Search

Predicting a patient’s hemoglobin level or degree of anemia using Electronic Health Records (EHRs) is a non-invasive and rapid approach. However, it presents challenges due to the irregular multivariate time series nature of EHRs, which often contain significant amounts of missing values and irregular time intervals. To address these issues, we introduce HgbNet, a model specifically designed to process irregular EHR data. HgbNet incorporates a NanDense layer with a missing indicator to handle missing values. Inspired by clinicians’ decision-making processes, the model employs three kinds of attention mechanisms to account for both local irregularity and global irregularity. We evaluate the proposed method using two real-world datasets across two use cases. HgbNet outperforms the best baseline results across all test scenarios, achieving an R2 score of $0.867~\pm $ 0.003 and $0.861~\pm $ 0.003 for hemoglobin level prediction, and an F1 score of $0.855~\pm $ 0.005 and $0.843~\pm $ 0.005 for anemia degree prediction under usecase 1 across two datasets. Additionally, we analyze the effect of the length of irregular time intervals on prediction performance and improve HgbNet’s performance at long intervals in usecase 2. These findings highlight the feasibility of estimating hemoglobin levels and anemia degree from EHR data, positioning HgbNet as an effective non-invasive anemia diagnosis solution that could potentially enhance the quality of life for millions of affected individuals worldwide.

The role of inter-satellite communication (ISC) is of great significance in the contemporary era of satellite technology. In this context, visible light communication (VLC) has emerged as a highly promising alternative to conventional radio frequency (RF) communication. This is especially true for small satellites like CubeSats, where constraints related to size, weight, and power (SWaP) are of utmost importance. However, VLC systems’ limited bandwidth makes it challenging to achieve large data rates and capacity. This can be evaded with the utilization of non-orthogonal multiple access (NOMA) as a fundamental technology to augment ISC based on VLC. Such schemes have already been evaluated and published in the public domain literature. There is, nevertheless, lack of information on the efficacy of utilizing cooperative NOMA (Co-NOMA) scheme in conjunction with VLC for a constellation of satellites, with limited power resources. This research study investigates the utilization of NOMA as the access technique for inter-satellite connections when communication is done through VLC under power restrictions, with a particular emphasis on reducing the bit error rate (BER) while enhancing the average sum rate (ASR) for distant satellites through Co-NOMA. The efficacy of the proposed methodology is verified through simulation results.

By improving image quality and resolution, Single Image Super-Resolution (SISR) models help advance understanding of underwater environments. Super-resolution techniques hold promise in addressing these issues by enhancing image details. The transformer architecture has recently gained considerable popularity in low-level vision tasks, including image super-resolution (SR). In this paper, we present a novel multi-distillation algorithm based on the swin transformer and wavelet transform (KDSWT) for underwater image super-resolution. A wavelet block is added to the swin transformer to alleviate the loss of detailed features because wavelet transforms can preserve the high-frequency components, subtle variations, and localized features that contribute to the overall richness in images. Based on a large dataset of high-resolution underwater images, a ViT-based student model is trained to learn the intricate features of underwater images from the teacher model. Using knowledge distillation, the student model learns to upscale input images and capture the domain-specific features essential to understanding underwater scenes. To address the computational cost issue in ViT, we introduce a lightweight network training strategy that combines multi-anchor distillation and progressive learning, enabling the lightweight network to achieve exceptional performance and the fastest inference time compared to recent lightweight state-of-art methods. Our experimental results show that KDSWT lightweight (student) significantly improves peak signal-to-noise ratio (PSNR) over the (original) teacher model while decreasing the number of parameters by 54% and the calculation cost by 22%. Concurrently, KDSWT-SR excels beyond state-of-the-art SR techniques on three benchmark underwater datasets, leading to noteworthy improvements in both PSNR and structural similarity index (SSIM).

This paper presents a supervised machine learning approach using eight popular classifiers for fault classification in power transmission lines. The classification of faults, indicated by the behavior of the electrical signals associated with them, plays a pivotal role in maintaining the reliability, stability, and security of electrical grids. However, most of the previous studies on fault analysis in power systems relied on proprietary data and lack of standardized benchmarks, hindering the comparison of algorithms and making performance more erratic. Moreover, the nonavailability of labeled data for all types of faults is the most problematic. This paper proposes to perform fault classification on a data set created from a real-time Simulink model to standardize performance and advance research in this area. A new strategy for non-intensive feature extraction is applied using relatively simpler techniques, eliminating computationally expensive techniques such as wavelets. Feature selection through dimensionality reduction techniques is used to improve model performance and more efficient use of computational resources. The performance of the learning algorithms (e.g. Decision Tree, Random Forest, etc.) has been analyzed with various preprocessing techniques (e.g. data scaling, transformation, etc.) and tuning of parameters, focusing on the accuracy and computational time ( $T_{c}$ ), for performance generalization and efficiency. Performing specific operations on data in sequence of steps provided flexibility and adaptability in processing the data, making it easy to train, evaluate, and validate the learning algorithms. The results demonstrated that the proposed scheme can be effectively used for fault classification with high accuracy and significant reductions in $T_{c}$ under various operating conditions. The study also determined the best estimator for each classifier when building and training the classifier models, offering a variety of options. Logistic Regression, Random Forest and Support Vector Machine were the outperforming classifiers and proved their potential for classifying faults in electric power transmission lines.

In optical communication systems, the Q-factor is an important performance metric to evaluate the performance of an optical link. In this paper, a deep learning-based eye diagram analyzer is proposed to estimate the Q-factor. CNN architectures such as LeNet, Wide ResNet, and Inception-v4 are used for ON-Off Keying (OOK) and Pulse Amplitude Modulation (PAM) formats’ eye diagrams. The performance of these architectures is evaluated in terms of accuracy, Mean Squared Error (MSE), and error tolerance. This work shows that Wide ResNet demonstrates better performance in both OOK and PAM4 transmission schemes, achieving MSE values of 0.00188 and 0.00036, respectively. Additionally, it attains a high R-squared (R2) value of 0.9998. This deep learning-based eye diagram analyzer may be a promising approach for analyzing and optimizing optical communication systems without extensive human intervention.

This paper introduces an innovative $4 \times 4$ dual-band Butler matrix (BM) characterized by compactness and an enhanced frequency ratio (K). The design employs meandered lines and an interdigital capacitor (IDC) unit-cell-based composite right/left-handed transmission-line (CRLH-TL) metamaterial (MTM) structure. The BM integrates compact dual-band 3 dB branch-line couplers (BLC), a 0 dB crossover, and dual-band ±45° phase shifters on a single Rogers RT5880 substrate having relative permittivity $\varepsilon _{r}$ of 2.2 and thickness $h$ of 0.787 mm. Simulations and measurement results demonstrate reflection and isolation coefficients exceeding −20 dB at all ports, with obtained insertion loss of −6±3 dB over the 0.7 GHz and 3.5 GHz frequency bands. The achieved output phase differences of ±45°, ±135°, ±135°, and ±45° at the designed frequencies indicate a maximum average phase tolerance of ±4.5° concerning the ideal values. Importantly, the BM’s overall dimensions are 143 mm $ \times 186$ mm, resulting in an impressive 78% size reduction compared to traditional T-shaped BM designs. The proposed configuration is designed and simulated using CST Microwave Studio, with the agreement between simulated and measured parameters highlighting design reliability and effectiveness. Additionally, a performance evaluation comparing the proposed BM with existing circuits reveals its suitability for sub-6 GHz 5G dual-band antenna array beamforming networks (BFN) due to its compact size and improved band ratio.

In this paper, a real-time adaptive clustering-based distributed voltage regulation scheme is presented for the voltage regulation of unbalanced distribution systems. In the proposed methodology, first, adaptive clusters for the distributed control are formulated based on voltage to injected power sensitivity, and a threshold criterion is designed that limits the size of the clusters to strongly controlled nodes only. Having performed the clustering, a multi-objective optimization problem is solved for power factor optimization within clusters for voltage regulation of all the control zones of the distribution system. Voltage quality factor and voltage unbalance indices are utilized to show the improvements in voltage profile and reduction in voltage unbalance, respectively. Furthermore, for optimally finding the tap positions for multiple OLTCs, another optimization problem is designed. An IEEE 123-node test feeder with multiple wind and solar DGs of varying capacities is used for the simulations, and two test cases with different scenarios are simulated to test the proposed scheme. All the results verify that the proposed scheme effectively regulates the voltage of the unbalanced system without massive remote data requirements and active power curtailment which ensures the stability of the system’s voltage profile and prevents OLTC-hunting even in highly dynamic operational scenarios.

Nowadays, cloud-based storage systems play a vital role in IoT data storage, processing, and sharing. Despite its contribution, the current cloud-based architecture may cause severe data leakage or jeopardize user privacy. Meanwhile, the cloud-based architecture heavily relies on a trusted third-party auditor (TPA) and runs in a centralized control manner. However, the TPA may not be a completely trustworthy entity, and a single point of failure might cause the centralized system to collapse. Fortunately, with the advent of blockchain technology, the decentralized storage model has gained popularity. A decentralized storage system successfully eradicates the rule of TPA, solves the problem of a single point of failure, and has many advantages over a centralized control architecture, such as low storage prices and high throughput. This study offers a blockchain-based decentralized distributed storage and sharing scheme that provides end-to-end encryption and fine-grained access control. In our proposed IoTChain model, fine-grained permission is based on attribute-based access control (A-BAC) policy by employing the Ethereum blockchain as an auditable access control layer. Smart contracts are tailored for the IoTChain model, which combines the Ethereum blockchain and the interplanetary file system (IPFS). We used an advanced encryption standard (AES) for encryption and the elliptic curve Diffie-Hellman key exchange protocol for secret key sharing between data owners and users. Also, the proof-of-work (PoW) consensus mechanism is replaced with a proof-of-authority (PoA) to minimize system transaction cost and boost system throughput. Additionally, our solution has been tested on the Ethereum official test network Rinkeby, and the results demonstrate that our approach is realistic and economical on the IoT data.

The increase in number of power transformer necessitates increased awareness among power system engineers to monitor transformer health in order to avoid complete failures that may results in prolonged outages. The utilities, nowadays, are exploring different ways of detecting these faults in an early stage. This paper aims to present a real-time monitoring technique for a three-phase transformer using data analytics. Accordingly, the use of signal to noise ratio (SNR) for fault detection in power transformers has been described. This technique determines the partial discharge with the help of sensors needed for unique identification of failed transformers in different test systems. The SNR for different transformer coils installed in field conditions has been compared and discussed in detail. The technique is expected to enable power transformer operators for detection of incipient faults and thus can avoid catastrophic failures and the resultant losses.

This study presents a novel approach for solving the economic dispatch (ED) problem in groups of generating units communicating through a communication network. The suggested strategy is a consensus-based dynamic event-triggered (ET) distributed optimization method. Our methodology considers the sharing of the local information between generators and their convex cost functions to address the total cost function and offers a decentralized optimization solution over a network. The proposed distributed method addresses the ED problem by considering the criterion of optimal cost and by offering efficient communication. Generating units are grouped according to their generation operational limits, that is, total capacity and dynamic ET distributed protocols are developed to ensure the consensus of cost variables among generating units, operating under normal capacity conditions. The remaining generating agents work on their operating limits, which are segregated through the sharing of flag information through a switching mechanism. Consequently, in contrast to the existing methods, the recommended protocol allows nodes to function in groups, based on the power supply, for ED with geographical clustering and capacity restrictions, in addition to handling the system constraints. Furthermore, the proposed technique employs a dynamic triggering method to manage bandwidth and guarantee the elimination of Zeno behavior. The simulation results validate the efficacy of the proposed approach. [ABSTRACT FROM AUTHOR]

New era ciphers employ substitution boxes (S-boxes) which assist in the provision of security for the plaintext in the encryption phase and transforming the ciphertext on the receiver side into original plaintext in the decryption phase. The overall security of a given cipher engaging an S-box greatly depends on the cryptographic forte of the respective S-box. Consequently, many researchers have used different innovative approaches to construct robust S-Boxes. In this article, an innovative and modest square polynomial transformation, the very first time, along with a novel affine transformation and a pioneering permutation approach to construct dynamic S-boxes is proposed. The proposed method has the capability to erect a huge number of robust S-boxes by applying minute changes in the parameters of transformation and permutation processes. An example S-Box is generated, and its recital analysis has been done using typical criteria including bijectivity, strict avalanche criterion, nonlinearity, bit independence criterion, linear probability, differential probability, and fixed-point analysis to check its cryptographic forte. This performance of the proposed S-box is placed side by side against state-of-the-art S-boxes to prove its strength. The performance and comparative analyses authenticate that the projected S-box possesses the true competence for its application in modern-day ciphers.

There exists a gap between existing security mechanisms and their ability to detect advancing threats. Antivirus and EDR (End Point Detection and Response) aim to detect and prevent threats; such security mechanisms are reactive. This approach did not prove to be effective in protecting against stealthy attacks. SCADA (Supervisory Control and Data Acquisition) security is crucial for any country. However, SCADA is always an easy target for adversaries due to a lack of security for heterogeneous devices. An attack on SCADA is mainly considered a national-level threat. Recent research on SCADA security has not considered “unknown threats,” which has left a gap in security. The proactive approach, such as threat hunting, is the need of the hour. In this research, we investigated that threat hunting in conjunction with cyber deception and kill chain has countervailing effects on detecting SCADA threats and mitigating them. We have used the concept of “decoy farm” in the SCADA network, where all attacks are engaged. Moreover, we present a novel threat detection and prevention approach for SCADA, focusing on unknown threats. To test the effectiveness of approach, we emulated several SCADA, Linux and Windows based attacks on a simulated SCADA network. We have concluded that our approach detects and prevents the attacker before using the current reactive approach and security mechanism for SCADA with enhanced protection for heterogeneous devices. The results and experiments show that the proposed threat hunting approach has significantly improved the threat detection ability.