This paper reviews key signal processing techniques in structural health monitoring (SHM), focusing on non-parametric time–frequency analysis, adaptive decomposition, and deconvolution methods. It examines the short-time Fourier transform (STFT), wavelet transform (WT), and Wigner–Ville distribution (WVD), highlighting their applications, advantages, and limitations in SHM. The review also explores adaptive techniques like empirical mode decomposition (EMD) and its variants (EEMD, MEEMD), as well as variational mode decomposition (VMD) and its improved versions (SVMD, AVMD), emphasizing their effectiveness in handling nonlinear and non-stationary signals. Additionally, deconvolution methods such as minimum entropy deconvolution (MED) and maximum correlated kurtosis deconvolution (MCKD) are discussed for mechanical fault diagnosis. The paper aims to provide a comprehensive overview of these techniques, offering insights for future research into SHM signal processing.

Gait authentication, especially sensor-based patterns, has been studied by researchers for decades. Nowadays, gait authentication has become an important facet of biometric systems due to the so-called unique characteristics of each user. With the development of various technologies (i.e., hardware, data processing, features extraction, and learning algorithms), the performance of sensor-based authentication methods is gradually improving. But we have found that the vulnerability of most existing methods can be compromised easily. In this paper, we propose a novel attack model, called one cycle attack, to bypass existing gait authentication methods. Firstly, the gait sequence is divided into multiple gait cycles. By adopting the K-mean algorithm, we get the average distance of each feature sample (extracted from the gait cycle) to its closest cluster center, and its result confirms that independent individuals may have similar gait cycles. Secondly, using six state-of-the-art models it was found that the adversarial gait cycle found with the clustering method can bypass the victim’s model rapidly. Furthermore, to improve the accuracy of sensor-based gait authentication methods to fight against attacks, we present a WPD-LSTM (Wavelet Packet Decomposition and Long Short-Term Memory) multi-cycle defense model which considers the contextual contents of the neighboring gait cycles in the gait sequence. Experimental results on two datasets (the largest public sensor-based gait database OU-ISIR and new dataset from our laboratory) show that our attack model can bypass most of the victims’ models within a limited number of attempts. Specifically, we can compromise 20%-80% of users within 5 attempts by utilizing imitation. On the contrary, the success rate of attackers has been greatly mitigated by deploying our multi-cycle defense model.

Deep learning has made important contributions to classification tasks applied to fault diagnosis. However, it is crucial to integrate the technologies into real industrial applications through cost-effective hardware. Edge computing, a new computing paradigm, has the potential to accelerate system response time, reduce bandwidth for transmission, and use fewer computing resources. In this article, the distillation quantization compression method based on energy entropy is applied to compress the convolutional neural network (CNN), which is deployed on a Cortex-M4 series microcontroller with only 192 KB of RAM and 512 KB of ROM. Additionally, based on the fault mechanism of rolling bearings, this article integrates the attention mechanism and envelope spectrum to verify the effectiveness of feature extraction by the CNN model, which effectively weaken invalid features in distillation quantization process. The experimental results show that the proposed method has excellent performance in terms of memory overhead and inference speed, which has great potential in industrial applications.

Photovoltaic (PV) power generation is greatly affected by meteorological environmental factors, with obvious fluctuations and intermittencies. The large-scale PV power generation grid connection has an impact on the source-load stability of the large power grid. To scientifically and rationally formulate the power dispatching plan, it is necessary to realize the PV output prediction. The output prediction of single power plants is no longer applicable to large-scale power dispatching. Therefore, the demand for the PV output prediction of multiple power plants in an entire region is becoming increasingly important. In view of the drawbacks of the traditional regional PV output prediction methods, which divide a region into sub-regions based on geographical locations and determine representative power plants according to the correlation coefficient, this paper proposes a multilevel spatial upscaling regional PV output prediction algorithm. Firstly, the sub-region division is realized by an empirical orthogonal function (EOF) decomposition and hierarchical clustering. Secondly, a representative power plant selection model is established based on the minimum redundancy maximum relevance (mRMR) criterion. Finally, the PV output prediction for the entire region is achieved through the output prediction of representative power plants of the sub-regions by utilizing the Elman neural network. The results from a case study show that, compared with traditional methods, the proposed prediction method reduces the normalized mean absolute error (nMAE) by 4.68% and the normalized root mean square error (nRMSE) by 5.65%, thereby effectively improving the prediction accuracy.

Power quality disturbances (PQDs) have a large negative impact on electric power systems with the increasing use of sensitive electrical loads. This paper presents a novel hybrid algorithm for PQD detection and classification. The proposed method is constructed while using the following main steps: computer simulation of PQD signals, signal decomposition, feature extraction, heuristic selection of feature selection, and classification. First, different types of PQD signals are generated by computer simulation. Second, variational mode decomposition (VMD) is used to decompose the signals into several instinct mode functions (IMFs). Third, the statistical features are calculated in the time series for each IMF. Next, a two-stage feature selection method is imported to eliminate the redundant features by utilizing permutation entropy and the Fisher score algorithm. Finally, the selected feature vectors are fed into a multiclass support vector machine (SVM) model to classify the PQDs. Several experimental investigations are performed to verify the performance and effectiveness of the proposed method in a noisy environment. Moreover, the results demonstrate that the start and end points of the PQD can be efficiently detected.

With the popularity of smartphones and the development of hardware, mobile devices are widely used by people. To ensure availability and security, how to protect private data in mobile devices without disturbing users has become a key issue. Mobile user authentication methods based on motion sensors have been proposed by many works, but the existing methods have a series of problems such as poor de-noising ability, insufficient availability, and low coverage of feature extraction. Based on the shortcomings of existing methods, this paper proposes a hybrid deep learning system for complex real-world mobile authentication. The system includes: (1) a variational mode decomposition (VMD) based de-noising method to enhance the singular value of sensors, such as discontinuities and mutations, and increase the extraction range of the feature; (2) semi-supervised collaborative training (Tri-Training) methods to effectively deal with mislabeling problems in complex real-world situations; and (3) a combined convolutional neural network (CNN) and support vector machine (SVM) model for effective hybrid feature extraction and training. The training results under large-scale, real-world data show that the proposed system can achieve 95.01% authentication accuracy, and the effect is better than the existing frontier methods.