Guided wave-based structural health monitoring is extensively utilized in various industrial applications to ensure the integrity of components within industrial systems. Among these monitoring techniques, principal component analysis (PCA) reconstruction methods are widely used for anomaly detection due to their computational efficiency and interoperability. However, existing PCA reconstruction methods are semisupervised anomaly detection approaches that require training on historical normal data and fail to detect anomalous signals within the training set. To address this limitation, this work proposes a correntropy PCA (C-PCA), enabling fully unsupervised anomaly detection on raw training data without requiring label information, when the dataset contains a high proportion of abnormal signals. This method allows anomaly detection on real-time measurements without the need for precleaned historical normal data or can also be used to generate clean data for existing semisupervised anomaly detection methods. In correntropy PCA, principal components are extracted from the correntropy matrix rather than the correlation matrix. The correntropy, representing the statistical dependence between samples of guided waves, is estimated utilizing a Gaussian kernel with a specified kernel width. Through the optimization of the kernel width, the correntropy PCA reconstruction method demonstrates superior anomaly detection performance compared with the standard PCA reconstruction method, especially in scenarios where training data are contaminated by a significant proportion of abnormal signals. Guidelines for the optimization of the kernel width are provided. The effectiveness of the correntropy PCA reconstruction-based anomaly detection method is validated using data collected from ten regions over an 80-day period, encompassing guided waves induced by damage occurring over durations ranging from 2 to 20 days.

Understanding and quantifying ecosystem services are crucial for sustainable environmental management, conservation efforts, and policy-making. The advancement of remote sensing technology and machine learning techniques has greatly facilitated this process. Yet, ground truth labels, such as biodiversity, are very difficult and expensive to measure. In addition, more easily obtainable proxy labels, such as land use, often fail to capture the complex heterogeneity of the ecosystem. In this paper, we demonstrate how land use proxy labels can be implemented with a soft, multi-label classifier to predict ecosystem services with complex heterogeneity.

Guided wave testing is a popular approach for monitoring the structural integrity of infrastructures. We focus on the primary task of damage detection, where signal processing techniques are commonly employed. The detection performance is affected by a mismatch between the wave propagation model and experimental wave data. External variations, such as temperature, which are difficult to model, also affect the performance. While deep learning models can be an alternative detection method, there is often a lack of real-world training datasets. In this work, we counter this challenge by training an ensemble of variational autoencoders only on simulation data with a wave physics-guided adversarial component. We set up an experiment with non-uniform temperature variations to test the robustness of the methods. We compare our scheme with existing deep learning detection schemes and observe superior performance on experimental data.

Ultrasonic guided waves are an attractive tool for structural health monitoring due to their capability to rapidly assess large regions of a structure. Yet, most guided wave based methods for detecting, locating, and classifying structural damage rely on our ability to accurately predict guided wave behavior. Characterizing and predicting guided wave behavior is difficult, particularly in mechanically complex materials such as fiber-reinforced composites. In this paper, we address this challenge through a sparse wavenumber analysis framework. Sparse wavenumber analysis integrates physics-based models, signal processing algorithms for compressive sensing, and a small number of local measurements to predict global wave behavior. We implement sparse wavenumber analysis for three wave systems: standing waves on a string, Lamb waves in an isotropic plate, and guided waves in a unidirectional, anisotropic plate. Through the use of simulation and experimental data, we show that sparse wavenumber analysis can accurately recover the sparse representations (i.e., the eigenmodes) of each system and then use these representations to predict global wave behavior. For the anisotropic plate, we accurately predict 149765 experimental time-domain measurements from only 36 local measurements.

One of the largest challenges for multichannel localization systems is developing methodologies that are robust to interference. Unlike noise, interference is not random and often has characteristics resembling the true signals of interest. Interference often originates from multipath propagation, jamming signals, or other sources. In this paper, we demonstrate that we can significantly improve localization performance in the presence of interference through the use of a random sensor topology and matched field processing. To show this, we apply concepts and results from random matrix theory and compressed sensing. We demonstrate theoretically that random sensor topologies allow us to achieve performance characteristics similar to those of random noise. Specifically, we show that the localization performance improves, with a high probability, at a rate proportional to the number of sensors in the system. We verify these results through simulation.

We present a multiresolution classification framework with semi-supervised learning for the indirect structural health monitoring of bridges. The monitoring approach envisions a sensing system embedded into a moving vehicle traveling across the bridge of interest to measure the modal characteristics of the bridge. To enhance the reliability of the sensing system, we use a semi-supervised learning algorithm and a semi-supervised weighting algorithm within a multiresolution classification framework. We show that the proposed algorithm performs significantly better than supervised multiresolution classification.

Matched field processing is a powerful tool for accurately localizing targets in dispersive media. However, matched field processing requires a precise model of the medium under test. In underwater acoustics, where matched field processing has been extensively studied, authors often resort to extremely detailed numerical models of the propagation medium, which are computationally expensive and impractical for many applications. As an alternative, this paper uses convex sparse recovery techniques to construct, directly from measured data, an accurate model of a plate medium based on its dispersion characteristics. From this data-driven model, the Green's function between two points can be readily predicted. We demonstrate the effectiveness of this model by localizing a source in a dispersive plate medium. The results visually illustrate our approach to significantly improve localization accuracy and reduce artifacts when compared to a conventional narrowband technique.

Detection of targets in complex environments is of importance in both radar and sonar applications. Recent work has shown that the use of Time Reversal (TR) techniques improves the performance of systems operating in deterministic channels with a significant multipath return. This paper extends those results to stationary random channels with significant multipath. We develop a TR-based approach and derive the Likelihood Ratio Test (LRT) for this approach. We compare this TR-LRT to an LRT derived through a “water filling” approach. We derive theoretical performance curves for the water filling LRT, and evaluate both the water filling and TR detectors with Monte Carlo simulations. For the scenarios tested, we show that TR achieves an SNR gain of 1–2dB over the water filling detector.

This paper is concerned with a moving target detection using time reversal in dense multipath environments. We show that the Doppler shift in the time reversal re-transmission simplifies the detector design, yet still achieves the focusing effect. Thus, the Doppler diversity is utilized to achieve high target detectability by time reversal.