Software-Defined Networking Intrusion Detection Systems (SDN-IDS) are designed to monitor and analyze network traffic within the centralized SDN architecture to detect malicious activities. The SDN-IDS system uses SDN controller programmability and worldwide network accessibility to achieve effective threat detection and security measures, which can be adjusted in real-time. The research introduces an improved SDN-IDS framework security solution, which uses Ant Colony Optimization (ACO) for feature selection and Kolmogorov-Arnold Networks (KAN) to perform classification. The proposed method is tested using the CICIDS2017 dataset, which simulates actual network traffic conditions. Data cleaning and min-max normalization are performed to enhance data quality and ensure stable model convergence. ACO identifies the most informative feature subset, which results in lower dimensionality and higher computational efficiency. The optimized features undergo classification through KAN, which successfully models complex nonlinear traffic patterns. The experimental results show better detection performance of 99.08% accuracy. The test results demonstrate that the proposed framework successfully delivers secure operations and scalable performance in SDN environments.

Storage, processing, and real-time analytics are becoming increasingly challenging due to the exponential growth of healthcare data generated by wearable devices, EHRs, medical imaging systems, and remote patient monitoring. Concerns of privacy, bandwidth, and latency arise when dealing with time-sensitive medical data stored in centralized cloud-based systems. An integrated Edge-Cloud computing architecture for distributed healthcare data processing is presented in this paper for safe, efficient, and real-time medical analytics. An Internet of Things (IoT) medical device network located close to healthcare facilities is the basis of the proposed solution. In order to decrease latency and alleviate network congestion, edge nodes perform filtering, preprocessing, anomaly identification, and rapid response generation. Important health warnings, such as aberrant vital signs, can be processed locally, which helps with therapeutic intervention. Storage on a grand scale, analytics, training models, and managing a population’s health are all made possible with cloud computing. By allocating workloads adaptively, we can keep computations on the edge and in the cloud in balance. Lightweight edge-based machine learning models and computationally intensive cloud-based deep learning algorithms both provide real-time decision support. The transfer and storage of patient data is protected via communication protocols and encryption. Distributed systems rely on data synchronization and orchestration modules to keep them reliable. When compared to cloud-only systems, Edge-Cloud hybrids outperform them in terms of response latency reduction, bandwidth optimization, and scalability. Healthcare monitoring may be maintained while fault tolerance and robustness are enhanced by distributed design.

Machine learning has been used for decades to analyze vast datasets, classify and cluster data, and make predictions using algorithms. One of its top use areas is cybersecurity, where it can help detect and prevent destructive threats such as malware. The use of machine learning in cybersecurity has proven to be a powerful tool in detecting and predicting malware attacks. In recent years, the number of Internet users has greatly increased and with it the number of malware attacks. This has made predicting malware a challenge. Consequently, to date, there is still a need to examine the numerous existing MLs’ performance. This study is presented to identify the best classification model for predicting malware using two datasets and 18 different classifiers belonging to six learning strategies. The results showed that the RandomForest classifier had the highest accuracy, precision, recall, F1-measure, and ROC Area metrics, Moreover, Trees and Bayes learning strategies showed the best predictive performance on the two datasets compared with the other five learning strategies.

Web 4.0 platforms introduce intelligent, decentralized agents and real-time interactions that increase both utility and attack surface. This paper presents a comprehensive, reproducible AI blockchain hybrid designed to (1) detect SQL injection attacks at scale using a textual TFIDF + machine-learning pipeline, (2) incorporate reputation signals from a real-world Bitcoin OTC trust dataset to compute a TrustAlert Score (TAS) that prioritizes alerts and guides logging policy, and (3) record privacy-preserving audit digests on blockchain, optionally attested via a zero-knowledge proof (ZKP) pipeline. We evaluate the system on a 148 k SQL corpus and Soc-SignBitcoinOTC reputation data. The detection module achieves high accuracy (0.9797), F1 (0.9807), and ROCAUC (0.9972). TAS effectively separates malicious from benign events (TAS AUC = 0.96) and enables selective on-chain logging to reduce cost. Blockchain benchmarks indicate that local (Ganache) throughput is adequate for batched logging, while public testnet (Goerli) exhibits significantly higher latency and gas usage; ZKP attachments further increase on-chain cost. We discuss practical deployment patterns (digest-only on-chain, Layer2 batching), propose evaluation extensions (transfer learning, adversarial red-teaming), and release reproducible scripts for the community.

IoT enables continuous patient monitoring and immediate response to healthcare needs by creating a connected environment where healthcare services can interact seamlessly and continuously with one another. However, IoT-based healthcare systems represent an attractive target for cybercriminals, as they hold sensitive medical information on individuals. An IoT-based healthcare platform can be compromised through data manipulation, unauthorized access, and denial-of-service (DoS) attacks. To keep this issue from occurring, this paper describes a secure solution for the continuous monitoring of patient health through the use of IoT technology. The Secure IoT-based Biomedical Monitoring Framework (SIBMF) will utilize lightweight encryption techniques and the Machine Learning-based Intrusion Detection System (MLIDS) to enhance security for IoT users. Physiological signals from patients’ bodies (ECG, heart rate, and temperature) will be captured by IoT sensors and transmitted in an encrypted format to a cloud-based server. IoT sensors will utilize a cryptographic algorithm to encrypt their biomedical data prior to sending it to a cloud-based server, ensuring the confidentiality of patient data. MLIDS will detect unauthorized suspect access to the IoT network by analyzing the features of the traffic created by the IoT devices. A prototype of the SIBMF was built in MATLAB, and various performance metrics (i.e., accuracy, detection rate, false positive rate, data encryption time, and latency) were assessed. The experimental results demonstrate that this approach significantly improves data security and will provide superior accuracy for monitoring patients’ health and has minimal computational requirements.

Companies are excited to maintain high-quality software while reducing the costs of production. DevOps is a contemporary software development life cycle paradigm in which development and operations teams join together during all stages of software development. Nonetheless, security is inadequately integrated inside DevOps. Although there have been attempts to amalgamate security with DevOps, resulting in the emergence of DevSecOps, considerable progress is necessary. The aim of Hybrid Intrusion Detection and Ensemble Learning System (HIDELS) is to present the incorporation of intrusion detection into the continuous monitoring phase of DevOps, hence enhancing DevSecOps. The integration comprises five machine learning (ML) models and assesses the performance of each model independently. Subsequently, the models are consolidated into an ensemble learning (EL) framework to improve overall robustness and provide more stable predictive outcomes. The results of the individual models show the decision tree (DT) classifier outperforming all the remaining models in terms of accuracy, precision, recall, and f1-score, with 99.5%, 99.5%, 99.7%, and 99.6% on average. Conversely, the EL model attained an average of 99.4%, 99.6%, 99.7%, and 99.7% for accuracy, precision, recall, and F1-score, respectively, exceeding the performance of all other individual ML models.

Diabetic Retinopathy (DR) poses a significant risk to vision among diabetic patients, necessitating early and accurate detection for effective intervention. This research introduced a novel methodology to detect the diabetic retinopathy in an efficient way. In contrast to the existing technique called Convolutional Neural Network (CNN), the proposed approach called Hybrid Learning assisted Diabetic Retinopathy Predictor (HLDRP) allows for easy and quick implementation and offers patients direct access to diagnostic equipment. By leveraging a hybrid learning model that incorporates both supervised and unsupervised learning techniques, HLDRP enhances the accuracy and reliability of DR diagnosis. The IoT component facilitates continuous monitoring and data acquisition from patients, while the hybrid learning model analyzes this data to detect early signs of DR and predict disease progression. Experimental results indicate that HLDRP offers superior diagnostic performance compared to traditional methods, with increased sensitivity and specificity. This integrated approach not only improves early detection but also enables personalized treatment planning, potentially transforming the management of diabetic retinopathy through innovative technology and advanced analytics. Through comprehensive evaluation using a diverse dataset of retinal images, HLDRP demonstrates significant improvements in diagnostic performance, including increased sensitivity and specificity compared to traditional detection methods. The results underscore HLDRP's potential to facilitate early intervention and personalized treatment, providing a valuable tool for healthcare professionals in managing and preventing diabetic retinopathy.

Unmanned Aerial Vehicles (UAVs) are increasingly deployed in mission-critical applications where reliable and secure Command and Control (C2) links are essential. Conventional UAV communication systems predominantly rely on wireless links, which are inherently vulnerable to radio-frequency (RF) jamming, spoofing, and interception. Fiber-optic tethered communication offers immunity to RF-based attacks while delivering high bandwidth and low latency; however, physical tether degradation or failure can disrupt operations and necessitate robust failover mechanisms. This paper presents FORTRESS-Fiber-Optimized Resilient Tethered Routing & Real-time Encryption with Seamless Switch-over-a hybrid fiber-RF UAV communication protocol that integrates predictive optical link-health monitoring, TimeSensitive Networking (TSN) scheduling, post-quantum cryptography (PQC), and a graceful failover strategy to maintain flight stability. FORTRESS continuously monitors the fiber link using pilot-tone attenuation tracking and micro-optical timedomain reflectometry (micro-OTDR), enabling early detection of degradation and proactive RF pre-arming. Simulation and hardware-in-the-loop (HIL) results demonstrate that FORTRESS achieves sub-millisecond C2 jitter under healthy conditions, extends early-warning lead time by up to 30 s before failure, and reduces failover-induced delay spikes by more than 50% compared to immediate RF switching. These results indicate that FORTRESS provides a scalable, secure, and resilient communication framework for UAV operations in contested environments.

This paper aims to investigate the application of PKE technologies and its development in the e-government setups along with analyzing the relevance of PKE technologies in supporting secure and efficient communications between government departments and the members of the public. Through analyzing RSA algorithm and advanced lattice-based cryptography, the paper highlights that application of PKE is also significant for defending the newly generated cyber threats while identifies advantages and possible usages of the method. Moreover, the review continues to discuss the actual usability issues associated with the architecture of PKI in egovernment, like key management difficulties and the amount of computation required by the encryption method. Challenges are investigated regarding the ways in which biometric data can be incorporated into e-government services in the future and how quantum-resistant cryptographic methods might potentially be incorporated to improve the stability of e-government services. Thus, when appraising the conceptual topicality of the review based on the analysis of a wide range of sources, it can be noted that the exploration of the importance of PKE in ensuring the provision of safe, effective, and clear e-gov services suggests the necessity of further development and improvement of cryptographic approaches as a response to the dynamic growth of threats in the sphere of the digital environment.

The exponential increase in Internet of Things (IoT) networks and environments has produce new security challenges that need to be solved, especially in cyberattacks detection within networks. This paper present anomaly detection mechanism to identify abnormal behavior within IoT environments. BoT-IoT dataset is employed to identify abnormalities. As Machine learning Algorithms have a significant role in developing detection techniques, we harness the power of various algorithms to detect whether the traffic is normally behave or maliciously behave. Various ML models such as Support Vector Machine (SVM), Random Forest (RF), and XGBoost are evaluated and compared using various measurements including accuracy, precision and recall. The findings of the research shows that the best models was … with significant accuracy of …, precision of …, and recall of …

The innovative Internet of Things (IoT) concept has become a worldwide phenomenon. The growing use of IoT devices over the past few decades has changed the way we interact with the embedded devices around us. The exponential rise of IoT devices and the data flowing from these devices is creating an impact on various fields such as healthcare, transportation, smart cities, and more. This paper aims to describe the potential and discuss challenges for using IoT-Aware models in the healthcare domain, identifying threats that should be mitigated to guarantee the successful applicability of those models. The most recent research on each component of the model is then presented, along with an assessment of its advantages, disadvantages, and applicability to a wearable IoT healthcare system. Security, privacy, wearability, and low-power operation are only a few of the difficulties that the IoT in healthcare must overcome. IoT challenges in healthcare include wearability, security, and privacy issues. Recommendations are also offered for future research paths. This paper concludes that IoT might assist governments in enhancing societal health care and business connections.

Many governmental and private organizations and institutions seek to keep pace with the technological development in their business to provide an electronic system that provides all the services and information customers need at any time and in any place, and also to save effort and time on it, through the use of cloud computing services and the services it provides for these institutions and organizations through the “payment-for- service” services and system management services provided by the service provider or by the user, or both. Despite the availability of cloud computing service, there are challenges and obstacles faced by the organization and service providers, and the most important of these challenges are issues of information security and privacy because of the negative impact on the client and the organization of breaches and tampering with this data. As a result, new models and analyses are needed to deal with the ever-increasing volume of data. Hence, this study came about to clarify the security challenges of big data.

Energy efficiency and reliable data delivery are central challenges in the design of Wireless Sensor Networks (WSNs), where battery-powered sensor nodes are prone to early depletion due to unbalanced communication loads. While clustering protocols such as LEACH and its enhanced variant LEACH-EA have improved energy utilization by rotating cluster head (CH) roles, issues of uneven energy consumption and node overuse persist, leading to energy “hotspots” and reduced network longevity. This paper presents an in-depth modification to LEACH-EA by introducing a random node re-clustering mechanism, triggered dynamically when a threshold proportion of nodes reach half of their initial energy. The protocol leverages a “dirty bit” signaling scheme and adaptive cluster head re-election coordinated by the base station, ensuring balanced communication load and prolonged operational life. Extensive simulation results demonstrate that the proposed approach significantly outperforms conventional protocols not only in network lifetime, first dead node round, and total data throughput, but also in advanced metrics including average energy consumption per round, standard deviation of residual energy, packet delivery ratio, and alive node rate. The findings confirm that random re-clustering is an effective, low-overhead enhancement for robust and scalable WSNs, with strong potential for real-world deployment.

The importance of data masking techniques has increased due to the rapid expansion of covert communications and the mass sharing of multimedia content. Steganography is a technique that allows for the safe and covert sharing of data. It's the science of incorporating data that is hard to see with the unaided eye. File types that support Steganography include audio, video, and image formats. Artificial Intelligence (AI) can enhance Steganography while also complicating the discovery of hidden data and augmenting steganalysis. By looking into these issues, this study intends to compare the various uses and techniques of AI and investigate the difficulties involved in incorporating AI technologies into Steganography. The study used the comparative research approach, comparing the literature with works that addressed Traditional Steganography in general. According to this Paper, countries need to educate themselves on classical Steganography as soon as possible to enhance information security and protection in the future.

Face landmarking is a primary goal of many projects that lead to face-preparing activities, like biometric recognition and mental state comprehension. Despite the inherent diversity of faces, the topic has proven to be extremely difficult due to a wide range of perplexing variables, such as position, expression, illumination, and occlusions. The aim of the HASAAM integrated fitting model is to find new solutions for the feature identification issue by combining the strengths of the Active Shape Model (ASM) and Active Appearance Model (AAM) to provide unique findings on the feature detection problem. In the first step, the ASM was used to identify the external shape landmarks of the face, and in the second stage, the AAM was used to identify the interior form landmarks. Then, the two kinds of landmarks are combined to generate the full face landmark. One issue with ASM is that it can’t produce an optimal global result since it will unavoidably converge to neighborhood minima. Nonetheless, ASM produces precise fitting results for the face’s exterior features, which need essential gradient values. In order to prevent AAM from fitting in order to address the local minima problem, ASM was used to identify these exterior landmarks of the face. The trials were run using the MORPH and LFPW datasets, which are available to the general public. In comparison to other techniques like ASM and AAM, the proposed hybrid model experiment result demonstrated effectiveness in extracting facial features with error rates of 2.2628% for the LFPW database and 2.7174% for the MORPH database. This better result helps to compensate for variations in shape and texture. The suggested hybrid method may be expanded to recognize gender, estimate head posture, estimate gaze, find facial feature points in the presence of significant pose fluctuations, and detect facial expressions.

The Internet of Things (IoT) is considered the base factor, that helps in developing many sectors. Additionally, IoT provides novel applications and services, i.e. smart homes, connected cars, and industrial automation. However, this development aggravates the security and privacy issues. This is because the connected devices may be vulnerable to cyber attacks in addition to generated data. Accordingly, it is of utmost importance to deploy appropriate security measures to alleviate these attacks. To overcome the mentioned problem, this work employs lightweight cryptography on limited processing capabilities of IoT devices i.e. limited memory, computing power, and battery life. These limited capabilities obstruct employing cryptography methods that require nontrivial computational resources. This work presents a multiround lightweight cryptography approach for IoT devices, which defines the security and protection of generated and transferred data. Multi-round cryptography mitigates the required device resources while ensuring robust protection and maintaining the stability of IoT devices.

This project focuses on designing a weather station and controllable wind turbine to harness the advantages of wind energy, which is one of the most important renewable energy sources in the world. The weather station is an Arduino-based tool and mobile App that measures temperature, humidity, wind speed and direction, and pressure. The methodology used is data analysis to give data and wind turbine control. The weather station provides the proper reads for the owner to steer the turbine in the right direction. Despite the physical, technical, and programming difficulties, both the weather station and mobile App perform well, achieving the project's objectives successfully. The accuracy of the results approaches 80%, with limitations highlighted in the thesis. Recommendations for further improvements and upgrades to the project are also discussed.

There has been a lot of activity in recent months to monetize generative AI applications due to their growth in popularity across multiple sectors. Specifically, businesses with a digital direct-to-consumer (D2C) channel, (e.g. e-commerce websites, Instagram shops), are strategically investing in AI tools to improve consumer engagement, enhance marketing effectiveness, and invest in product development. Technology is undergoing a revolution thanks to generative AI, which creates highly personalized and realistic content automatically for a range of media. This study explores the panorama of generative AI applications in e-commerce; including task and technology classification, and maps the main challenges and concerns associated with generative AI applications in e-commerce, covering social, behavioral, and technical limitations.

The number of cars on the road in the world is currently increasing. With the widespread usage of vehicles, a slew of issues inevitably arises. The damage of the road can be annoying to the drivers, and congestion in traffic can waste both fuel and time. It's worth noting that one of the primary causes of car damage and aging is the state of the road surface. Driving comfort and vehicle controllability can be harmed by poor road conditions. As a result, the system of monitoring road status is crucial to reducing congestion in traffic, reducing accident rates, and protecting cars from hazardous road conditions. This paper proposes using a framework based on cloud-fog network architecture, The fog computing network architecture is a dispersed network that may process data at the fog layer to minimize cloud computing's computational load. In addition, using small and large threshold values ( ), Main Authority (MA) deals with the major conditions, and the Sub Authority (SA) deals with the minor conditions.