic prediction assisted single-copy routing (PASR) scheme that can be instantiated for different mobility models. PASR employs an effective greedy algorithm, aggressive chronological projected graph (ACPG), which can capture the features of network mobility patterns and provide guidance on how to use historical information. As a result, prediction assisted single-copy online routing protocols instantiated from ACPG are energy efficient and cognizant of the underlying mobility patterns. We demonstrate the advantages of PASR in underwater sensor networks with various mobility models.