Article
Acoustics
Yanfeng Lang, Zhibo Yang, Xuefeng Chen
Summary: Amplitude and phase characteristics are widely used in Lamb wave imaging, but they are dependent on the baseline and affected by artifacts. Instantaneous frequency, although often neglected, is also an important source of damage information. A dispersive instantaneous frequency imaging algorithm is proposed to suppress the undesired mode and eliminate baseline dependence.
JOURNAL OF SOUND AND VIBRATION
(2023)
Article
Acoustics
Dingcheng Ji, Fei Gao, Jiadong Hua, Jing Lin
Summary: In this paper, a novel approach using reflections and diffractions of Lamb waves was proposed to identify the crack location and size. The interaction mechanism between the crack and Lamb wave was thoroughly analyzed, revealing that both reflections and diffractions carry valuable damage information. The signals were classified into two groups using a threshold, and an overcomplete dictionary of waveforms was constructed to extract the damage information based on different propagation distances.
Article
Multidisciplinary Sciences
Kwanchai Pakoksung, Anawat Suppasri, Fumihiko Imamura
Summary: This study analyzed the atmospheric wave and tsunami triggered by the eruption of the Hunga Tonga-Hunga Ha'apai volcano in Tonga. Frequency characteristics and modal analysis were conducted using observation data, revealing the arrival time and dominant periods of the Lamb waves and tsunamis. The findings provide valuable insights and information on the nonseismic and far-field effects of tsunamis generated by volcanic eruptions.
SCIENTIFIC REPORTS
(2023)
Article
Engineering, Mechanical
Pan Zhang, Shaoguang Li, Alfredo Nunez, Zili Li
Summary: This paper presents a solution method based on finite element modeling to predict multimodal dispersive waves in a free rail. Modal behaviors, wavenumber-frequency dispersion relations, and group velocities of six types of propagative waves are derived and discussed in detail in the frequency range of 0-5 kHz. The ODS measurement approach and SMAW approach are proposed to experimentally study wave propagation and dispersion characteristics.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2021)
Article
Materials Science, Characterization & Testing
Jiadong Hua
Summary: Lamb wave is a promising tool for structural health monitoring and nondestructive testing. Sparse decomposition is proposed to address the overlap problem caused by dispersion and multi-mode characteristics. Broadband Lamb wave sparse decomposition is proposed to overcome the limitation of predicting propagating wave packets with severe amplitude modulation and velocity dispersion. Tone burst models are used as dictionary atoms and non-zero values in the time-frequency matrix are extracted for wave packet reconstruction.
JOURNAL OF NONDESTRUCTIVE EVALUATION
(2023)
Article
Engineering, Mechanical
A. Nokhbatolfoghahai, H. M. Navazi, R. M. Groves
Summary: This paper proposes an adaptive dictionary learning framework to enhance the performance of the sparse reconstruction method in real complex engineering structures. The framework combines analytical modeling with training data sets and learning methods. Experimental evaluation on an anisotropic composite plate showed the potential of the framework for improved health monitoring of complex structures.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2022)
Article
Mechanics
Alvaro Gonzalez-Jimenez, Luca Lomazzi, Francesco Cadini, Alessio Beligni, Claudio Sbarufatti, Marco Giglio, Andrea Manes
Summary: Ultrasonic guided waves-based Structural Health Monitoring (SHM) is a promising solution for damage diagnosis in plate-like structures, with the Reconstruction Algorithm for Probabilistic Inspection (RAPID) being a widely used tomographic algorithm for active damage detection and localization. However, the algorithm may produce artefacts due to non-uniform distributions of the sensing network density. A proposed processing and spatial filtering technique in this article effectively mitigates this issue, validated using a numerical database for a large carbon fiber reinforced polymer (CFRP) with a through-the-thickness hole.
COMPOSITE STRUCTURES
(2021)
Article
Instruments & Instrumentation
Wen Qiu, Lei Xu, Yaozhong Liao, Qiao Bao, Qiang Wang, Zhongqing Su
Summary: This study proposes a sparse, triangle-shaped sensor array to identify, orient, and assess the degree of structural damage in composite constructions. The damage-scattered Lamb waves are recorded by the sparse sensor array and then fed into the SVM classification method. The location and severity of the damage can be determined by training the SVM model.
SMART MATERIALS AND STRUCTURES
(2023)
Article
Chemistry, Multidisciplinary
Alex Vu, Leonard J. Bond, Sunil K. Chakrapani
Summary: This study numerically investigates the mode conversion between fundamental Lamb and Rayleigh waves in quarter and half spaces. Using finite element analysis, the propagation of fundamental Lamb waves in a plate attached to a quarter space and the subsequent mode conversion to Rayleigh waves are studied. B-Scans show beat-like phenomenon for the R -> L conversion and generation length for the L -> R conversion. The study confirms that grazing incidence of bulk waves and scattering/diffraction play important roles in the mode conversion phenomenon.
APPLIED SCIENCES-BASEL
(2023)
Article
Acoustics
Shiqian Chen, Kaiyun Wang, Zhike Peng, Chao Chang, Wanming Zhai
Summary: A generalized dispersive mode decomposition (GDMD) method is proposed in this paper to accurately estimate GDs and fully separate overlapped modes of dispersive signals. By formulating the mode decomposition issue as an optimal dispersion compensation problem based on a defined model, the dispersion effects of all the modes can be fully eliminated, resulting in a high-quality TF distribution.
JOURNAL OF SOUND AND VIBRATION
(2021)
Article
Materials Science, Composites
Asif Khan, Heung Soo Kim
Summary: This article proposes a framework using Lamb waves and unsupervised autonomous features for the damage assessment and effect of temperature variations in laminated composites. The extracted features are confirmed to have discriminative capabilities through supervised and unsupervised machine learning. Through visualization and damage localization, physically consistent results are revealed in the feature space.
JOURNAL OF THERMOPLASTIC COMPOSITE MATERIALS
(2023)
Article
Multidisciplinary Sciences
Robin S. Matoza, David Fee, Jelle D. Assink, Alexandra M. Iezzi, David N. Green, Keehoon Kim, Liam Toney, Thomas Lecocq, Siddharth Krishnamoorthy, Jean-Marie Lalande, Kiwamu Nishida, Kent L. Gee, Matthew M. Haney, Hugo D. Ortiz, Quentin Brissaud, Leo Martire, Lucie Rolland, Panagiotis Vergados, Alexandra Nippress, Junghyun Park, Shahar Shani-Kadmiel, Alex Witsil, Stephen Arrowsmith, Corentin Caudron, Shingo Watada, Anna B. Perttu, Benoit Taisne, Pierrick Mialle, Alexis Le Pichon, Julien Vergoz, Patrick Hupe, Philip S. Blom, Roger Waxler, Silvio De Angelis, Jonathan B. Snively, Adam T. Ringler, Robert E. Anthony, Arthur D. Jolly, Geoff Kilgour, Gil Averbuch, Maurizio Ripepe, Mie Ichihara, Alejandra Arciniega-Ceballos, Elvira Astafyeva, Lars Ceranna, Sandrine Cevuard, Il-Young Che, Rodrigo De Negri, Carl W. Ebeling, Laslo G. Evers, Luis E. Franco-Marin, Thomas B. Gabrielson, Katrin Hafner, R. Giles Harrison, Attila Komjathy, Giorgio Lacanna, John Lyons, Kenneth A. Macpherson, Emanuele Marchetti, Kathleen F. McKee, Robert J. Mellors, Gerardo Mendo-Perez, T. Dylan Mikesell, Edhah Munaibari, Mayra Oyola-Merced, Iseul Park, Christoph Pilger, Cristina Ramos, Mario C. Ruiz, Roberto Sabatini, Hans F. Schwaiger, Dorianne Tailpied, Carrick Talmadge, Jerome Vidot, Jeremy Webster, David C. Wilson
Summary: The eruption of the Hunga volcano in Tonga on 15 January 2022 caused an explosion in the atmosphere, generating a range of atmospheric waves that were observed globally. The eruption produced significant infrasound, audible sound, and ionospheric perturbations, and contributed to the occurrence of tsunamis. The exceptional observations of the atmospheric waves are highlighted in this study.
Article
Multidisciplinary Sciences
Hang Fan, Fei Gao, Wenhao Li, Kun Zhang
Summary: Traditional Lamb wave inspection methods are limited by the availability of dispersion curves and baseline recordings. In this study, a two-step strategy utilizing adaptive multiple signal classification (MUSIC) and sparse reconstruction is proposed to overcome this limitation. The method reconstructs the multimodal Lamb waves in the f-k domain using random measurements, computes the phase and group velocities, and establishes steering vectors for potential scattering Lamb waves. An adaptive window is used to extract local wave components, highlighting scattering features. Experiment and simulation results demonstrate the accuracy and resolution of damage localization without prior dispersion data and baseline recordings.
Article
Chemistry, Physical
Frank H. G. Stolze, Keith Worden, Graeme Manson, Wieslaw J. Staszewski
Summary: The use of a diffused Lamb wave field is a challenging yet effective method for fatigue-crack detection in multi-riveted strap-joint aircraft panels. The panel is equipped with low-profile surface-bonded piezoceramic transducers, which extract information on fatigue damage through various amplitude characteristics of Lamb waves. Statistical outlier analysis is performed to detect damage, and simplified wave scattering modeling is used to explain complex response features. The results demonstrate the potential and limitations of this method for reliable fatigue-crack detection in complex aircraft components.
Article
Mathematics, Applied
Barbara Zupancic, Yulia Prokop, Anatolij Nikonov
Summary: This paper presents a newly developed numerical engineering approach for analyzing elastic wave dispersion in composite plates with high-contrast properties of the layers. The results show that the developed computational methodology accurately captures the complexity of the dispersion phenomena and provides valuable insights into the frequency range in which vibration modes can be activated.
FINITE ELEMENTS IN ANALYSIS AND DESIGN
(2021)
Article
Engineering, Electrical & Electronic
Ayobami S. Edun, Naveen Kumar Tumkur Jayakumar, Samuel R. Kingston, Cynthia M. Furse, Michael A. Scarpulla, Joel B. Harley
Summary: Spread spectrum time domain reflectometry (SSTDR) has traditionally been used for detecting hard faults in transmission lines, with little research on impedance issues of circuit elements in the middle of the line. This study considers transmission lines with different impedances on each wire and provides accurate analytical expressions for reflection coefficients.
IEEE SENSORS JOURNAL
(2021)
Biographical-Item
Engineering, Electrical & Electronic
Hagit Messer, Arye Nehorai, Jeff Krolik, Jose M. F. Moura, Al Hero, Joseph Tabrikian
IEEE SIGNAL PROCESSING MAGAZINE
(2021)
Article
Engineering, Electrical & Electronic
Athina Petropulu, Jose M. F. Moura, Rabab Kreidieh Ward, Theresa Argiropoulos
Summary: Signal processing (SP) is a powerful technology that has significantly impacted the digital world and brought about major changes in our lives. The growth of digital SP since the mid-1960s, supported by integrated circuits and digital computers, has led to groundbreaking advances in various fields, profoundly influencing society. The IEEE Signal Processing Society (SPS), as the leading professional society, has played a crucial role in advancing the theory and applications of SP through its publications, conferences, and educational activities. It has fostered collaboration and knowledge sharing among researchers, practitioners, and students, empowering the growth of SP.
IEEE SIGNAL PROCESSING MAGAZINE
(2023)
Editorial Material
Engineering, Electrical & Electronic
Geert Leus, Antonio G. Marques, Jose M. F. Moura, Antonio Ortega, David Shuman
IEEE SIGNAL PROCESSING MAGAZINE
(2023)
Article
Engineering, Electrical & Electronic
Stefan Vlaski, Soummya Kar, Ali H. Sayed, Jose M. F. Moura
Summary: This article reviews the significant advances in networked signal and information processing (SIP) that have allowed decision making, inference, optimization, control, and learning to be extended to distributed agent environments. As interacting agents cooperate, new collective behaviors emerge from their local decisions and actions. Furthermore, networked agents have been shown to match the performance of cloud or federated solutions while offering improved privacy, increased resilience, and conserved resources through cooperation and sharing.
IEEE SIGNAL PROCESSING MAGAZINE
(2023)
Proceedings Paper
Computer Science, Artificial Intelligence
Wei Li, Fangzhou Wang, Jose M. F. Moura, R. D. Blanton
Summary: This study proposes a method of modeling global floorplanning as an SDP problem and solves the rank constraint problem by introducing a direction matrix, which improves the quality of the solution. Furthermore, a series of techniques are introduced to enhance the flexibility, accuracy, and efficiency of the algorithm.
2023 60TH ACM/IEEE DESIGN AUTOMATION CONFERENCE, DAC
(2023)
Proceedings Paper
Computer Science, Artificial Intelligence
John Shi, Jose M. F. Moura
Summary: In both Discrete Signal Processing (DSP) and Graph Signal Processing (GSP), sampling is a crucial step for signal reconstruction. This paper presents a unified GSP sampling theory that bridges the gap between the vertex and spectral domains, similar to DSP sampling. By exploring the steps of DSP sampling in GSP and considering the impact of different sampling choices, the paper provides new insights and intuition on both DSP and GSP sampling. It also highlights the limitations of current spectral sampling methods in the GSP literature.
2022 56TH ASILOMAR CONFERENCE ON SIGNALS, SYSTEMS, AND COMPUTERS
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Shreyas Chaudhari, Jose M. F. Moura
Summary: Direction of arrival (DoA) estimation is a well-studied problem with various applications. Traditional algorithms require prior knowledge of the number of transmitters or sufficient measurements, which limits their performance. Recently, a deep learning approach has been proposed, but it still needs prior knowledge or a large number of snapshots. We propose a new deep learning approach that can determine the number and positions of transmitters, outperforming traditional and recent deep learning methods in low-SNR and low-snapshot scenarios.
2022 56TH ASILOMAR CONFERENCE ON SIGNALS, SYSTEMS, AND COMPUTERS
(2022)
Article
Engineering, Electrical & Electronic
John Shi, Jose M. F. Moura
Summary: Vertex based and spectral based GSP sampling methods have been studied recently. This paper introduces a unified graph signal sampling theory that starts from the spectral domain and includes dual versions in both the vertex and spectral domains. The theory shows how GSP sampling reduces to DSP sampling when the graph is a directed time cycle graph. Simple examples illustrate the impact of choices available in GSP sampling.
IEEE TRANSACTIONS ON SIGNAL PROCESSING
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Rajshekhar Das, Yu-Xiong Wang, Jose M. F. Moura
Summary: Few-shot classification aims to classify categories of a novel task by learning from just a few labelled examples, and a new finetuning approach based on contrastive learning can significantly boost few-shot generalization.
2021 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2021)
(2021)
Proceedings Paper
Acoustics
Shreyas Chaudhari, Harideep Nair, Jose M. F. Moura, John Paul Shen
Summary: The study introduces a neuromorphic approach for unsupervised time series clustering, emphasizing ultra-low power and continuous online learning for edge devices. Performance evaluation on a subset of UCR Time Series Archive datasets shows that the proposed method outperforms or compares similarly to existing algorithms, while being more amenable for efficient hardware implementation.
2021 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP 2021)
(2021)
Proceedings Paper
Computer Science, Theory & Methods
Alireza Chamanzar, Xujin Liu, Lavender Y. Jiang, Kimon A. Vogt, Jose M. F. Moura, Pulkit Grover
Summary: The study presents a non-invasive deep learning approach for tracking cortical spreading depressions (CSDs) in scalp electroencephalography (EEG) signals. The method, named CSD spatially aware convolutional network or CSD-SpArC, combines convolutional neural network and graph neural network to extract temporal features and spatial structure of EEG signals, achieving high accuracy in tracking CSDs.
2021 10TH INTERNATIONAL IEEE/EMBS CONFERENCE ON NEURAL ENGINEERING (NER)
(2021)
Article
Engineering, Electrical & Electronic
Mark Cheung, John Shi, Oren Wright, Lavendar Y. Jiang, Xujin Liu, Jose M. F. Moura
IEEE SIGNAL PROCESSING MAGAZINE
(2020)
Article
Engineering, Electrical & Electronic
Dusan Jakovetic, Dragana Bajovic, Joao Xavier, Jose M. F. Moura
PROCEEDINGS OF THE IEEE
(2020)
Article
Engineering, Electrical & Electronic
Joao Domingos, Jose M. F. Moura
IEEE TRANSACTIONS ON SIGNAL PROCESSING
(2020)