Enabling High-Speed Communication between Autonomous Underwater Vehicles

加拿大曾经有一个蓬勃发展的水下声学研究计划，用于敌舰回声定位，由加拿大国防研究与发展局(DRDC)在哈利法克斯领导，但20世纪90年代初冷战结束后，这些计划的资金突然中断。最近，人们对海底勘探的各种应用越来越感兴趣，例如海港监视、海洋学行为分析和近海设备监测。例如，在海上时代，为在芬迪湾高效发电而部署涡轮机方面投入了大量资金。监测野生动物和海底设备对于评估环境和经济影响至关重要。为了避免电缆的负担，水声通信应运而生。在拟议的研究计划中，我们打算开发最先进的设备，以便在海底环境中进行声音通信。为了延长部署的水声设备的寿命，需要考虑低功率技术，但这将固有地限制传输范围。为了帮助增加通信范围，我们建议使用自主水下机器人(AUV)。然而，由于技术限制，这些设备的通信能力目前受到强烈限制。因此，一旦AUV发射，通常没有反馈机制，AUV的任务无法实时监控。需要允许视频传输的可靠通信链路，并且将需要100kbps量级的高数据速率。现有的水声调制解调器的吞吐量非常有限，约为10kbps，并且消耗大量功率，从而限制了AUV中电池的寿命。水下通信的技术挑战部分是由于用于水下传输的设备操作频率较低，但主要是因为声音在水下的传播不可预测。与人们熟知的电波传播信道相比，水声信道的损伤是相当严重的。人们可以想象，在洋流和海浪创造了动态变化的环境的海洋中，很难预测信号的失真。目前控制信号传播的物理现象的准确模型包括频率相关吸收、多径到达、由于移动性和小尺度波动引起的多普勒频移。这项拟议的工作旨在使用在定制处理器上实现的新算法将水下通信的吞吐量提高至少一个数量级。为了减少失真，所提出的数字算法将需要补偿水下传播的极端损伤。此外，发射器和接收器将配备多个换能器和水听器，以提高数据速率。使用多个传感器将需要大量的内存和重要的计算资源。为了满足这些要求，信号处理将在定制的完全集成平台上进行编程。此外，为了允许与处理器的模拟接口，从而最大限度地减少片外互连的数量，集成电路还将保持高分辨率的数据转换器。这项研究计划将在通信、信号处理和超大规模集成(VLSI)技术领域带来创新的解决方案。加拿大被3大片水域包围，这项研究将使加拿大受益匪浅。这项研究将与哈利法克斯当地工业合作进行，用于需要海底监测的商业和科学应用。

Canada once had a thriving research program in underwater acoustics for the echolocation of enemy vessels led by Defence Research & Development Canada (DRDC) in Halifax but the end of the Cold War in the early ’90s brought a sudden interruption to the funding of these programs. Recently there has been an increasing interest in subsea exploration for a variety of applications, such as seaport surveillance, oceanographic behaviour analysis and monitoring of offshore equipment. For example, in the Maritimes, there is significant investment in the deployment of turbines for efficient energy generation in the Bay of Fundy. Monitoring of the wildlife and subsea equipment is crucial to assess environmental and economic impacts. To avoid the burden of cables, acoustic communication is proposed. In the proposed research program, we intend to develop state-of-the-art equipment to communicate acoustically in subsea environments. To extend the lifetime of deployed underwater acoustic equipment, low power techniques shall be considered, but will inherently limit transmission range. To help increase the communication range, we propose to use autonomous underwater vehicles (AUVs). However, communication capabilities to these devices are currently strongly restricted, due to technology limitations. As a result, once an AUV is launched, there is typically no feedback mechanism and the AUVs' mission cannot be monitored and controlled in real-time. A reliable communication link that allows video transmission is needed and would require a high data rate on the order of 100 kbps. Existing underwater acoustic modems have very limited throughput, on the order of 10 kbps and consume significant power, thus limiting the lifetime of the batteries in AUVs. The technical challenges for underwater communication are, in part, due to the low frequency of operation of the equipment used for underwater transmission, but primarily to the unpredictable propagation of sound underwater. In comparison to the well-understood radio-electric propagation channel, the underwater acoustic channel impairments are quite severe. One can imagine that in the ocean where currents and waves create a dynamically varying environment, it is difficult to predict the distortion of the signal. An accurate model of the physical phenomena that govern the signal propagation currently includes frequency dependent absorption, multipath arrival, Doppler shift due to mobility and small-scale fluctuations. The proposed work aims to improve the throughput of underwater communication by at least one order of magnitude using novel algorithms implemented on custom processors. In order to mitigate distortion, the proposed digital algorithms will need to compensate for the underwater propagation extreme impairments. Also, the transmitter and receiver will be equipped with multiple transducers and hydrophones to increase the data rate. Using multiple transducers will require a significant amount of memory and important computation resources. To satisfy these requirements, the signal processing will be programmed on a custom fully integrated platform. Additionally, to allow an analog interface to the processor and consequently minimize the number of off-chip interconnections, the integrated circuit will also hold high-resolution data converters. This research program shall lead to innovative solutions in the fields of communication, signal processing and very large scale integration (VLSI) technology. Canada being surrounded by 3 large bodies of water will benefit greatly from the research developed in this project. This research will be conducted in collaboration with local industry in Halifax for commercial and scientific applications that require sub sea monitoring.