On the Relationship Between the Underwater Acoustic and Optical Channels

Diamant, Roee; Campagnaro, Filippo; Grazia, Michele De Filippo De; Casari, Paolo; Testolin, Alberto; Calzado, Violeta Sanjuan; Zorzi, Michele

doi:10.1109/twc.2017.2756055

Cited by 37 publications

(20 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…C. Benchmark models 1) Support Vector Machines: For the binary fish detection task, the performance of deep learning models is validated against Support Vector Machines (SVMs) [18] -a more traditional popular class of machine-learning models that have also been successfully applied to underwater signals (e.g., [9]). In particular, we test two types of SVMs with both linear and non-linear kernels (polynomial and radial basis function), trained through sequential minimal optimization [50] for a maximum of 30 objective evaluations using a 4-fold crossvalidation scheme.…”

Section: ) Lstm Training and Testingmentioning

confidence: 99%

“…In the context of detection at low SCR levels, machine learning techniques can offer high potential, providing an efficient, data-driven approach for solving complex acoustic detection problems, even in challenging underwater sea environments [9]. In particular, deep learning enables training artificial neural networks composed of many processing layers that learn high-level representations of the data by exploiting multiple levels of abstraction [10].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Detecting Submerged Objects Using Active Acoustics and Deep Neural Networks: A Test Case for Pelagic Fish

Testolin

Kipnis

Diamant

2022

IEEE Trans. on Mobile Comput.

Self Cite

View full text Add to dashboard Cite

The accurate detection and quantification of submerged targets has been recognized as a key challenge in marine exploration, one that traditional census approaches cannot handle efficiently. Here we present a deep learning approach to detect the pattern of a moving fish from the reflections of an active acoustic emitter. To allow for real-time detection, we use a convolutional neural network, which provides the simultaneous labeling of a large buffer of signal samples. This allows to capture the structure of the reflecting signal from the moving target and to separate it from clutter reflections. We evaluate system performance both on synthetic (simulated) data, as well as on real data recorded over 50 sea experiments in a variety of sea conditions. When tested on real signals, the network trained on simulated patterns showed non-trivial detection capabilities, suggesting that transfer learning can be a viable approach in these scenarios, where tagged data is often lacking. However, training the network directly on the real reflections with data augmentation techniques allowed to reach a more favorable precision-recall trade-off, approaching an ideal detection bound. We also evaluate an alternative model based on recurrent neural networks which, despite exhibiting slightly inferior performance, could be applied in scenarios requiring on-line processing of the reflection sequence.

show abstract

Section: ) Lstm Training and Testingmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Detecting Submerged Objects Using Active Acoustics and Deep Neural Networks: A Test Case for Pelagic Fish

Testolin

Kipnis

Diamant

2022

IEEE Trans. on Mobile Comput.

Self Cite

View full text Add to dashboard Cite

show abstract

“…With its kernel 'trick', SVR can capture highly non-linear relations with only a few user-defined parameters. As shown in our recent works for optic-acoustic classification and sonar target detection [28,29], SVR can be used successfully for seemingly non-related datasets.…”

mentioning

confidence: 90%

Cross-Sensor Quality Assurance for Marine Observatories

et al. 2020

Self Cite

View full text Add to dashboard Cite

Measuring and forecasting changes in coastal and deep-water ecosystems and climates requires sustained long-term measurements from marine observation systems. One of the key considerations in analyzing data from marine observatories is quality assurance (QA). The data acquired by these infrastructures accumulates into Giga and Terabytes per year, necessitating an accurate automatic identification of false samples. A particular challenge in the QA of oceanographic datasets is the avoidance of disqualification of data samples that, while appearing as outliers, actually represent real short-term phenomena, that are of importance. In this paper, we present a novel cross-sensor QA approach that validates the disqualification decision of a data sample from an examined dataset by comparing it to samples from related datasets. This group of related datasets is chosen so as to reflect upon the same oceanographic phenomena that enable some prediction of the examined dataset. In our approach, a disqualification is validated if the detected anomaly is present only in the examined dataset, but not in its related datasets. Results for a surface water temperature dataset recorded by our Texas A&M—Haifa Eastern Mediterranean Marine Observatory (THEMO)—over a period of 7 months, show an improved trade-off between accurate and false disqualification rates when compared to two standard benchmark schemes.

show abstract

“…The maximum range of the optical communication also depends on how the attenuation coefficient changes along the water column: in this work, we used the values of the attenuation coefficient depicted in Figure 5a and measured during the ALOMEX'15 research cruise offshore the coast of Morocco (latitude 30 • 42.520' N and longitude 10 • 18.680' W). The ALOMEX'15 cruise was organized by the NATO STO Centre of Marine Research and Experimentation (CMRE), and its dataset was first presented in [18].…”

Section: Optical Modem Simulatormentioning

confidence: 99%

Wireless Remote Control for Underwater Vehicles

Campagnaro

Signori

Zorzi

2020

JMSE

Self Cite

View full text Add to dashboard Cite

Nowadays, the increasing availability of commercial off-the-shelf underwater acoustic and non-acoustic (e.g., optical and electromagnetic) modems that can be employed for both short-range broadband and long-range low-rate communication, the increasing level of autonomy of underwater vehicles, and the refinement of their underwater navigation systems pave the way for several new applications, such as data muling from underwater sensor networks and the transmission of real-time video streams underwater. In addition, these new developments inspired many companies to start designing hybrid wireless-driven underwater vehicles specifically tailored for off-shore operations and that are able to behave either as remotely operated vehicles (ROVs) or as autonomous underwater vehicles (AUVs), depending on both the type of mission they are required to perform and the limitations imposed by underwater communication channels. In this paper, we evaluate the actual quality of service (QoS) achievable with an underwater wireless-piloted vehicle, addressing the realistic aspects found in the underwater domain, first reviewing the current state-of-the-art of communication technologies and then proposing the list of application streams needed for control of the underwater vehicle, grouping them in different working modes according to the level of autonomy required by the off-shore mission. The proposed system is finally evaluated by employing the DESERT Underwater simulation framework by specifically analyzing the QoS that can be provided to each application stream when using a multimodal underwater communication system specifically designed to support different traffic-based QoSs. Both the analysis and the results show that changes in the underwater environment have a strong impact on the range and on the stability of the communication link.

show abstract

On the Relationship Between the Underwater Acoustic and Optical Channels

Cited by 37 publications

References 36 publications

Detecting Submerged Objects Using Active Acoustics and Deep Neural Networks: A Test Case for Pelagic Fish

Detecting Submerged Objects Using Active Acoustics and Deep Neural Networks: A Test Case for Pelagic Fish

Cross-Sensor Quality Assurance for Marine Observatories

Wireless Remote Control for Underwater Vehicles

Contact Info

Product

Resources

About