Enhancing Industrial Wireless Communication Security Using Deep Learning Architecture‐Based Channel Frequency Response

Physical layer security (PLS) in wireless communication systems has attracted extensive research attentions in recent years. Unlike cryptography-based methods applied in upper-layer in network, PLS methods are applied in physical layers and can provide information-theoretic security by utilizing the randomness of signals and wireless channels. In this survey, we provide a comprehensive review in the domain of physical layer authentication (PLA) in wireless communication systems, including the concepts, several key techniques of typical PLA architectures as well as future challenges and research trends in more sophisticated communication systems. The survey begins with an overview of the background and basic concepts of PLA, such as the general model of wireless security communication system, typical frameworks of key-based/less PLA systems, and the common attack models. We then discuss the major concerns and key techniques that are applied in PLA systems, where three types of authentication schemes are considered, i.e., the authentication based on channel information, radio-frequency and identity watermarks. Basic models and representative research results about key approaches and techniques applied to the authentication systems above are subsequently covered. Finally, the associated challenges and potential research trends of PLA in future communication systems are presented at the end of the survey paper.

In online courses, discussion forums play a key role in enhancing student interaction with peers and instructors. Due to large enrolment sizes, instructors often struggle to respond to students in a timely manner. To address this problem, both traditional Machine Learning (ML) (e.g., Random Forest) and Deep Learning (DL) approaches have been applied to classify educational forum posts (e.g., those required urgent responses vs. that did not). However, there lacks an in-depth comparison between these two kinds of approaches. To better guide people to select an appropriate model, we aimed at providing a comparative study on the effectiveness of six frequently-used traditional ML and DL models across a total of seven different classification tasks centering around two datasets of educational forum posts. Through extensive evaluation, we showed that (i) the up-to-date DL approaches did not necessarily outperform traditional ML approaches; (ii) the performance gap between the two kinds of approaches can be up to 3.68% (measured in F1 score); (iii) the traditional ML approaches should be equipped with carefully-designed features, especially those of common importance across different classification tasks. Based on the derived findings, we further provided insights to help instructors and educators construct effective classifiers for characterizing educational forum discussions, which, ultimately, would enable them to provide students with timely and personalized learning support.

In the research of hyperspectral image classification, deep learning-based methods have achieved promising results. However, due to the high cost and difficulty in obtaining labeled hyperspectral image samples, traditional deep learning methods experience a significant performance decline under few-shot conditions. This paper proposes a deep few-shot learning method, referred to as Wavelet-Based Multi-stream Spatial-Frequency Fusion Network, to address the issue of small sample size in hyperspectral image classification. Through a multi-stream structure, features at various scales are obtained, effectively alleviating the issue of insufficient feature representation caused by limited training samples. These features are then fed into a multi-level wavelet transformation module integrated with a frequency-domain attention module. By decomposing the input into high-frequency and low-frequency components via wavelet transformation, and subsequently inputting these components into the frequency-domain attention module, a richer representation of features in the frequency domain is achieved. Finally, a Spatial-Frequency Fusion module is employed to not only extract abundant feature information but also fuse features from different scales. This process results in feature maps that encompass more comprehensive global feature information, thereby significantly enhancing classification accuracy under small sample conditions. The proposed method has been compared with traditional machine learning approaches and recent deep learning methods on the Houston, Dioni, and University of Pavia datasets. Under training conditions using only 1 to 5, 10, 15, and 20 samples per class, the experimental results demonstrate the superiority of the method proposed in this paper.

Trealize the Industry 4.0 vision and enable mobile connectivity and flexible deployment in harsh industrial environments, wireless communication is essential. But before wireless communications technology can be widely deployed for critical control applications, first it must be assessed, and that requires a comprehensive characterization of the wireless channel. This can be done by analyzing large amounts of wireless data collected from different industrial environments. In this article, we discuss the possibilities offered by a recently published industrial wireless data set. This data set is more exhaustive than measurements previously reported. We show two cases of how those data have been applied to improve latency performance and to investigate the feasibility of physical-layer security techniques for wireless communication in industrial environments.

Detecting major depressive disorder presence using passively-collected wearable movement data in a nationally-representative sample

Industrial control systems rely on wireless devices and sensors, necessitating critical security. Physical layer authentication (PLA) is a promising mechanism for device authentication, utilizing its unique spatiotemporal characteristics and channel state randomness, which offers unforgeability and high informatics security with low computational overhead and efficiency in resource-constrained scenarios. However, existing PLA mechanisms face challenges in complex industrial wireless environments, including insufficient accuracy, computational complexity, inadequate noise consideration, and poor performance. To address these challenges, we propose a convolutional denoising autoencoder (CDAE) model that reduces feature dimensions, eliminates noise, and extracts key vectors. The weighted <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$k$ </tex-math></inline-formula> -nearest neighbor algorithm classifies the extracted vectors for comprehensive authentication in control system networks. Accurate authentication enables efficient detection of malicious attacks. Simulation experiments show that using CDAE-extracted feature vectors achieves over 95% accuracy with only 1% training samples, surpassing channel state information-based authentication by 46.15%, validating the proposed mechanism’s effectiveness.

The rise of social media platforms has led to an increase in the flow and dissemination of information, but it has also made generating and spreading rumors easier. Rumor detection requires understanding the context and semantics of text, dealing with the evolving nature of rumors, and processing vast amounts of data in real-time. Deep learning (DL)-based techniques exhibit a higher accuracy in detecting rumors on social media compared to many traditional machine learning approaches. This study presents a systematic review of DL approaches in rumor detection, analyzing datasets, pre-processing methods, feature taxonomy, and frequently used DL methods. In the context of feature selection, we categorize features into three areas: text-based, user-based, and propagation-based. Besides, we surveyed the trends in DL models for rumor detection and classified them into convolutional neural networks (CNN), recurrent neural networks (RNN), graph neural networks (GNN), and other methods based on the model structure. It offers insights into effective algorithms and strategies, aiming to guide researchers, developers, social media users, and governments in detecting and preventing the spread of false information. The study contributes to enhancing research in this field and identifies potential areas for future exploration.

Small data, big challenges: Machine- and deep-learning strategies for data-limited drug discovery.

Early detection of lung cancer remains a critical challenge in medical imaging, where accurate classification between benign and malignant pulmonary nodules can significantly impact patient outcomes. This study presents a comprehensive comparative analysis of six distinct machine learning approaches for lung cancer classification using 3D computed tomography (CT) data. The methodologies evaluated include 2D convolutional neural networks (CNN), 3D CNN architectures, multi-slice processing techniques, and traditional machine learning algorithms including Random Forest and Logistic Regression. Our experimental framework utilized synthetic CT volumes designed to replicate characteristic patterns of benign and malignant lung lesions. The synthetic dataset comprised 120 volumetric samples with distinct spatial patterns representing clinical features such as nodule morphology, tissue density variations, and anatomical distribution. Results demonstrate that traditional machine learning approaches, particularly Random Forest classifiers, achieved superior performance with accuracy rates exceeding 85%, while deep learning models showed variable performance depending on architecture complexity and data preprocessing strategies. The findings suggest that careful consideration of model architecture selection and data representation is crucial for effective lung cancer classification systems.

Physical-Layer Authentication (PLA) attracts a lot of research interests because of its significant advantages over upper-layer authentication mechanisms: high security and low complexity. The PLA schemes can be categorized into passive and active schemes. In this paper, we extensively leverage the advantages of both the active and passive schemes as a reference scheme, named as the Direct Hybrid (DH) scheme. Although the DH scheme improves the authentication performance of the prior PLA schemes, it has limitations, e.g., high communication overhead. Then, we further propose two hybrid PLA schemes to overcome the limitations of the DH scheme. The first proposed scheme further uses the advantage of the Challenge-Response Authentication Mechanism (CRAM) scheme, named as the CR-based Hybrid (CRH) scheme. Although both DH and CRH schemes significantly improve the authentication performance of the prior PLA schemes, they do not address one significant limitation of the active scheme, i.e., to set the power allocation of a tag empirically. Thus, based on the CRH scheme, we further propose the Adaptive CR-Based Hybrid (ACRH) scheme to adaptively set the parameter instead of the empirical setting. Moreover, we provide the theoretical analysis of the proposed schemes over wireless fading channels and derive their closed-form expressions in terms of the Probability of Detection (PD), Probability of False Alarm (PFA), and optimal threshold, respectively. At last, we discuss the advantages and disadvantages of the proposed schemes and give some useful suggestions for seeking a better tradeoff. Our experimental results show that, in comparison with the active scheme, the DH scheme has better robustness, and the CRH scheme has better both robustness and compatibility but it sacrifices the security. The ARCH scheme achieves a better tradeoff than the remaining schemes.

The classification of clinical notes into specific diagnostic categories is critical in healthcare, especially for mental health conditions like anxiety and adjustment disorder. In this study, we compare the performance of various artificial intelligence models, including both traditional machine learning approaches (random forest, support vector machine, K-nearest neighbors, decision tree, and eXtreme Gradient Boost) and deep learning models (DistilBERT and SciBERT), to classify clinical notes into these two diagnoses. Additionally, we implemented three oversampling strategies: No Oversampling, Random Oversampling, and Synthetic Minority Over-sampling Technique (SMOTE), to assess their impact on model performance. Hyperparameter tuning was also applied to optimize model accuracy. Our results indicate that oversampling techniques had minimal impact on model performance overall. The only exception was SMOTE, which showed a positive effect specifically with Bidirectional Encoder Representations from Transformers (BERT)-based models. However, hyperparameter optimization significantly improved accuracy across the models, enhancing their ability to generalize and perform on the dataset. The decision tree and eXtreme Gradient Boost models achieved the highest accuracy among machine learning approaches, both reaching 96%, while the DistilBERT and SciBERT models also attained 96% accuracy in the deep learning category. These findings underscore the importance of hyperparameter tuning in maximizing model performance. This study contributes to the ongoing research on AI-assisted diagnostic tools in mental health by providing insights into the efficacy of different model architectures and data balancing methods.

Accurate prediction of small interfering RNA (siRNA) silencing efficiency is critical for accelerating nucleic acid drug development. Over the past two decades, the field has transitioned from empirical sequence-derived heuristics to data-driven methods such as deep learning and graph neural networks. In this review, we systematically discuss current research from three integrated perspectives: data resources, design rules, and computational algorithms. Firstly, we summarize strengths, biases, and gaps in existing datasets, including low-throughput saturation scanning data, high-throughput panels, and specialized repositories for chemically modified siRNAs. Next, we outline the evolution of siRNA design principles, from classical guidelines such as GC content windows, terminal thermodynamic asymmetry, and positional base preferences, to more sophisticated thermodynamic-structural features incorporating target mRNA accessibility, and finally, to advanced clinical GalNAc-siRNA strategies involving seed destabilization and stereochemically purified backbones. Regarding computational algorithms, the progression is categorized into three stages: early linear regression models, traditional machine learning approaches, and modern deep learning frameworks, alongside various specialized algorithms tailored for chemically modified siRNAs. Current studies demonstrate steadily improving prediction accuracy, yet significant challenges persist, including insufficient data coverage for disease-relevant targets and hard-to-transfect cells, limited paired measurements of bare versus modified sequences, and inadequate integration of off-target effects and immunogenicity metrics. Moving forward, there is an urgent need to establish highquality, dose- and time-resolved datasets across diverse tissues, embed biophysical priors into interpretable deep-learning architectures, and adopt multi-task modeling approaches to optimize efficiency, safety, and delivery simultaneously. Collectively, these advancements promise to propel siRNA design beyond liver-centric applications towards precision therapeutics targeting multiple tissues and diseases.

IntroductionAccurate classification of brain tumors from MRI scans is a critical task for improving patient outcomes. Machine learning (ML) and deep learning (DL) methods have shown promise in this domain, but their relative performance remains unclear.MethodsThis study evaluates several ML and DL techniques using the BraTS 2024 dataset. The models assessed include traditional algorithms such as Random Forest and advanced deep learning architectures including Simple CNN, VGG16, VGG19, ResNet50, Inception-ResNetV2, and EfficientNet. Preprocessing strategies were applied to optimize model performance.ResultsThe Random Forest classifier achieved the highest accuracy of 87%, outperforming all deep learning models, which achieved accuracy in the range of 47% to 70%. This indicates that traditional ML approaches can sometimes surpass state-of-the-art DL methods in tumor classification tasks.DiscussionThe findings highlight the importance of model selection and parameter tuning in automated brain tumor diagnosis. While deep learning models are generally considered standard for image analysis, Random Forest demonstrated superior performance in this context. This underscores the need for fine-grained consideration of dataset characteristics, computational resources, and diagnostic requirements.ConclusionThe study shows that carefully selected and optimized ML approaches can improve tumor classification and support more accurate and efficient diagnostic systems for brain tumor patients.

A communication system in the railway track environment provides constant channel coefficient property as trains travel on a predetermined route and speed. This channel characteristic provides advantages in designing a channel estimation. This paper proposes a channel estimation algorithm based on a deep learning network called the Convolutional Neural Network (CNN). The CNN has trained with an average Channel Frequency Response (CFR) dataset on railway track environments with different multi-path fading and noises. The CFR can be estimated using a known pilot symbol as conventional methods. The estimated CFR is also used to select the right CFR for multi-path fading compensation. The CNN will then classify the estimated CFR by recognizing estimated channel characteristics and determining the most matched estimated channel for equalization of received signals in a data channel. The simulation results show that the performance of the deep learning algorithms outperforms that of the conventional algorithms. Furthermore, the proposed method delivers better Bit Error Rate (BER) performance since the deep learning-based channel estimation can categorize the features of the channel characteristics with different multi-path and doppler shifts.

The proliferation of Internet of Things (IoT) devices has created unprecedented cybersecurity vulnerabilities, with botnets emerging as a critical threat to network infrastructure. This study focuses on traditional machine learning and deep learning approaches, proposes a novel ensemble framework to address these issues, integrating Convolutional Neural Network (CNN), Bidirectional Long Short-Term Memory (BiLSTM), Random Forest (RF), and Logistic Regression (LR) via a weighted soft-voting mechanism. Our approach introduces a Quantile Uniform transformation to reduce feature skewness, a multi-layered feature selection method to enhance discriminative power, an individual performance of deep learning-traditional machine learning and a hybrid models (ensemble models) for robust detection. Evaluated on BOT-IOT, CICIOT2023, and IOT23 datasets, the framework achieves 100% accuracy on BOT-IOT, 99.2% on CICIOT2023, and 91.5% on IOT23, outperforming state-of-the-art models by up to 6.2%. These contributions advance IoT security by enabling scalable, high-performance detection adaptable to diverse network scenarios, with practical optimizations for real-world deployment.