<rss version="2.0" xmlns:atom="https://www.w3.org/2005/Atom">
  <channel>
    <title>Transport Research International Documentation (TRID)</title>
    <link>https://trid.trb.org/</link>
    <atom:link href="https://trid.trb.org/Record/RSS?s=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" rel="self" type="application/rss+xml" />
    <description></description>
    <language>en-us</language>
    <copyright>Copyright © 2026. National Academy of Sciences. All rights reserved.</copyright>
    <docs>http://blogs.law.harvard.edu/tech/rss</docs>
    <managingEditor>tris-trb@nas.edu (Bill McLeod)</managingEditor>
    <webMaster>tris-trb@nas.edu (Bill McLeod)</webMaster>
    <image>
      <title>Transport Research International Documentation (TRID)</title>
      <url>https://trid.trb.org/Images/PageHeader-wTitle.jpg</url>
      <link>https://trid.trb.org/</link>
    </image>
    <item>
      <title>Ship-radiated noise separation via auditory scene analysis-guided deep learning</title>
      <link>https://trid.trb.org/View/2773659</link>
      <description><![CDATA[Ship-radiated noise is an important acoustic source in passive underwater monitoring, but marine recordings often contain overlapping components from multiple sources, which can hinder target-related analysis. To address this problem, this paper develops a ship-radiated noise separation method that integrates auditory scene analysis with deep learning. Following the computational auditory scene analysis framework, the proposed method consists of auditory segmentation and auditory grouping. In the segmentation stage, a Gammatone filter bank provides time-frequency decomposition, and Dense-UNet extracts deep features and estimates separation masks from the resulting time-frequency representation. In the auditory grouping stage, the separation masks estimated by the network are first mapped from the Gammatone frequency-band domain onto the linear frequency axis. These masks are then applied to the complex spectrogram of the original mixed signal, which helps reduce reconstruction distortion when converting the separated auditory-band representations back to the full-band waveform. The proposed method was evaluated on the ShipsEar and DeepShip datasets. Under the tested conditions, the method consistently achieved competitive performance across multiple objective metrics compared to other commonly used models. Further analysis also revealed that the method exhibits enhanced capability in recovering low-frequency line-spectrum structures, which are crucial for ship-radiated noise characterization.]]></description>
      <pubDate>Mon, 28 Sep 2026 08:37:09 GMT</pubDate>
      <guid>https://trid.trb.org/View/2773659</guid>
    </item>
    <item>
      <title>Underwater acoustic signal recognition method jointly driven by dual-adaptive reduced-order variational mode decomposition and multiscale direction trend Lempel-Ziv complexity</title>
      <link>https://trid.trb.org/View/2763107</link>
      <description><![CDATA[Underwater acoustic signal is easily influenced by complex marine environment and present nonlinear and non-stationary characteristics, so it is difficult to fully explore its internal dynamic characteristics. Therefore, signal recognition method based on dual-adaptive reduced-order variational modal decomposition (D-ARVMD) and multiscale direction trend Lempel-Ziv complexity(MDTLZC) is proposed. Firstly, D-ARVMD is proposed, and dynamic optimization of decomposition parameters is realized by residual energy ratio attenuation rate and spectrum structure perception, which significantly improves the anti-aliasing ability and fidelity of signal decomposition. Secondly, direction trend Lempel-Ziv complexity(DTLZC) is proposed, which uses the first-order difference vector in phase space and adaptive coding to accurately capture signal evolution trend. Subsequently, MDTLZC is proposed through the idea of multiscale analysis to realize multiscale fine characterization of signal dynamics characteristics. Finally, the Spearman correlation coefficient is used to screen the dominant eigenmodes, which are fused with the original signal features to realize the complementary advantages of dynamic information. The verification on the measured data sets of ship radiated noise and marine mammal acoustic signal shows that the recognition method has achieved high accuracy, which provides an efficient and reliable new idea for complex underwater acoustic signal recognition.]]></description>
      <pubDate>Wed, 16 Sep 2026 16:15:38 GMT</pubDate>
      <guid>https://trid.trb.org/View/2763107</guid>
    </item>
    <item>
      <title>Gaussian process estimation of underwater acoustic fluctuations: Experimental validation on the Iceland–Faroe polar front</title>
      <link>https://trid.trb.org/View/2659525</link>
      <description><![CDATA[Forecasting underwater acoustic propagation in oceanic frontal areas is a difficult task due to their unstable dynamics. In this work, we propose to fit a Gaussian Process model, with a kernel derived from a structure model, to infer the position of the front from profiler data. Samples from the Gaussian Process can be used to generate sound-speed fields. Parabolic equation simulations on those samples show a good agreement with experimental acoustic data in propagation parallel to and across the front. As it can be intuitively expected, the discrepancy is a bit higher for across-front propagation due to strong range-dependence. However, these discrepancies are statistically due to Gaussian Process samples which proportion do not exceed 10% of the simulated data.]]></description>
      <pubDate>Tue, 01 Sep 2026 14:02:30 GMT</pubDate>
      <guid>https://trid.trb.org/View/2659525</guid>
    </item>
    <item>
      <title>Ecological effects and mitigation strategies for underwater noise from shipping activities</title>
      <link>https://trid.trb.org/View/2719353</link>
      <description><![CDATA[In recent years, concerns about the harmful effects of underwater radiated noise (URN) on marine ecosystems have grown. Shipping activities are the most significant contributors to continuous low-frequency noise (<1000 Hz). This noise overlaps with the hearing range of marine organisms, disrupting vital processes like foraging, mating and predator-prey interactions. Over time, such disturbances may lead to population declines and shifts in community dynamics. Ship-generated noise dominates low-frequency sound pollution in marine environments due to its persistent and far-reaching nature, affecting more organisms over larger areas than other noise sources. Despite its critical impact, there are no mandatory regulations to limit ship noise, aside from non-binding guidelines by the International Maritime Organization (IMO). This article explores the ecological impacts of URN, emphasizing the role of ship propulsion systems in noise generation and its effects on marine life. It reviews mitigation strategies, including quieter propulsion technologies and operational practices and highlights the need for noise limits on new ships and improved maintenance for existing vessels.]]></description>
      <pubDate>Tue, 01 Sep 2026 14:02:30 GMT</pubDate>
      <guid>https://trid.trb.org/View/2719353</guid>
    </item>
    <item>
      <title>Rate-Splitting Multiple Access with Backward Retransmission for Underwater Acoustic Networks: Efficient and Reliable Communication</title>
      <link>https://trid.trb.org/View/2711574</link>
      <description><![CDATA[The underwater acoustic networks (UANs) hold great promise for the development of the ocean and effectively support future communication demands for seamless global coverage. Nevertheless, limited underwater frequency resources and conventional orthogonal multiple access techniques cannot fulfill the exploding demand for data transmission, posing great challenges to the efficient operation of UANs. Fortunately, rate splitting multiple access (RSMA) with retransmission protocol can achieve concurrent connections and high reliability. Therefore, in this article, we first summarize representative multiple access solutions for UANs and illustrate the advantages and suitability of RSMA in underwater. Then, a cluster-based UAN architecture is presented as a case study to evaluate the characteristics and diverse applications in underwater. Furthermore, we introduce a RSMA scheme with backward retransmission (BRQ) and describe the integration of the two mechanisms in detail, aiming at managing serious interference and maintaining link stability. Simulation results demonstrate the feasibility and effectiveness of the proposed scheme compared with other benchmarks. Open issues and future directions are also discussed for UANs adopting nonorthogonal retransmission.]]></description>
      <pubDate>Mon, 31 Aug 2026 10:31:31 GMT</pubDate>
      <guid>https://trid.trb.org/View/2711574</guid>
    </item>
    <item>
      <title>A lightweight multi-scale gated expert network for ship-radiated noise recognition</title>
      <link>https://trid.trb.org/View/2710047</link>
      <description><![CDATA[Ship-radiated noise recognition remains challenging because segmented training recordings often contain redundant samples, single-resolution spectral inputs may not capture acoustic structures at different temporal scales, and static feature fusion cannot adjust scale contributions across samples. A lightweight multi-scale gated expert network is proposed for underwater ship-radiated noise recognition. The method integrates three task-oriented designs: a hard-random sample reconstruction strategy based on probe-estimated sample difficulty, a dual-resolution log-Mel input constructed with two hop sizes, and a gated multi-scale expert network with short-, medium-, and long-term temporal receptive-field branches. A Transformer encoder is used for temporal context modeling, and gating regularization is introduced to reduce near-uniform routing and single-branch dominance. Under a unified experimental protocol, the proposed model achieves accuracies of 98.43 ± 0.08% on DeepShip and 98.65 ± 0.07% on ShipsEar with 0.482M parameters. Additional MMSI-level vessel-disjoint evaluation on DeepShip, controlled acoustic perturbation tests, statistical significance analysis, and inference-cost evaluation are conducted to examine performance under stricter and degraded conditions. Ablation results indicate that sample reconstruction, dual-resolution representation, and gated multi-scale fusion each contribute to the final performance. The results support a favorable accuracy–complexity trade-off on the evaluated public datasets, while the conclusions remain limited to public benchmark protocols and controlled perturbation settings rather than open-sea deployment conditions.]]></description>
      <pubDate>Wed, 10 Jun 2026 16:38:53 GMT</pubDate>
      <guid>https://trid.trb.org/View/2710047</guid>
    </item>
    <item>
      <title>An ocean knowledge-driven adversarial learning method to decouple ocean ambient noise and multipath effect for underwater acoustic target recognition</title>
      <link>https://trid.trb.org/View/2707632</link>
      <description><![CDATA[Ocean ambient noise and multipath effect severely impact the characteristics of ship radiated noise, reducing the accuracy and robustness of passive underwater acoustic target recognition methods. This paper proposes an ocean knowledge-driven adversarial learning method to decouple ocean ambient noise and multipath effect for underwater acoustic target recognition. By applying the ocean knowledge-driven mechanism across multiple deep layers, the method learns the frequency distribution of ocean ambient noise and the principles of multipath propagation. Guided by this physical relationship, the spectrum of ship radiated noise is reconstructed. The deep frequency features from ship radiated noise are then extracted to operate the recognition task. The adversarial game design enables the models to distinguish environmental interference in the deep feature domain and to enhance nonlinear correlations with ocean knowledge. Experimental results show that the proposed method effectively decouples environmental interference and enhances physical interpretability. Visualization results confirm the establishment of knowledge-based physical connections. Recognition experiments demonstrate that the method achieves superior accuracy across various SNR and multipath conditions, improving accuracy by up to 11.37% compared with the baseline. The recognition performance on real-world data further demonstrates the practicality and generalizability of the proposed method.]]></description>
      <pubDate>Mon, 01 Jun 2026 09:14:40 GMT</pubDate>
      <guid>https://trid.trb.org/View/2707632</guid>
    </item>
    <item>
      <title>External Acoustic Black Holes to reduce structural vibration and radiated sound by a plate in contact with water</title>
      <link>https://trid.trb.org/View/2681482</link>
      <description><![CDATA[Ship noise arises from various sources, including propeller cavitation, hydrodynamic flow, and onboard machinery. Vibrations from rotating and reciprocating machines can propagate through the ship’s structure and radiate into the marine environment, significantly contributing to underwater noise pollution. This study investigates the potential of Acoustic Black Holes (ABH) to mitigate structural vibrations and, consequently, reduce underwater radiated noise. ABHs are passive vibration-damping devices that utilize a unique mechanism to concentrate and dissipate vibrational energy. While their potential on structures in air is recognized, the effect on structures in contact with water has yet to be investigated. A series of experiments was conducted on a rectangular steel plate subjected to controlled excitation using an electrodynamic shaker. The results show a noticeable reduction in structural vibration levels across a broad frequency range (50–5000 Hz) with the implementation of ABHs. However, this reduction does not directly translate into a corresponding decrease in underwater radiated noise. The inconsistency between vibration reduction and acoustic radiation was analysed and discussed. These findings provide insight into the effectiveness and limitations of ABHs in underwater noise mitigation strategies.]]></description>
      <pubDate>Thu, 14 May 2026 17:05:12 GMT</pubDate>
      <guid>https://trid.trb.org/View/2681482</guid>
    </item>
    <item>
      <title>Quantum machine learning for passive sonar: Spectral feature extraction and Hybrid convolution neural network classification of ship propeller signals</title>
      <link>https://trid.trb.org/View/2697578</link>
      <description><![CDATA[Underwater acoustic classification of propeller-driven vessels is hindered by environmental variability, limited labeled data, and the operator-dependent nature of Detection of Envelope Modulation on Noise (DEMON) preprocessing. We present a two-stage framework consisting of (i) a Spectral Amplitude Variation (SAV) algorithm that extracts discriminative frequencies through temporal variance of spectral amplitudes without manual bandwidth selection, and (ii) a Hybrid Quantum-Convolutional Neural Network (HQ-CNN) that integrates a 10-qubit parameterized quantum circuit with classical convolutional layers. Mathematical analysis shows that SAV exhibits cubic noise-power dependence compared with DEMON quartic dependence, providing improved robustness in low-SNR regimes. On the ShipEar and DeepShip benchmarks, SAV achieves detection rates of 98.22% and 99.04%, respectively, with false-alarm rates of 3.5% and 3.2%, while operating from 140x to 175x faster than DEMON-Hilbert. The HQ-CNN attains classification accuracies of 93.7% (ShipEar) and 96.8% (DeepShip), outperforming a classical CNN baseline by 3.7 and 5.2 percentage points. Source-stratified 5-fold cross-validation with BCa bootstrap (B = 10,000) confirms statistical significance (p<0.05, d = 3.10 and 4.83). Systematic comparison against seven classical regularization techniques demonstrates that the HQ-CNN yields near-zero generalization gaps (0.2% and 0.15%). Analysis of the trained quantum circuit reveals that individual qubits specialize in distinct frequency bands consistent with established underwater acoustic source mechanisms. All quantum-circuit evaluations were performed via noiseless statevector simulation and therefore represent an upper bound on performance attainable with current NISQ hardware.]]></description>
      <pubDate>Thu, 30 Apr 2026 16:39:45 GMT</pubDate>
      <guid>https://trid.trb.org/View/2697578</guid>
    </item>
    <item>
      <title>Cavitation underwater radiated noise methodology applied to a propeller operating in-behind condition</title>
      <link>https://trid.trb.org/View/2653052</link>
      <description><![CDATA[Concerns about shipping noise influence on marine life are motivating mitigation efforts, increasing interest in numerical methods for underwater radiated noise (URN) assessment. A common approach is to use incompressible flow solvers with the Ffowcs Williams-Hawkings (FW-H) method, though reliability is shown to depend on modelling choices. To advance on this matter, model and full scale simulations are performed for a fully appended ship, operating in mildly cavitating conditions, using the incompressible Reynolds-averaged Navier–Stokes (RANS) framework to investigate pressure pulse and URN levels. Predictions are compared with model scale experiments and full scale sea trial measurements. The permeable FW-H method is used, while also an alternative approach modelling cavitation as a monopole source combined with the solid FW-H formulation and corrections for Lloyd’s mirror effect is proposed and tested. Further, the effect of blockage on cavitation is investigated and found to affect induced pressure pulse and URN levels due to effects on the ship wakefield in the tunnel section. Although good agreement is found for cavity dynamics and pressure pulses, significant discrepancies are found in URN levels between measurements and numerical predictions, for both methods. Reasons for these differences are partly understood and discussed, while other questions remain open.]]></description>
      <pubDate>Wed, 08 Apr 2026 13:57:21 GMT</pubDate>
      <guid>https://trid.trb.org/View/2653052</guid>
    </item>
    <item>
      <title>From sky to underwater: Can passive acoustic monitoring replace plane-based aerial photography for quantifying human use of inland waters?</title>
      <link>https://trid.trb.org/View/2677284</link>
      <description><![CDATA[Large inland waters concentrate most of the alterations linked to human activities and management strategies are needed to maintain a good level of ecosystem services. However, assessing human presence is often challenging – particularly in large socio-ecosystems – due to detection thresholds, spatial resolution limits, and the high cost of traditional monitoring methods. Although passive acoustics-based technologies have advanced considerably, they remain underutilized in freshwater management and we lack direct comparisons between sound levels and exhaustive counts from traditional methods. In the present study, we tested whether underwater Passive Acoustic Monitoring (PAM), which consists of deploying acoustic recorders underwater to record the soundscape, could be used to monitor boating activity on a large multi-use lake. We deployed acoustic recorders at five different locations of Lake Bourget, the largest natural lake in France, where Plane-based Aerial Photography (PAP) was carried out during the summer to provide comprehensive counts of human presence. Sound pressure levels increased with PAP-derived motorboat numbers and the strength of the relationship was high (Pearson's correlation coefficients between 0.60 and 0.79) to very high (0.80 <r < 1) at four locations, and very low (−0.19 <r < 0.19) to low (0.20 <r < 0.39) at the fifth. Averaging sound pressure levels over the one-third octave band centered at 1584 Hz rather than the 0.01–20 kHz frequency band slightly improved the correlations. Conversely, motorboat numbers estimated from spectrogram visual inspection underestimated the actual numbers of motorboats. Compared to PAP, underwater PAM does not provide exhaustive counts, has a low spatial resolution, and assessing detection thresholds can be challenging. However, it can be deployed almost continuously over much longer periods and may prove more cost-effective in the long term. From the managers' perspective, underwater PAM could be used instead of or in combination with visual approaches to quantify inland water use.]]></description>
      <pubDate>Fri, 27 Mar 2026 14:27:56 GMT</pubDate>
      <guid>https://trid.trb.org/View/2677284</guid>
    </item>
    <item>
      <title>Model scale noise measurements of the research vessel Nawigator XXI propeller</title>
      <link>https://trid.trb.org/View/2682224</link>
      <description><![CDATA[In the last decades, the prediction and mitigation of ships underwater radiated noise have become an important matter of study for research institutions in naval hydrodynamics, mostly because of the adverse impacts that anthropogenic noise may have on marine ecosystems. Propellers are commonly recognized as the main noise sources, when cavitating, therefore large effort is dedicated to the prediction of propeller cavitation and noise. Model scale noise measurements represent the oldest approach to perform this task, and its effectiveness is generally rather good. Yet the extrapolation of full scale noise from model tests presents many challenges associated to the many scale effects involved, as well as the effects of the adopted facility and experimental procedure. In this framework, the availability of comparisons between model test results and sea trials, as well as comparative studies among various facilities is of utmost importance to define and improve proper testing and scaling procedures for cavitating propeller noise. This work presents the results of the extensive experimental campaign carried out at the University of Genova cavitation tunnel on the model propeller of the research vessel Nawigator XXI. This propeller will be the object of the Round Robin test program planned within the International Towing Tank Conference, therefore current work will provide valuable information on this important test case in the view of future analyses and comparisons.]]></description>
      <pubDate>Wed, 25 Mar 2026 11:44:46 GMT</pubDate>
      <guid>https://trid.trb.org/View/2682224</guid>
    </item>
    <item>
      <title>Time-varying ship-radiated noise generation via flow matching</title>
      <link>https://trid.trb.org/View/2682791</link>
      <description><![CDATA[Ship-radiated noise generation has witnessed notable advancements in performance, particularly with the advent of neural network models. Most previous approaches generate ship-radiated noise relying on simulating the physical vibrations of ship components, which were conventionally assumed to remain stable under the ship type and operating condition. However, the noise emitted by real-world ships exhibits inherent temporal non-stationarity, changing dynamically over time even under invariant operating conditions. The modeling and evaluation of such time-varying characteristics remain largely unexplored. To fill this gap, this paper presents three key contributions. First, a flow matching-based generative model is proposed for synthesizing time-varying ship-radiated noise by learning smooth probability trajectories in a continuous latent space. Unlike traditional physics-driven approaches, the proposed method generates realistic acoustic waveforms using only reference audio clips together with LOFAR and DEMON prompts. Second, an evaluation strategy is introduced that measures long-term temporal fluctuations by comparing the distributional evolution of low-frequency features between real and generated signals. Finally, experiments on the DeepShip dataset demonstrate that with as few as 16 function evaluations, the proposed method synthesizes perceptually realistic ship-radiated noise, and closely matches the time-varying characteristics of real-world recordings.]]></description>
      <pubDate>Wed, 25 Mar 2026 11:44:46 GMT</pubDate>
      <guid>https://trid.trb.org/View/2682791</guid>
    </item>
    <item>
      <title>Increasing diffusiveness in cavitation tunnel with Schroeder diffusers for improved sound power estimate of directional sources at high frequency</title>
      <link>https://trid.trb.org/View/2634013</link>
      <description><![CDATA[This study focuses on the measurement of underwater radiated noise at high frequencies in a test section of a cavitation tunnel. These measurements, performed to predict the noise source level of a ship at the full scale, are carried out in a test section of a cavitation tunnel with acoustic pressure sensors. The directivity of the sources at a model-scale is one of the causes of the inaccuracy and high deviation that we can observe. We investigate by theory and simulations the use of Schroeder’s diffusers at wall location. The results obtained show an increase of the diffusiveness in the test section. The analysis of the diffusion coefficient and the acoustic level distributions show that a directional source could be considered as an omnidirectional source within a given bandwidth. Further studies are needed to extend this bandwidth with other designs or concepts.]]></description>
      <pubDate>Tue, 10 Mar 2026 09:55:43 GMT</pubDate>
      <guid>https://trid.trb.org/View/2634013</guid>
    </item>
    <item>
      <title>Standardized underwater measurements of a ferry’s source levels in the St. Lawrence’s shallow coastal environment</title>
      <link>https://trid.trb.org/View/2667440</link>
      <description><![CDATA[Accurate characterization of ship source levels is essential to assess the acoustic footprint of maritime traffic in biologically sensitive regions. This study presents standardized underwater radiated noise (URN) measurements for the N.M. Saaremaa I, a diesel-electric ferry scheduled to operate in the St. Lawrence Estuary, a critical habitat for endangered belugas. Using two seabed-mounted digital hydrophones and a shallow-water measurement protocol adapted from recent standards, we derived monopole source levels (mSLs) across 37 decidecade frequency bands under various operating conditions, including constant-speed transits, acceleration runs, and idle states. Our results show a consistent increase in mSLs with vessel speed, reaching slopes of 2–4 dBkn⁻¹, with strongest correlations above 500  Hz. Acceleration events produced band emissions exceeding 180  dB re 1  µPa²   ·    m²  at 25  Hz, while mid-to-high frequency levels ( ≳  1  kHz) were often lower than during steady-speed operation. These findings highlight the dual nature of acceleration noise, potentially degrading acoustic habitat for baleen whales while increasing collision risk for odontocetes near terminals. This work provides one of the first in-situ acoustic profiles of a ferry ahead of service in a high-use marine mammal corridor and offers empirical input for noise modelling, regulatory assessments, and mitigation planning in shallow coastal environments.]]></description>
      <pubDate>Wed, 25 Feb 2026 14:28:19 GMT</pubDate>
      <guid>https://trid.trb.org/View/2667440</guid>
    </item>
  </channel>
</rss>