Due to the water's characteristics, especially salt water, electromagnetic fields are quickly dissipated and absorbed in underwater environment. Thus, to transfer data in this medium, acoustic signals are mostly used due to their capability to transmit data over a long range. However, at very short range, electromagnetic signals enable high rate wireless data transfer in the underwater medium. The objective of this paper is to investigate underwater contactless data transfer technologies between an Autonomous Underwater Vehicle (AUV) and its docking station. On one hand, we consider 5 GHz Wi-Fi based transmission prototype and provide experimental end-to-end data performance for several transmission range, rotation angles, non-perfect alignment and transmit power. On the other hand we investigate the achievable rate of an experimental Magnetic Induction (MI) prototype performing contactless power and data transfer.

Deployment of autonomous underwater vehicle have became a focal point for many industrial application. The main challenge to the autonomous underwater vehicle is the finite mission time during the operation in a harsh environment. The most efficient way to reduce the operational and maintenance costs, and increase the efficiency of the system, the autonomous under water vehicles have to be recharged under water during the operation. This paper proposes an analysis of one side compensation topology and indexes some parameters to be considered when designing an inductive contactless power transfer system for AUV charging.

This paper deals with the binary classification of seals videos. To address this problem, we propose a novel system based on two phases : offline and online. The offline phase aims to generate a trained model. For this end, we use a transfer learning of convolutional neural network approach. Especially, we employed the VGG-16 deep network trained on the large ImageNet database and transfer their hyperparameters to recognize seals in the video. Finally, fully connected layer followed by a softmax method are used to decide if the seals exist or not in a given frame. For the online phase, we extract different information from a given video based on the file video name and on the trained model in the offline phase. The evaluation of the proposed method in a seals videos database demonstrates promising results.

This paper investigates the data gain brought by the multiple-input–multiple-output (MIMO) principle for underwater acoustic (UWA) communication. By considering a class of acoustic MIMO channels where each fading coefficient is modeled by a Rice distribution law with correlation factor depending on channel Doppler spread, we derive the achievable rate of such system that takes into account both channel estimation error and training sequence overhead. Model parameters are then estimated from an experiment campaign in a MIMO shallow water channel conducted in the roadstead of Brest, France. The system achievable rate is evaluated for several MIMO architectures and channel configurations and then compared against the conventional single-input–multiple-output transmission. The achievable rate gain is finally put in perspective with end-to-end data rate performance of a single-carrier MIMO transmission system experimented at sea.

This paper describes the use of the novel Orthogonal Chirp Division Multiplex (OCDM) waveform for an end-to-end high data rate underwater acoustic transmission system. The described OCDM based transmission is evaluated over a realistic underwater acoustic (UWA) channel and compared fairly against the conventional orthogonal frequency division multiplex (OFDM). By efficiently exploiting multi-paths diversity of the channel, OCDM is shown to provide better robustness than OFDM especially in case of a underloaded configuration when only a subset of N chirp waveforms are modulated leading to an interesting trade-off between spectral efficiency and robustness against UWA channel impairments.

This paper describes an end-to-end underwater acoustic transmission system based on MIMO (multiple input multiple output) principle. The described transmission system is experimented over various sea conditions and compared against conventional SIMO (single input multiple output) mode operating with a single transmission stream and multiple receive sensors. By analyzing the optimal performance of the proposed decoder, we select a MIMO and SIMO mode achieving similar robustness against underwater acoustic channel and demonstrate over longterm experiment, that, as foreseen by theory, MIMO technology provide a data rate gain at a same robustness level.

In the framework of the development of a MIMO (Multi-input Multi-output) capable underwater acoustic modem, this paper describes on the one hand a remotely operated experimental platform at sea in real conditions able to transmit and receive real-time multi-streams signals and the other hand an analysis of the experimental channel capacity gain brought by the MIMO approach with respect to traditional single transmitter system. The original remotely operated generation/acquisition system provides an extensive characterization of the MIMO underwater acoustic channel for various transmission ranges and sea states. The capacity analysis extracted from real conditions channel estimates confirm that MIMO approach in experimental scenario leads to a substantial capacity gain with respect to single transmitter even by taking account signal overhead. As foreseen by theory, the MIMO capacity gain is also shown to be dependent from transmission range and transducers arrays configuration.

Today, cabled seafloor observatories are installed at many sites around the globe, gathering different types of sensors in the marine environment where a Global Positioning System (GPS) signal is not accessible. Accurate time marking of ocean sensor data is highly important in many marine applications. This paper presents a smart GPS emulator based on the IEEE-1588 Precision Time Protocol (PTP). The device was designed and implemented to be able to provide accurate timing data (trigger + time code) to any ocean sensor as a broadband seismometer. In this case, accurate location and magnitude of a detected earthquake are dependent on the accuracy of the data timemarks. The performance of time synchronization is tested, using a commercial broadband seismometer, and the results are presented. These tests are based on a comparison of the synchronization trigger between master and slave clocks as well as the analysis of the data acquired by the seismometer. The work presented here leads to an improved performance of the ocean-bottom seismometers as well as tsunami warning systems.

Argos link (satellite) http://www.argos-system.org Global coverage, tracking and monitoring system, polar orbiting satellites, relatively low bandwidth (approx. from few kilobits/day to few kilobits/second), cost for an intensive use. Iridium link (satellite) http://www.iridium.com Global coverage, 66 satellites in low-Earth orbit (LEO), relatively low bandwidth (approx. few kilobits/second), and cost for an intensive use. 3G/4G networks For onshore or coastal applications, uplink speed is usually lower than downlink, bandwidth is not guaranteed, ISPs usually provide data plans with a monthly data usage limit. Private wireless radio link Private link, high bandwidth, guaranteed bandwidth, unlimited data usage, limited distance and restricted coverage. Private telecom marine cable Certainly the best solution for a permanent observatory (virtually unlimited data + power) but very expensive.

Sea Test Base / Celadon is a non-profit organization, which offers testing facilities mainly based on a coastal non-cabled observatory, dedicated for designing, testing and qualifying marine sensors and marine instruments in real conditions. This platform can be considered as an experimental laboratory at sea, available 24/7/365 from the Internet. In a first step, we describe, analyse and compare the differences in terms of architecture and services between conventional marine cabled, non-cabled observatories and Sea Test Base facilities. In a second step, we present the results obtained with this platform to design, improve and qualify the hardware and the algorithms for a Multiple-Input and Multiple-Output (MIMO) modem for underwater acoustic communication. In conclusion, we present the upcoming extension of Sea Test Base, which consists of a shared open platform based on a mesh network including buoys and a pontoon, dedicated to underwater acoustic experiments at sea.

A power-over-fiber system using a single fiber optic of 10km devoted to transmit the energy to supply a remote equipment and also the up/down-stream data exchanged between a shore station and the equipment, is presented in this paper. Firstly, we present an overview of the developed quasi-all-optical architecture and its main communication protocol: the SPI. Then, we focus on its low consumption remote equipment and more precisely on a device called “the interface instrument” which is made of two main units. One realizes the optical toelectrical conversion to provide the power supply. The other one manages the optical data. Afterwards, experimental characterizations of the prototype are given and discussed.

This paper describes two methods for characterizing underwater optical elements under both hyper-baric pressure and high optical power. Commercial optical feedthroughs to be tested were inserted in a hyper-baric vessel and a 600 bar pressure was applied as an optical power of several Watts was sent over the optical fibers. An unexpected decrease of the optical losses was evidenced.

ABSTRACTA 10‐km long extension using power‐over‐fiber technology and dedicated to sea floor observatories is presented with a particular focus on the electrical both ends units. We propose a solution using a serial peripheral interface 3 wires protocol for large range of low‐power sensors like hydrophones. © 2013 Wiley Periodicals, Inc. Microwave Opt Technol Lett 55:2562–2568, 2013

Historically, observation in Marine Science was mainly based on in situ measurements made mainly over ship surveys and shore measurements. Unfortunately, ship surveys can only be episodic, and are constrained by weather and by the constant rise of ship-time cost. As the data provided by non-communicating moorings are stored in the measurement system, a ship intervention is needed to recover both the mooring and the data after several acquisition months. Further to the rather successful mediumand short-term deployment of these traditional devices, scientists have expected the development of long-term observations and permanent marine system-monitoring tools so as to gain more insight into the observed processes. By providing additional information, satellite technology can partly solve this gap between the reality and expectations. However, even though satellite images provide information over a large time frame (from minutes to years) and a wide range of spatial resolutions (from metres to thousands of kilometres), they only cover the upper layer of the sea. An Open-Sea Observatory is a complementary tool that allows one to make, in the water column and on the seafloor, long-term measurements of many environmental parameters and to acquire them in real-time, or near real-time. In addition to this real-time data transmission, these systems permit remote intervention by humans when needed, and thus can be considered as 2-way communicating devices. Because of these two characteristics, observatories are innovative systems that bring internet to the ocean and make the ocean reality visible to the human eye. According to our definition of an Open-Sea observatory, other very useful observation tools such as gliders, floats, repeated profiler transects, etc. will not be considered in this chapter to only focus on such ocean observatories. Observatory initiatives have been spreading worldwide since the 1990s. In Europe, several initiatives started twenty years ago so as to upgrade free-fall systems from the sea surface (the so-called “landers”) to make them 2-way communicating and to develop bottom

Stated that a trend in the sensor technology is the development of intelligent sensors also called smart sensors. The development of such sensors do not only rely on the hardware development but also on the software. The later should so meet the requirements on low costs and of quality. This paper presents our approach to model the software of a smart sensor and to generate the code for the embedded real-time application. It will also describe how the use of a domain-specific modelling methodology enabled us to achieve a high level of modularity which will permit to save costs and development time.