Search

This study presents a comprehensive review of the potential of multimodal deep learning (DL) in medical diagnosis, using COVID-19 as a case example. Motivated by the success of artificial intelligence applications during the COVID-19 pandemic, this research aims to uncover the capabilities of DL in disease screening, prediction, and classification, and to derive insights that enhance the resilience, sustainability, and inclusiveness of science, technology, and innovation systems. Adopting a systematic approach, we investigate the fundamental methodologies, data sources, preprocessing steps, and challenges encountered in various studies and implementations. We explore the architecture of deep learning models, emphasising their data-specific structures and underlying algorithms. Subsequently, we compare different deep learning strategies utilised in COVID-19 analysis, evaluating them based on methodology, data, performance, and prerequisites for future research. By examining diverse data types and diagnostic modalities, this research contributes to scientific understanding and knowledge of the multimodal application of DL and its effectiveness in diagnosis. We have implemented and analysed 11 deep learning models using COVID-19 image, text, and speech (ie, cough) data. Our analysis revealed that the MobileNet model achieved the highest accuracy of 99.97% for COVID-19 image data and 93.73% for speech data (i.e., cough). However, the BiGRU model demonstrated superior performance in COVID-19 text classification with an accuracy of 99.89%. The broader implications of this research suggest potential benefits for other domains and disciplines that could leverage deep learning techniques for image, text, and speech analysis., Comment: 43 pages

Image inpainting is a widely used technique in computer vision for reconstructing missing or damaged pixels in images. Recent advancements with Generative Adversarial Networks (GANs) have demonstrated superior performance over traditional methods due to their deep learning capabilities and adaptability across diverse image domains. Residual Networks (ResNet) have also gained prominence for their ability to enhance feature representation and compatibility with other architectures. This paper introduces a novel architecture combining GAN and ResNet models to improve image inpainting outcomes. Our framework integrates three components: Transpose Convolution-based GAN for guided and blind inpainting, Fast ResNet-Convolutional Neural Network (FR-CNN) for object removal, and Co-Modulation GAN (Co-Mod GAN) for refinement. The model's performance was evaluated on benchmark datasets, achieving accuracies of 96.59% on Image-Net, 96.70% on Places2, and 96.16% on CelebA. Comparative analyses demonstrate that the proposed architecture outperforms existing methods, highlighting its effectiveness in both qualitative and quantitative evaluations., Comment: 34 pages

The Internet of Things (IoT) has transformed healthcare, facilitating remote patient monitoring, enhanced medication adherence, and chronic disease management. However, this interconnected ecosystem faces significant vulnerabilities with the advent of quantum computing, which threatens to break existing encryption standards protecting sensitive patient data in IoT-enabled medical devices. This chapter examines the quantum threat to healthcare IoT security, highlighting the potential impacts of compromised encryption, including privacy breaches, device failures, and manipulated medical records. It introduces post-quantum cryptography (PQC) and quantum-resistant techniques like quantum key distribution (QKD), addressing their application in resource-constrained healthcare IoT devices such as pacemakers, monitoring tools, and telemedicine systems. The chapter further explores the challenges of integrating these solutions and reviews global efforts in mitigating quantum risks, offering insights into suitable PQC primitives for various healthcare use cases., Comment: 24 pages

Cybersecurity has emerged as a critical global concern. Intrusion Detection Systems (IDS) play a critical role in protecting interconnected networks by detecting malicious actors and activities. Machine Learning (ML)-based behavior analysis within the IDS has considerable potential for detecting dynamic cyber threats, identifying abnormalities, and identifying malicious conduct within the network. However, as the number of data grows, dimension reduction becomes an increasingly difficult task when training ML models. Addressing this, our paper introduces a novel ML-based network intrusion detection model that uses Random Oversampling (RO) to address data imbalance and Stacking Feature Embedding based on clustering results, as well as Principal Component Analysis (PCA) for dimension reduction and is specifically designed for large and imbalanced datasets. This model's performance is carefully evaluated using three cutting-edge benchmark datasets: UNSW-NB15, CIC-IDS-2017, and CIC-IDS-2018. On the UNSW-NB15 dataset, our trials show that the RF and ET models achieve accuracy rates of 99.59% and 99.95%, respectively. Furthermore, using the CIC-IDS2017 dataset, DT, RF, and ET models reach 99.99% accuracy, while DT and RF models obtain 99.94% accuracy on CIC-IDS2018. These performance results continuously outperform the state-of-art, indicating significant progress in the field of network intrusion detection. This achievement demonstrates the efficacy of the suggested methodology, which can be used practically to accurately monitor and identify network traffic intrusions, thereby blocking possible threats., Comment: Accepted in Journal of Big Data (Q1, IF: 8.1, SCIE) on Jan 19, 2024

Data from interconnected vehicles may contain sensitive information such as location, driving behavior, personal identifiers, etc. Without adequate safeguards, sharing this data jeopardizes data privacy and system security. The current centralized data-sharing paradigm in these systems raises particular concerns about data privacy. Recognizing these challenges, the shift towards decentralized interactions in technology, as echoed by the principles of Industry 5.0, becomes paramount. This work is closely aligned with these principles, emphasizing decentralized, human-centric, and secure technological interactions in an interconnected vehicular ecosystem. To embody this, we propose a practical approach that merges two emerging technologies: Federated Learning (FL) and Blockchain. The integration of these technologies enables the creation of a decentralized vehicular network. In this setting, vehicles can learn from each other without compromising privacy while also ensuring data integrity and accountability. Initial experiments show that compared to conventional decentralized federated learning techniques, our proposed approach significantly enhances the performance and security of vehicular networks. The system's accuracy stands at 91.92\%. While this may appear to be low in comparison to state-of-the-art federated learning models, our work is noteworthy because, unlike others, it was achieved in a malicious vehicle setting. Despite the challenging environment, our method maintains high accuracy, making it a competent solution for preserving data privacy in vehicular networks., Comment: 7 pages

Quantum computing is emerging as a significant threat to information protected by widely used cryptographic systems. Cryptographic methods, once deemed secure for decades, are now at risk of being compromised, posing a massive threat to the security of sensitive data and communications across enterprises worldwide. As a result, there is an urgent need to migrate to quantum-resistant cryptographic systems. This is no simple task. Migrating to a quantum-safe state is a complex process, and many organisations lack the in-house expertise to navigate this transition without guidance. In this paper, we present a comprehensive framework designed to assist enterprises with this migration. Our framework outlines essential steps involved in the cryptographic migration process, and leverages existing organisational inventories. The framework facilitates the efficient identification of cryptographic assets and can be integrated with other enterprise frameworks smoothly. To underscore its practicality and effectiveness, we have incorporated case studies that utilise graph-theoretic techniques to pinpoint and assess cryptographic dependencies. This is useful in prioritising crypto-systems for replacement., Comment: 24 Pages

Intelligent transportation and autonomous mobility solutions rely on cooperative awareness developed by exchanging proximity and mobility data among road users. To maintain pervasive awareness on roads, all vehicles and vulnerable road users must be identified, either cooperatively, where road users equipped with wireless capabilities of Vehicle-to-Everything (V2X) radios can communicate with one another, or passively, where users without V2X capabilities are detected by means other than V2X communications. This necessitates the establishment of a communications channel among all V2X-enabled road users, regardless of whether their underlying V2X technology is compatible or not. At the same time, for cooperative awareness to realize its full potential, non-V2X-enabled road users must also be communicated with where possible or, leastwise, be identified passively. However, the question is whether current V2X technologies can provide such a welcoming heterogeneous road environment for all parties, including varying V2X-enabled and non-V2X-enabled road users? This paper investigates the roles of a propositional concept named Augmenting V2X Roadside Unit (A-RSU) in enabling heterogeneous vehicular networks to support and benefit from pervasive cooperative awareness. To this end, this paper explores the efficacy of A-RSU in establishing pervasive cooperative awareness and investigates the capabilities of the available communication networks using secondary data. The primary findings suggest that A-RSU is a viable solution for accommodating all types of road users regardless of their V2X capabilities., Comment: 13 pages

In this paper have developed a novel hybrid hierarchical attention-based bidirectional recurrent neural network with dilated CNN (HARDC) method for arrhythmia classification. This solves problems that arise when traditional dilated convolutional neural network (CNN) models disregard the correlation between contexts and gradient dispersion. The proposed HARDC fully exploits the dilated CNN and bidirectional recurrent neural network unit (BiGRU-BiLSTM) architecture to generate fusion features. As a result of incorporating both local and global feature information and an attention mechanism, the model's performance for prediction is improved.By combining the fusion features with a dilated CNN and a hierarchical attention mechanism, the trained HARDC model showed significantly improved classification results and interpretability of feature extraction on the PhysioNet 2017 challenge dataset. Sequential Z-Score normalization, filtering, denoising, and segmentation are used to prepare the raw data for analysis. CGAN (Conditional Generative Adversarial Network) is then used to generate synthetic signals from the processed data. The experimental results demonstrate that the proposed HARDC model significantly outperforms other existing models, achieving an accuracy of 99.60\%, F1 score of 98.21\%, a precision of 97.66\%, and recall of 99.60\% using MIT-BIH generated ECG. In addition, this approach substantially reduces run time when using dilated CNN compared to normal convolution. Overall, this hybrid model demonstrates an innovative and cost-effective strategy for ECG signal compression and high-performance ECG recognition. Our results indicate that an automated and highly computed method to classify multiple types of arrhythmia signals holds considerable promise., Comment: 23 pages

Network intrusion detection systems (NIDSs) play an important role in computer network security. There are several detection mechanisms where anomaly-based automated detection outperforms others significantly. Amid the sophistication and growing number of attacks, dealing with large amounts of data is a recognized issue in the development of anomaly-based NIDS. However, do current models meet the needs of today's networks in terms of required accuracy and dependability? In this research, we propose a new hybrid model that combines machine learning and deep learning to increase detection rates while securing dependability. Our proposed method ensures efficient pre-processing by combining SMOTE for data balancing and XGBoost for feature selection. We compared our developed method to various machine learning and deep learning algorithms to find a more efficient algorithm to implement in the pipeline. Furthermore, we chose the most effective model for network intrusion based on a set of benchmarked performance analysis criteria. Our method produces excellent results when tested on two datasets, KDDCUP'99 and CIC-MalMem-2022, with an accuracy of 99.99% and 100% for KDDCUP'99 and CIC-MalMem-2022, respectively, and no overfitting or Type-1 and Type-2 issues., Comment: Accepted in the Journal of Information Security and Applications (Scopus, Web of Science (SCIE) Journal, Quartile: Q1, Site Score: 7.6, Impact Factor: 4.96) on 7 December 2022

Cancer is a fatal disease caused by a combination of genetic diseases and a variety of biochemical abnormalities. Lung and colon cancer have emerged as two of the leading causes of death and disability in humans. The histopathological detection of such malignancies is usually the most important component in determining the best course of action. Early detection of the ailment on either front considerably decreases the likelihood of mortality. Machine learning and deep learning techniques can be utilized to speed up such cancer detection, allowing researchers to study a large number of patients in a much shorter amount of time and at a lower cost. In this research work, we introduced a hybrid ensemble feature extraction model to efficiently identify lung and colon cancer. It integrates deep feature extraction and ensemble learning with high-performance filtering for cancer image datasets. The model is evaluated on histopathological (LC25000) lung and colon datasets. According to the study findings, our hybrid model can detect lung, colon, and (lung and colon) cancer with accuracy rates of 99.05%, 100%, and 99.30%, respectively. The study's findings show that our proposed strategy outperforms existing models significantly. Thus, these models could be applicable in clinics to support the doctor in the diagnosis of cancers., Comment: Accepted for publication in the Special Issue of Expert Systems with Applications (IF:6.954, Cite:12.70) How to Cite: Md. Alamin Talukder, Md. Manowarul Islam, Md Ashraf Uddin, Arnisha Akhter, Khondokar Fida Hasan, Mohammad Ali Moni. "Machine Learning-based Lung and Colon Cancer Detection using Deep Feature Extraction and Ensemble Learning", Expert Systems with Applications. 2022 Jun 1

The concept of traditional farming is changing rapidly with the introduction of smart technologies like the Internet of Things (IoT). Under the concept of smart agriculture, precision agriculture is gaining popularity to enable Decision Support System (DSS)-based farming management that utilizes widespread IoT sensors and wireless connectivity to enable automated detection and optimization of resources. Undoubtedly the success of the system would be impacted on crop productivity, where failure would impact severely. Like many other cyber-physical systems, one of the growing challenges to avoid system adversity is to ensure the system's security, privacy, and trust. But what are the vulnerabilities, threats, and security issues we should consider while deploying precision agriculture? This paper has conducted a holistic threat modeling on component levels of precision agriculture's standard infrastructure using popular threat intelligence tools STRIDE to identify common security issues. Our modeling identifies a noticing of fifty-eight potential security threats to consider. This presentation systematically presented them and advised general mitigation suggestions to support cyber security in precision agriculture.

Artificial Intelligence (AI) is one of the disruptive technologies that is shaping the future. It has growing applications for data-driven decisions in major smart city solutions, including transportation, education, healthcare, public governance, and power systems. At the same time, it is gaining popularity in protecting critical cyber infrastructure from cyber threats, attacks, damages, or unauthorized access. However, one of the significant issues of those traditional AI technologies (e.g., deep learning) is that the rapid progress in complexity and sophistication propelled and turned out to be uninterpretable black boxes. On many occasions, it is very challenging to understand the decision and bias to control and trust systems' unexpected or seemingly unpredictable outputs. It is acknowledged that the loss of control over interpretability of decision-making becomes a critical issue for many data-driven automated applications. But how may it affect the system's security and trustworthiness? This chapter conducts a comprehensive study of machine learning applications in cybersecurity to indicate the need for explainability to address this question. While doing that, this chapter first discusses the black-box problems of AI technologies for Cybersecurity applications in smart city-based solutions. Later, considering the new technological paradigm, Explainable Artificial Intelligence (XAI), this chapter discusses the transition from black-box to white-box. This chapter also discusses the transition requirements concerning the interpretability, transparency, understandability, and Explainability of AI-based technologies in applying different autonomous systems in smart cities. Finally, it has presented some commercial XAI platforms that offer explainability over traditional AI technologies before presenting future challenges and opportunities., Comment: Book_Chapter, Springer Nature

LoRa (Long-Range) is an LPWAN (low-power wide-area network) protocol that is part of the IoT family that focusses on long-range communication of up to 14km, albeit with delay-inherent transmissions. Three IoT-based time synchronisation methodologies are analysed, and their efficacy measured through a systematic critical literature review. These include a GNSS-based method, an off-the-shelf GPS hardware resampling method, and the LongShoT method, within the context of vehicular ad-hoc networks (VANET), wireless sensor networks (WSN), and long-range wide area network (LoRaWAN) respectively. Although two of the three methods are not LoRaWAN-specific, the findings obtained from the research are applied to the context of LoRa in the proposed methodology. A methodology for selecting a time synchronisation methodology with regards to LoRa specifically is posited, whereby each requirement of synchronisation objective, energy consumption and costs, scenario and security analysis, application requirements, microcontroller requirements and transceiver requirements are taken into consideration. These are then followed by a fine-grain approach to the selection of a particular time-sync method. The resultant methodology may not only have implications in the field of research, where practitioners may adopt this literature review as a baseline understanding of time synchronisation methods and obstacles encountered toward LoRa, however developers of applications for the LoRaWAN may adapt the analysed methods outlined within., Comment: 23 Pages

The recent advancement of cloud technology offers unparallel strength to support intelligent computations and advanced services to assist with automated decisions to improve road transportation safety and comfort. Besides, the rise of machine intelligence propels the technological evolution of transportation systems one step further and leads to a new framework known as Cognitive Internet of Vehicles (C-IoV). The redefined cognitive technology in this framework promises significant enhancements and optimized network capacities compared with its predecessor framework, the Internet of Vehicles (IoV). CIoV offers additional security measures and introduces security and privacy concerns, such as evasion attacks, additional threats of data poisoning, and learning errors, which may likely lead to system failure and road user fatalities. Similar to many other public enterprise systems, transportation has a significant impact on the population. Therefore, it is crucial to understand the evolution and equally essential to identify potential security vulnerabilities and issues to offer mitigation towards success. This chapter offers discussions framing answers to the following two questions, 1) how and in what ways the penetration of the latest technologies are reshaping the transportation system? 2) whether the evolved system is capable of addressing the concerns of cybersecurity? This chapter, therefore, starts presenting the evolution of the transportation system followed by a quick overview of the evolved CIoV, highlighting the evolved cognitive design. Later it presents how a cognitive engine can overcome legacy security concerns and also be subjected to further potential security, privacy, and trust issues that this cloud-based evolved transportation system may encounter., Comment: 19 pages, book chapter

Vulnerable road users (VRUs) such as pedestrians, cyclists and motorcyclists are at the highest risk in the road traffic environment. Globally, over half of road traffic deaths are vulnerable road users. Although substantial efforts are being made to improve VRU safety from engineering solutions to law enforcement, the death toll of VRUs' continues to rise. The emerging technology, Cooperative Intelligent Transportation System (C-ITS), has the proven potential to enhance road safety by enabling wireless communication to exchange information among road users. Such exchanged information is utilized for creating situational awareness and detecting any potential collisions in advance to take necessary measures to avoid any possible road casualties. The current state-of-the-art solutions of C-ITS for VRU safety, however, are limited to unidirectional communication where VRUs are only responsible for alerting their presence to drivers with the intention of avoiding collisions. This one-way interaction is substantially limiting the enormous potential of C-ITS which otherwise can be employed to devise a more effective solution for the VRU safety where VRU can be equipped with bidirectional communication with full C-ITS functionalities. To address such problems and to explore better C-ITS solution suggestions for VRU, this paper reviewed and evaluated the current technologies and safety methods proposed for VRU safety over the period 2007-2020. Later, it presents the design considerations for a cellular-based Vehicle-to-VRU (V2VRU) communication system along with potential challenges of a cellular-based approach to provide necessary recommendations., Comment: 16 pages, 5 Figures

Vehicular Networks are one of the enabling technologies for cooperative Intelligent Transportation Systems. For accurate and precise time synchronization in vehicular networks, Global Navigation Satellite System (GNSS) is getting increasing attention for effective use of its time dissemination to vehicular networks. In urban road scenarios, the blockage of satellite signals affects precise positing due to erroneous geometry. However, for timing, it creates less deflection as long as a single satellite is visible. With the recent development of Multi-GNSS receivers, signal reception no longer relies on a single GNSS constellation but all the existing four GNSS systems. This increases the number of visible satellites in the sky, thus increasing the coverage of satellite signals. Developing a method of GNSS time synchronization, we have conducted a systematic field-based study in this paper to determine the attainable timing accuracy in vehicular networks using commercial-grade GNSS receivers and off-the-shelf communication devices. Our experimental results show that the attainable timing accuracy from GNSS services in consumer-grade on-board devices can be as good as 2 microseconds. A momentary outage of GNSS time solution due to signal blockage on the road can lead a clock error of up to sub-microseconds, which still meets the desired requirement for most vehicular applications., Comment: 10 pages, 16 figures

The new wave of computing allows users to explore their time in the Internet of Things (IoT) by connecting their smart devices over the network for data transfer without human interventions. While this swing increases the pace in IoT, time synchronization became a demanding feature on IoT devices for real-time applications. In this paper, we describe two synchronization protocols in IoT, its features and the advantages over the other. In addition, this work reveals the importance of time synchronization in the IoT platform and the effects of bad syncs in real-time applications. We start our research by analysing the widely using Simple Network Time Protocol (SNTP) and its performance in terms of offset and delay to examine the accuracy and reliability to define the synchronization., Comment: 43 pages

Over the past few years, we have experienced great technological advancements in the information and communication field, which has significantly contributed to reshaping the Intelligent Transportation System (ITS) concept. Evolving from the platform of a collection of sensors aiming to collect data, the data exchanged paradigm among vehicles is shifted from the local network to the cloud. With the introduction of cloud and edge computing along with ubiquitous 5G mobile network, it is expected to see the role of Artificial Intelligence (AI) in data processing and smart decision imminent. So as to fully understand the future automobile scenario in this verge of industrial revolution 4.0, it is necessary first of all to get a clear understanding of the cutting-edge technologies that going to take place in the automotive ecosystem so that the cyber-physical impact on transportation system can be measured. CIoV, which is abbreviated from Cognitive Internet of Vehicle, is one of the recently proposed architectures of the technological evolution in transportation, and it has amassed great attention. It introduces cloud-based artificial intelligence and machine learning into transportation system. What are the future expectations of CIoV. To fully contemplate this architectures future potentials, and milestones set to achieve, it is crucial to understand all the technologies that leaned into it. Also, the security issues to meet the security requirements of its practical implementation. Aiming to that, this paper presents the evolution of CIoV along with the layer abstractions to outline the distinctive functional parts of the proposed architecture. It also gives an investigation of the prime security and privacy issues associated with technological evolution to take measures., Comment: 6 pages

Time synchronization in communication networks provides a common time frame among all nodes, thussupporting various network functions such as message transmission, channel scheduling and resourcesharing in real-time and in correct order. In vehicular ad-hoc networks (VANETs) for connected andautomated vehicles, network nodes must be synchronized to exchange time-critical vehicle locationsand warning messages for various road safety applications. However, this is challenging as the datacommunication systems need to achieve low latency and high reliability under the conditions ofhigh user dynamics and density. While many synchronization techniques have been developed forgeneral communications networks, it is necessary to understand the requirements for VANET timesynchronization and the applicability of existing time synchronization techniques in VANET applications.This paper provides a survey on theory and practice of time synchronization in VANETs. It addressessome key factors in VANET time synchronization such as requirements analysis, precision, accuracy,availability, scalability and compatibility, and highlights the advantages of Global Navigation SatelliteSystem (GNSS) in VANET time synchronization. Through this survey, some insights are developed intoexisting and emerging protocols for time synchronization in VANETs., Comment: 13 pages, Vehicular Communications (2018)

Time synchronization is critical for the operation of distributed systems in networked environments. It is also demanded in vehicular ad-hoc networks (VANETs), which, as a special type of wireless networks, are becoming increasingly important for emerging cooperative intelligent transport systems. Global navigation satellite system (GNSS) is a proven technology to provide precise timing information in many distributed systems. It is well recognized to be the primary means for vehicle positioning and velocity determination in VANETs. However, GNSS-based time synchronization is not well understood for its role in the coordination of various tasks in VANETs. To address this issue, this paper examines the requirements, potential benefits, and feasibility of GNSS time synchronization in VANETs. The availability of GNSS time synchronization is characterized by almost 100% in our experiments in high-rise urban streets, where the availability of GNSS positioning solutions is only 80%. Experiments are also conducted to test the accuracy of time synchronization with 1-PPS signals output from consumer grade GNSS receivers. They have shown 30-ns synchronization accuracy between two receivers of different models. All these experimental results demonstrate the feasibility of GNSS time synchronization for stringent VANET applications., Comment: 10 pages