Nazmul Hasan

Wireless Sensor Networks (WSNs) are widely used for real-time monitoring in industrial automation, environmental sensing, and healthcare. However, sensor faults pose significant challenges, potentially leading to compromised data integrity and reduced system reliability. Traditional fault detection methods often rely on large labeled datasets, which are scarce in real-world scenarios. This paper proposes a novel data-efficient framework that integrates unsupervised learning for fault detection with few-shot learning for fault classification, addressing the challenge of a limited number of fault data in WSNs. For fault detection, we introduce the LSTM Autoencoder with a Temporal Attention Mechanism (LATAM) model, trained on normal data, that identifies anomalies based on reconstruction errors and a dynamic thresholding technique. For fault classification, we employ a few-shot learning approach using Convolutional Kernel Transform (MiniROCKET) for efficient feature extraction, and Prototypical Networks to classify faults with a minimal number of labeled samples. Due to the lack of comprehensive public fault datasets, the proposed framework was evaluated on a dataset combining real-world healthy temperature sensor readings with synthetically generated faults. Experimental results demonstrate that the LATAM model achieves high accuracy in fault detection, with an F-score of 98.90%. The few-shot classification approach performs effectively in low-data scenarios, achieving F-scores of 94.18% in a 15-shot setting and 84.75% in a one-shot setting. Compared to the traditional 1D CNN, LSTM, and Matching Networks methods, the proposed framework exhibits superior performance, particularly in data-scarce environments. By eliminating reliance on large labeled datasets and pre-trained models, this approach effectively addresses the challenges posed by data scarcity in WSNs.

Ensuring reliable data collection from sensor nodes is vital for the success of Internet of Things (IoT) applications such as smart cities, industrial automation, and other data-driven systems. As sensor data underpins key decision-making processes, undetected faults can compromise system performance, leading to misinterpretations and costly operational disruptions. Therefore, timely and accurate fault classification is essential to preserve the integrity and resilience of IoT deployments. This article proposes a novel method for classifying five common types of sensor faults across homogeneous sensor nodes using a combination of federated learning (FL) and transfer learning (TL). The approach supports two categories of client nodes: source nodes with labeled data and target nodes without labels. Federated training at source nodes exploits dataset similarity with the unlabeled target node, while a central server adaptively weights each client’s contribution based on its relevance to the target sensor before aggregation. Across various source-target configurations, the proposed method consistently outperforms baseline approaches. It yields 2%–11% higher accuracy compared to the widely used FedAvg method. In addition, it improves over a centralized LSTM-based model by +3.5% to +5.4%, and surpasses traditional classifiers—such as support vector machine (SVM), decision tree, random forest, and K-nearest neighbors (KNN)—by up to +33.6% in accuracy and +40.2% in ${F}1$ -score, especially in target nodes lacking labeled data. These results highlight the effectiveness of the proposed approach in enabling robust, scalable, and privacy-preserving sensor fault classification in real-world IoT systems.

This paper introduces a few-shot transfer learning approach to fault classification in wireless sensor networks (WSNs) with a minimal number of fault samples. WSNs are susceptible to various faults, such as drift, stuck, bias, and spike faults, erratic behavior, and data loss, which can compromise system reliability. Conventional deep learning fault diagnosis methods have achieved promising results, however, the majority of these approaches require a substantial amount of labeled training data, which is not available in real-world scenarios. To address this, we propose a novel method that combines convolutional neural networks (ResNet-18, VGG-16, and MobileNetV2 backbone architectures) with prototypical networks for fault diagnosis in WSNs with only a few fault samples. By transforming time-series data into Gramian Angular Field images, our approach leverages pre-trained deep learning models to extract feature-rich embeddings. These embeddings are then classified using a prototypical network, which enhances the system’s ability to diagnose faults even with a limited amount of labeled data. The proposed model is lightweight and deployable on Internet of Things (IoT) devices, ensuring efficient fault classification with minimal computational resources. Experimental results demonstrate the model’s high accuracy and robustness across various fault types, highlighting its potential for scalable and adaptive IoT applications.

The Internet of Things (IoT) is witnessing a surge in sensor-equipped devices. The data generated by these IoT devices serve as a critical foundation for informed decision-making, real-time insights, and innovative solutions across various applications in everyday life. However, data reliability is often compromised due to the vulnerability of sensors to faults arising from harsh operational conditions that can adversely affect the subsequent operations that depend on the collected data. Hence, the identification of anomalies within sensor-derived data holds significant importance in the IoT context. This article proposes a sensor fault detection method using a Long Short-Term Memory autoencoder (LSTM-AE). The AE, trained on normal sensor data, predicts a 20-step window, generating three statistical features via SHapley Additive exPlanations from the estimated steps. These features aid in determining potential faults in the predicted steps using a machine learning classifier. A secondary classifier identifies the type of fault in the sensor signal. Experimentation on two sensor datasets showcases the method’s functionality, achieving fault detection accuracies of approximately 93% and 97%. It is possible to attain a perfect fault classification performance by slightly modifying the feature calculation approach. In a univariate prediction scenario, our proposed approach demonstrates good fault detection and classification performance.

Sleep stage classification plays a pivotal role in predicting and diagnosing numerous health issues from human sleep data. Manual sleep staging requires human expertise, which is occasionally prone to error and variation. In recent times, availability of polysomnography data has aided progress in automatic sleep-stage classification. In this paper, a hybrid deep learning model is proposed for classifying sleep and wake states based on a single-channel electroencephalogram (EEG) signal. The model combines an artificial neural network (ANN) and a convolutional neural network (CNN) trained using mixed-input features. The ANN makes use of statistical features calculated from EEG epochs, and the CNN operates on Hilbert spectrum images generated during each epoch. The proposed method is assessed using single-channel Pz-Oz EEG signals from the Sleep-EDF database Expanded. The classification performance on four randomly selected individuals shows that the proposed model can achieve accuracy of around 96% in classifying between sleep and wake states from EEG recordings.

In this Internet of Things (IoT) era, the number of devices capable of sensing their surroundings is increasing day by day. Based on the data from these devices, numerous services and systems are now offered where critical decisions depend on the data collected by sensors. Therefore, error-free data are most desirable, but due to extreme operating environments, the possibility of faults occurring in sensors is high. So, detecting faults in data obtained by sensors is important. In this paper, a digital twin inspired detection approach is proposed, and its ability to detect a single type of fault in several sensor is analyzed. The digital equivalent of the sensor is developed using a Generative Adversarial Network (GAN). As GANs inherently performs well with images, Gramian Angular Field (GAF) encoding is used to convert timeseries data to image. The GAF encoding preserves the temporal relations of the timeseries data. The GAN is trained with the GAF images. The trained GAN model acts as the virtual representation of the sensor, and the discriminator network of the GAN model, once it has learned the pattern of normal data, is used as the fault detector. The performance of the virtual sensor is promising because it successfully generates data for normal conditions. The best fault detection accuracy achieved by the proposed model is 98.7%, which makes this GAN-based digital twin inspired approach a promising candidate for sensor fault detection.

Total Volatile Base Nitrogen (TVB-N) and Trimethylamine Nitrogen (TMA-N) contents of Vetki (Lates calcarifer) were investigated following Conway’s Microdiffusion technique at 4 hour interval in Fish Quality Control Laboratory, Fisheries and Marine Resource Technology Discipline, Khulna University. Observed TVB-N and TMA-N value did not vary significantly among different size classes. At fresh condition, TVB-N content was observed as 11.106±0.8 mg, 11.162±0.8 mg, and 11.162±0.8 mg in small, medium and large size classes of fish respectively and it increased to 59.087±0.8 mg, 68.417±1.5 mg and 75.967±2.7 mg for 100 g^-1 respectively at complete spoilage stage (after 28 hrs). At fresh condition, TMA–N content was 6.7±0.016 mg, 11.069±1.79 mg and 11.172±1.01 mg for 100 g^-1 in small, medium and large size classes of fish respectively and it increased to 48.893±3.01 mg, 52.492±.089 mg and 57.717±1.95 mg for 100 g^-1 respectively at complete spoilage stage. TVB-N and TMA-N values started to increase gradually in all size groups with the passage of time but the rate of increase became faster after 20th hour and continued till the final stage was reached.

In this paper, we have investigated the recirculation flow of a nanofluid in the developed flow zones of an infinite periodic nanotube with a hexagonal cross-section. A numerical analysis is commenced to identify the existence of single and double recirculation zones for various geometric parameters. The boundary element method (BEM) has been formulated for an unbound interim nanotube for solving the governing equations. We have developed the codes for the BEM method in FORTRAN 90, and the graphs have been plotted in MATLAB 2016. We have various geometric parameters to inaugurate circumstances for the onset of recirculation. We have found recirculation flow through this type of periodic tube for a set of geometric parameters such as amplitude, wavelength and throat radius etc. Firstly recirculation flow in the expansion region of the nanotube is prophesied to arrive beyond a critical amplitude and second order recirculation zone is also predicted for still higher amplitudes. The recirculation flow has great importance in the application, for example, it can be used in the particle separation process.

The ubiquitous computing era is characterized by wide deployment of resource constraint devices. Secure communication between such devices encounters a big challenge due to their resource limitation. The lightweight cryptographic algorithms are developed to address this challenge. In this work, implementation of LBlock, a lightweight encryption algorithm is investigated in Altera DE1 FPGA board. The encryption algorithm involves a Feistal structure and requires 32 clock cycles to encrypt 64-bit data. The implementation results show that it requires 326 Logic Elements (LEs) in Cyclone II EP2C20F484C7 platform. The maximum clock frequency was reported as 156.32 MHz and a corresponding power consumption of 107.62 mW. Also, the throughput is found as 312 Mbps. These results are better as compared to some other lightweight encryption such as XTEA and Hummingbird for the same FPGA platform.