Load forecasting in Smart Grids (SG) is a major module of current energy management systems, that play a vital role in optimizing resource allocation, improving grid stability, and assisting the combination of renewable energy sources (RES). It contains the predictive of electricity consumption forms over certain time intervals. Load Forecasting remains a stimulating task as load data has exhibited changing patterns because of factors such as weather change and shifts in energy usage behaviour. The beginning of advanced data analytics and machine learning (ML) approaches; particularly deep learning (DL) has mostly enhanced load forecasting accuracy. Deep neural networks (DNNs) namely Long Short-Term Memory (LSTM) and Convolutional Neural Networks (CNN) have achieved popularity for their capability to capture difficult temporal dependencies in load data. This study designs a Short-Load Forecasting scheme using a Hybrid Deep Learning and Beluga Whale Optimization (LFS-HDLBWO) approach. The major intention of the LFS-HDLBWO technique is to predict the load in the SG environment. To accomplish this, the LFS-HDLBWO technique initially uses a Z-score normalization approach for scaling the input dataset. Besides, the LFS-HDLBWO technique makes use of convolutional bidirectional long short-term memory with an autoencoder (CBLSTM-AE) model for load prediction purposes. Finally, the BWO algorithm could be used for optimal hyperparameter selection of the CBLSTM-AE algorithm, which helps to enhance the overall prediction results. A wide-ranging experimental analysis was made to illustrate the better predictive results of the LFS-HDLBWO method. The obtained value demonstrates the outstanding performance of the LFS-HDLBWO system over other existing DL algorithms with a minimum average error rate of 3.43 and 2.26 under FE and Dayton grid datasets, respectively.
Intelligent Cyber-Physical Transportation Systems (ICTS) rely deeply on data exchange between moving vehicles and fixed infrastructure. While exchanging data between vehicles in the network, security is a main constraint. ICTS encompasses many facilities and claims, such as managing road traffic, traveler data, public transportation systems, and autonomous vehicles. It is anticipated that ICTS will make significant contributions to future smart cities and urban design in road and traffic security, transportation and transport optimization, energy efficiency, and environmental pollution control. Flexibility, variety of quality-of-service needs, and massive volumes of data generated all present challenges when working with ITS. Previous works on cyber-physical attack detection lag in ensuring security at various levels. To monitor In-Vehicle Networks (IVN), Vehicle-to-Vehicle (V2V) Communications, and Vehicle-to-Infrastructure (V2I) Networks for malicious behavior, we introduce a deep learning-based Intrusion Detection System (IDS) for ICTS. In this research, we look into how Deep Learning (DL), a relatively new but rapidly growing subject, could help improve ICTS. This research investigates whether deep learning can enhance the security of autonomous vehicles and other forms of smart transportation. We improve the smart transport network using the deep learning method, simulate it, and statistically study its data transmission economy, accuracy projections of the future, and path-switching strategy. To identify malicious activity in the core networks of autonomous vehicles (AV), we have developed an ensemble Long-Short Term Memory (LSTM) technique based on Deep Learning architecture. The car hacking dataset (used to monitor communications within a vehicle) and the UNSWNB15 dataset (used to monitor communications with the outside world) are used to test and compare the performance of the proposed IDS. The experimental findings showed that our suggested system outperformed other intrusion detection methods by achieving 100% accuracy in detecting all sorts of assaults on the auto-hacking dataset and 99% on the UNSW-NB15 dataset.
Guided bone regeneration (GBR) is a reliable technique used to treat ridge deficiencies prior or during implant placement. Injectable-platelet rich fibrin (i-PRF) laced with a bone substitute (sticky bone) has heralded the way for advancing the outcomes of bone regeneration. This study evaluated the efficacy of sticky bone in horizontal ridge augmentation with and without collagen membrane. A total of 20 partially edentulous patients (Group-I n = 10; Group-II n = 10) that indicated GBR were included, and the surgical procedure was carried out. In Group-I, the sticky bone and collagen membrane were placed in ridge-deficient sites and Group-II received only sticky bone. At the end of 6 months, 20 patients (Group-I (n = 10); Group-II (n = 10)) completed the follow-up period. A CBCT examination was performed to assess changes in the horizontal ridge width (HRW) and vertical bone height (VBH). A statistically significant increase in HRW (p < 0.05) was observed in both groups with mean gains of 1.35 mm, 1.55 mm, and 1.93 mm at three levels (crest, 3 mm, and 6 mm) in Group-I and 2.7 mm, 2.8 mm, and 2.6 mm at three levels in Group-II. The intergroup comparison revealed statistical significance (p < 0.05) with respect to HRW and KTW (Keratinised tissue width) gains of 0.775 at the 6-month follow-up. Sticky-bone (Xenogenic-bone graft + i-PRF) served as a promising biomaterial in achieving better horizontal bone width gain.
Lung and colon cancers are deadly diseases that can develop concurrently in organs and undesirably affect human life in some special cases. The detection of these cancers from histopathological images poses a complex challenge in medical diagnostics. Advanced image processing techniques, including deep learning algorithms, offer a solution by analyzing intricate patterns and structures in histopathological slides. The integration of artificial intelligence in histopathological analysis not only improves the proficiency of cancer detection but also holds the potential to increase prognostic assessments, eventually contributing to effective treatment strategies for patients with lung and colon cancers. This manuscript presents an Improved Water Strider Algorithm with Convolutional Autoencoder for Lung and Colon Cancer Detection (IWSACAE-LCCD) on HIs. The major aim of the IWSACAE-LCCD technique aims to detect lung and colon cancer. For noise removal process, median filtering (MF) approach can be used. Besides, deep convolutional neural network based MobileNetv2 model can be applied as a feature extractor with IWSA based hyperparameter optimizer. Finally, convolutional autoencoder (CAE) model can be applied to detect the presence of lung and colon cancer. To enhance the detection results of the IWSACAE-LCCD technique, a series of simulations were performed. The obtained results highlighted that the IWSACAE-LCCD technique outperforms other approaches in terms of different measures.
29 August 2025 in EURASIP Journal on Wireless Communications and Networking
The rapid evolution of 5G networks has introduced unparalleled connectivity, speed, and efficiency. However, these advancements bring critical cybersecurity challenges due to increased attack surfaces, network slicing vulnerabilities, and adaptive threat vectors. Traditional IDS and ML-based mechanisms lack the adaptability required for real-time threat mitigation in such complex environments. To address this, we propose SecureNet-RL, an AI-powered cyber resilience model that leverages reinforcement learning (RL) for proactive, real-time threat hunting in 5G networks. The framework integrates Deep Q-Networks and Proximal Policy Optimization with federated multi-agent learning to enable distributed, adaptive defense across network slices. SecureNet-RL agents autonomously learn from network traffic patterns, identify anomalies, and mitigate threats without human intervention. The system incorporates adversarial robustness through reward shaping and GAN-generated attack simulations to counter evolving threats. Evaluated using DDoS, insider, and 0-day threat scenarios over NS3-based 5G slices, the model consistently achieves 95.8% detection accuracy, < 50 ms mitigation latency, and a false-positive rate of just 4.3%—surpassing traditional IDS (78.2%) and supervised ML (88.5%). It also demonstrates efficient resource utilization with 45% CPU and 1.5 GB memory, ensuring scalability in large-scale 5G deployments. Key contributions include: (1) real-time adaptive RL-based threat hunting, (2) federated learning integration for decentralized scalability, (3) adversarial robustness via reward and attack modeling, and (4) comprehensive validation on multi-modal threat datasets. These findings confirm that SecureNet-RL offers a scalable, intelligent, and self-improving cyber defense solution tailored for next-generation 5G networks.