Abstract

The Internet of Things (IoT) leads to the next phase of human interaction with technology. With the help of the IoT, physical objects can be given the ability to generate, receive, and seamlessly trade data with one another. The IoT includes a wide variety of applications, each of which focuses on automating a specific task and works to give inanimate objects the ability to act independently of human intervention. The currently available and upcoming IoT applications hold a great deal of promise for enhancing the level of convenience, productivity, and automation enjoyed by users. High levels of security, privacy, authentication, and the ability to recover from attacks are required for the implementation of such a world in a manner that is constantly expanding. In this light, it is necessary to make the necessary adjustments to the architecture of IoT applications to accomplish end-to-end security in IoT environments. In this article, a comprehensive review of the security-related challenges and potential sources of danger posed by IoT applications is provided. Following a discussion of security concerns, a variety of new and established technologies that are focused on achieving a high degree of trust in the applications of the IoT are covered. Machine learning, fog computing, edge computing, and blockchain are just a few of the technologies that help the IoT provide greater security.

Introduction

The Internet of Things (IoT) is one of the disruptive technologies of the 21st century that has revolutionized every business dimension across the globe. The Internet has become an essential component in daily activities such as e-commerce, e-learning, e-conferences, etc. According to [1], the number of IoT devices deployed may exceed 21.3 billion by 2022. Intelligent interfaces connect IoT devices in order to collaborate, gather, communicate, and store data. Deploying IoT devices enhances business productivity by optimizing operational procedures and maximizing the utilization of available resources. IoT brings innovation in the way things are monitored and managed remotely with the help of real-time data acquired from the sensors. With the unrestrained volumes of data generated by billions of connected devices, storage, management, and security are arduous. Security aspects such as authentication, authorization, privacy, confidentiality, availability, and integrity form the basis for information exchange in a trusted environment. Authentication is the process of identifying the genuineness of an entity. The authorization process allows only authenticated users to access the resources. The protection of privacy prevents hackers from accessing the user’s private information. Data confidentiality is an outcome of authentication and authorization. It prevents sensitive data from being reclaimed and exploited. A device’s availability confirms that IoT devices and services are resilient to attacks and are accessible all the time. Data integrity is the process of providing a tamper-proof platform for information communication, exchange, and storage.

Conclusion and future directions

In this paper, the methods that are currently available in the literature are discussed, and this paper also presents the methods that are currently available. The subsequent step is an in-depth discussion of the numerous security risks and the importance of threat modeling. The subsequent topic is a discussion of the various IoT applications’ necessary security precautions. It reviews the prior research on the numerous security issues and the solutions put forth by various researchers. The holes that were found while conducting this literature review served as inspiration for the current research work.

7.1. Research challenges
The billions of IoT devices deployed across the globe collect personal and sensitive data and exchange it with other networks using intelligent interfaces. Providing device authentication, authorization, data privacy, and security in such an un trusted environment is a challenge. Centralized trusted infrastructure cannot scale to the dynamic and ever intensifying traffic and, thus, leading to bottlenecks in the network. Distributed authentication can scale to the increasing demands of the IoT networks, but requires distributed trusted systems. The majority of the work in the IoT security domain focuses on embracing the security mechanisms of WSNs and conventional Internet-based applications. However, these mechanisms are far from being implemented in real-time scenarios due to the specific challenges of IoT, such as

Abstract

In recent years, with the increasing demand for intelligent society, intelligent photonics has developed rapidly. Machine learning (ML), as a subset of artificial intelligence (AI), has played an important role in the intelligent evolution of optical fiber sensors. Its impact extends beyond enhancing sensor performance by introducing innovative problem-solving approaches. Specifically, ML algorithms have become instrumental in signal demodulation and elevating the efficacy of discrete and distributed sensors, and have also greatly promoted the development of optical fiber speckle pattern processing. This paper presents the latest advancements in ML-based optical fiber sensors, outlines the problems faced by conventional demodulation methods and the common ML algorithms applied in optical fiber sensors, and emphasizes key applications. Additionally, this paper delves into the challenges and future development of this emerging research direction.

Introduction

Due to advancements in powerful computing tools, hardware, widespread use of cloud data, and the evolution of the Internet of Things, the accessibility of large datasets has significantly improved. This has promoted the emergence of many efficient machine learning (ML) algorithms. These algorithms, aided by continuous advancements in data analysis technology, have demonstrated effectiveness in training models and solving specific problems. If ML algorithms are complex and deep enough in terms of layers, then ML can become so-called deep learning (DL) [1]. Regardless of the name, ML seems to have a wide range of applications in modern society, because the achievements of artificial intelligence (AI) often exceed those achieved by manual control design.

In 2021, Nature Photonics further advanced its exploration of the hot topic of optics/photonics through a focused issue called “Machine learning of Light”. As an important type of microphysics, optics seems to have similarities in complexity with some of the “black box” processes in ML. In the Q&A session, David Pille and Aydogan Ozcan discussed the two primary drivers behind the current application of ML in photonics [2]. One direction is to use AI methods on hardware to design optical structures and devices with specific performance. Another direction is to use optical systems as fully optical/hybrid statistical inference models to achieve AI calculations.

Optics and photonics applied to ML computing offer unique advantages, allowing for ultrafast calculations at extreme frame rates and low energy consumption [3], [4], [5], [6]. For instance, Feldmann et al. [7] proposed a fully optical neural synaptic system. They use wavelength division multiplexing technology to implement a scalable circuit architecture for photonic neural networks (NNs) and use optical pulse signals to regulate the phase change materials that encapsulate neurons. This phase change material can regulate transmittance and reflectivity through a controllable laser. This system can perform both supervised and unsupervised learning. The work successfully demonstrates direct pattern recognition in the full optical system. This photonic NN has the potential to leverage the inherent high-speed and high bandwidth of optical systems, thereby enabling the direct processing of full optical communication and visual data. However, logic gates, computation, and NNs are all linear (passive) systems that cannot address nonlinear (active) issues. The lack of effective nonlinear optical processes as activation functions for many nodes poses challenges for all-optical implementation [8].

Conclusion and future perspectives

This paper reviews the application of the ML algorithm in optical fiber sensors. In recent years, the advent of ML has greatly impacted optical development. Compared with static demodulation algorithms and complex spatial optical paths, the ML algorithm offers better signal processing mechanisms for optics. ML algorithms enable automated processing, analysis, and recognition of spectrums, thereby enhancing the efficiency and accuracy of spectrum analysis. Using ML algorithms to classify and identify the spectrum of different substances, enabling qualitative and quantitative analysis of substances. By learning from known corresponding spectrums and building models, ML algorithms can predict and analyze unknown spectrums. ML algorithms can extract and recognize features from the spectrum, analyze the content and proportion of various components, and reveal the intrinsic nature of substances. With the continuous development of spectroscopy and ML algorithms, the application prospects of ML in spectroscopy will become increasingly widespread. This advancement will contribute to more efficient and accurate analytical methods in chemical analysis, environmental monitoring, food safety, and other fields, promoting progress in scientific research and expanding application fields.

Abstract

Despite providing unparalleled connectivity and convenience, the exponential growth of the Internet of Things (IoT) ecosystem has triggered significant cybersecurity concerns. These concerns stem from various factors, including the heterogeneity of IoT devices, widespread deployment, and inherent computational limitations. Integrating emerging technologies to address these concerns becomes imperative as the dynamic IoT landscape evolves. Machine Learning (ML), a rapidly advancing technology, has shown considerable promise in addressing IoT security issues. It has significantly influenced and advanced research in cyber threat detection. This survey provides a comprehensive overview of current trends, methodologies, and challenges in applying machine learning for cyber threat detection in IoT environments. Specifically, we further perform a comparative analysis of state-of-the-art ML-based Intrusion Detection Systems (IDSs) in the landscape of IoT security. In addition, we shed light on the pressing unresolved issues and challenges within this dynamic field. We provide a future vision with Generative AI and large language models to enhance IoT security. The discussions present an in-depth understanding of different cyber threat detection methods, enhancing the knowledge base of researchers and practitioners alike. This paper is a valuable resource for those keen to delve into the evolving world of cyber threat detection leveraging ML and IoT security.

Introduction

Brendan O’Brien astutely observed, ”If you think the Internet has changed your life, think again. The Internet of Things is about to change it all over again!” [1]. This is indeed the case, as the Internet of Things (IoT) has heralded unprecedented connectivity. The advancements in sensor technology, wireless communication, and data analytics have spurred an exponential increase in connected devices. This influx of connectivity, brought about by integrating IoT into various industries, cities, and households, promotes unmatched efficiency and convenience. As the backbone of IoT, sensors and actuators acquire and convert data from the physical world into digital signals. These compact devices amass a diverse range of data, thereby enabling real-time monitoring and control of numerous systems and processes.

However, the rapid proliferation and extensive integration of IoT devices into everyday life have ushered in various security challenges. These issues must be robustly addressed to ensure the safety and reliability of this expanding ecosystem. The sheer volume and variety of IoT devices and their often inconsistent security features and protocols engender a fragmented environment teeming with potential attack vectors. IoT devices frequently prioritize low cost and user simplicity over security, making them susceptible to breaches and exploitation. As a result, these devices are at risk of various cyber threats, including data breaches, Distributed Denial-of-Service (DDoS) attacks, and malware infections. Any security breach in these devices could significantly compromise privacy and crucial infrastructure systems, given the sensitive nature of the data they handle. Moreover, IoT devices, potentially serving as entry points, might allow attackers to infiltrate broader networks, amplifying the potential impact of security breaches. Another primary concern is the security of communication routes between IoT devices and networks, as many IoT devices utilize wireless communication protocols susceptible to interception or manipulation. These vulnerabilities can be exacerbated by the resource constraints of specific IoT devices, which prevent them from adopting contemporary encryption and authentication techniques. Furthermore, the long lifespan and widespread deployment of IoT devices compound the difficulty of managing security upgrades and patches, as many devices may not receive regular updates or may be difficult to access for maintenance. This could lead to an increased number of outdated or vulnerable devices, further exacerbating security concerns [2].

Conclusion

IoT devices have become fundamental to everyday life, offering increased connectivity and convenience. However, this accelerated development in IoT devices also brought several security challenges that need to be addressed to guarantee the safety and reliability of these interconnected systems. The survey was organized into five sections detailing the role of machine learning in enhancing IoT security. We have provided an overview of the current trends in ML for cyber threat detection in IoT environments. Furthermore, we survey recent cyber detection methods, define them, highlight the approach, and detail the attack surface utilized along with their evaluations. The utilized ML techniques are also discussed, defined, and compared regarding advantages and drawbacks in the relevant use cases. Moreover, open issues were discussed briefly, concluding that further research and development is required as no current solution can address the issues related to IoT cyber threat detection. A holistic strategy combining diverse techniques and strategies is required as a final recommendation. We present this survey as a reference for the current advancements in the field and point out the direction being taken. Our goal is for the survey to create a more secure and resilient IoT environment by identifying the current issues, understanding their root causes and possible implications, and developing innovative solutions.

Abstract

Employing the Internet of Things (IoT) in healthcare provides many advantages for patient monitoring and analysis of the patient’s health with the help of generated data. The basic role of IoT in healthcare is to ease the patient’s life by giving them a monitor over their medical condition. The use of IoT in medical devices requires a focus on the end-user. Medical devices such as glucose meters are designed to record the data of the patient and their vital signs. The generated information can be used to enhance decision-making for the physician. The collected information about the patient can be at risk due to certain security issues during the transferring of information can compromise the identity and social life of the patient. This review explores the IoT regarding its structural requirement and its role in various fields with special emphasis on healthcare. The security and privacy issues than can hinder the utilization of IoT at its potential and ways to overcome these issues are being addressed.

Role of IoT in medical devices

Medical Devices (MDs) are any instrument, reagent, apparatus, implement, machine or any other in-vitro reagent or calibrator, software which is intended to be used alone or in combination for the diagnosis, mitigation, prevention, treatment or alleviation of diseases in human beings.1 Generally, MDs are classified based on the risk associated with them. As per USFDA MDs are classified into three categories I, II, III where category I has lowest risk and category III has highest risk as shown in Table 12 (see Fig. 4)A report from Fortune Business Insights states that from 2018 to 2025 the Medical Device Industry will experience major investment in Research & Development of new technology from pharmaceutical players, approvals from regulatory authorities will also contribute to increase this number. The medical device global market was USD 42.5 Billion in 2018 and is estimated to reach USD 612.7 Billion by 2025. CAGR is expected at 5.4 % from 2018 to 2025.3

Further Medical Devices have been showing an increasing growth in the previous decade and with promising technology and innovations being made, it will keep growing more. Although there is conventional research and development being done by existing companies. With the entry of giants like Apple and Google, new technological advancements in the form of Artificial Intelligence, Data Analytics and Internet of Things (IoT) are being implemented in this sector.4

Conclusion

The amalgamation of a variety of technology is required to build an IoT. Due to this complexity, the system becomes more vulnerable to security and privacy threats. The generic engineering framework involved in patient monitoring remotely should meet the security requirements with the help of cryptographic algorithms. These algorithms should be device-specific to nurture the design & environment of patient-specific needs. The security and safety issues of a patient which arise from the application of IoT cannot be solved with the help of a single solution. To practice, the best security controls an informed user is also necessary. The management ability of the network and device automation can be of great help to deliver patient well-being and care. Delivering healthcare to patients efficiently and effectively, independent of time and location requires a new paradigm that is opened by IoT. The existing foundations of telemedicine have now been extended and better assessment of normal public health in a cohort sense has shifted the passive involvement of patient data gathering. Improving the outcomes for the patient and reducing the care cost are being balanced against the privacy and confidentiality of data and threats of cyber security. There is a vast source of data collection and potential to transfer clinical data, but there is a challenge to triage this data for clinical use. The cyber security and safety of patient data need to move faster than regulation and implementation of policy. The way forward for IoT needs superior standards every day to use the connected devices.

Abstract

     Concrete is one of the most widely used materials in various civil engineering applications. Its global production rate is increasing to meet demand. Mechanical properties of concrete are among important parameters in designing and evaluating its performance. Over the past few decades, machine learning has been used to model real-world problems. Machine learning, as a branch of artificial intelligence, is gaining popularity in many scientific fields such as robotics, statistics, bioinformatics, computer science, and construction materials. Machine learning has many advantages over statistical and experimental models, such as optimal accuracy, high-performance speed, responsiveness in complex environments, and economic cost-effectiveness. Recently, more researchers are looking into deep learning, which is a group of machine learning algorithms, as a powerful method in matters of diagnosis and classification. Hence, this paper provides a review of successful ML and DL model applications to predict concrete mechanical properties. Several modeling algorithms were reviewed highlighting their applications, performance, current knowledge gaps, and suggestions for future research. This paper will assist construction material engineers and researchers in selecting suitable and accurate techniques that fit their applications.

Introduction

     Concrete is the most widely used building material worldwide. With population growth and urbanization, the demand for concrete is expected to reach 18 billion by 2050 [[1], [2], [3]]. In order to improve the design of concrete structures, it is necessary to gain a better understanding of concrete performance, relying on accurate evaluation of its mechanical properties. Among the various properties of concrete, compressive strength has been considered a direct indicator of performance. It is directly related to the safety and performance of the structure throughout its life cycle [1,4,5].

     Concrete is a complex system of combinations of different components (coarse and fine aggregates, water, cement, and additional mixtures) that are randomly distributed throughout the concrete matrix [[6], [7], [8], [9]]. This heterogeneous feature makes it difficult to accurately predict certain mechanical properties, especially compressive strength [[10], [11], [12]]. The most direct way to evaluate the compressive strength of concrete is through physical tests performed on specimens cured to the desired age [13,14]. Such a method for evaluating the compressive strength needs time while being affected by other factors related to specimen fabrications and test operators. Moreover, the test tends to damage the specimens. Empirical regression methods were therefore proposed to predict compressive strength [[15], [16], [17]], but the disadvantage of this method is the non-linear relationship between the concrete mixture and the concrete’s compressive strength. This prevents an accurate regression expression. Numerical simulation is another method that can predict the behavior of concrete. However, a good prediction of concrete behavior is not easy to achieve due to the non-linearity and randomness [[18], [19], [20]].

Conclusions and recommendations

     The present study investigated the performance of different ML/DL models to predict the mechanical properties of concrete. Among the wide range of ML/DL models, support vector machines, decision trees, evolutionary algorithms, and artificial neural networks were examined due to the popularity of these models as well as their frequent use in the field of civil engineering. Therefore, the performance of these models in various studies to evaluate the compressive strength, tensile strength, shear strength, flexural strength, and elastic modulus was evaluated. A comparison of experimental models with ML/DL methods shows that ML/DL models with better updating capability and the ability to analyze large datasets perform better. On the other hand, statistical models are not good for predicting complex structures due to their high cost and time-consuming nature. According to ML/DL methods, selecting the appropriate model for predicting the target strength of concrete should be made by considering different criteria. A review of various studies shows that the relationship between concrete components and mechanical strength is influential in choosing the forecasting model. Therefore, models that can respond in nonlinear space should be used if the relationship is nonlinear. In these cases, SVM and ANN models can be used for their acceptable performance in a non-linear environment with fewer errors. To achieve more accurate results, optimization of these models with metaheuristic algorithms can also be used. But if we need the transparency of the model and the explicit mathematical formula between input and output, we can use decision tree models and evolutionary algorithms. On the other hand, the use of ensemble models to optimize these models, although it results in higher accuracy, increases computation time and model complexity. The combined SVM and ANN models, although they increase the computation time, have accurate results in the face of extensive data. Therefore, based on the study, the best option for predicting the strength of concrete in terms of model accuracy and model implementation is the combined use of SVM and ANN combined models.

Abstract

     Artificial Intelligence (AI) and Machine learning (ML) represents an important evolution in computer science and data processing systems which can be used in order to enhance almost every technology-enabled service, products, and industrial applications. A subfield of artificial intelligence and computer science is named machine learning which focuses on using data and algorithms to simulate learning process of machines and enhance the accuracy of the systems. Machine learning systems can be applied to the cutting forces and cutting tool wear prediction in CNC machine tools in order to increase cutting tool life during machining operations. Optimized machining parameters of CNC machining operations can be obtained by using the advanced machine learning systems in order to increase efficiency during part manufacturing processes. Moreover, surface quality of machined components can be predicted and improved using advanced machine learning systems to improve the quality of machined parts. In order to analyze and minimize power usage during CNC machining operations, machine learning is applied to prediction techniques of energy consumption of CNC machine tools. In this paper, applications of machine learning and artificial intelligence systems in CNC machine tools is reviewed and future research works are also recommended to present an overview of current research on machine learning and artificial intelligence approaches in CNC machining processes. As a result, the research filed can be moved forward by reviewing and analysing recent achievements in published papers to offer innovative concepts and approaches in applications of artificial Intelligence and machine learning in CNC machine tools.

Introduction

     CNC machining operation is one of the most important part-production methodologies, and it is often referred to as the engine of modern manufacturing processes. The automotive and medical sectors, aerospace, gas and oil, and warehousing services, are using the CNC machining operations to create parts in different applications [1]. CNC machining is generally used in manufacture of every machine, molded part, or finished product as one of the most important manufacturing processes. CNC machinery has paved the way in manufacturing and machining, allowing businesses to achieve their goals and targets in a variety of ways. However, because manufacturing methodologies is always evolving and new technologies are being introduced, it is critical to consider future of CNC machining operations [2, 3]. Machine learning (ML) is the study of computer algorithms that gives computers the capacity to automatically learn from data and prior experiences in order to find patterns and make predictions without human involvement. The ML and applications in different areas of study are considered to be a component of artificial intelligence [4], [5], [6].

     Machine learning and artificial intelligence in particular raise plenty of concerns about the future of CNC machining operations and how these concepts will evolve future works of manufacturing companies [7]. The way a machine learns, adapts, and optimizes output can also be influenced by real-time data, analytics, and deep learning. Data sets are essential for operators to understand how a machine works and, eventually, how a whole floor of machines works together [8, 9]. Due to the development of affordable, reliable, and resilient sensors and acquisition and communication systems, novel implementations of machine learning approaches for tool condition monitoring can be presented [10]. Machine learning systems are capable of completely examine data and identify various types of areas which should be modified. Machine tools are increasingly being equipped with edge computing options to record internal drive signals at high frequency in order to supply the necessary vast quantity of data for the use of machine learning techniques in manufacturing [11]. Productivity and efficiency are two areas where artificial intelligence can modify CNC machine tools operations in order to enhance accuracy of CNC machining operations[12]. Machines can generate and analyze production data and provide real-time findings to human operators are effective devices for increasing productivity in part production processes. As a result, shop owners can quickly adjust the way a machine operates using the modified data generated by advanced machine learning algorithms in order to enhance productivity of part manufacturing [13]. Having more knowledge and making better decisions in process planning strategies means less downtime on the work floor during process of part production. Production and maintenance process of part manufacturing using CNC machine tools can be developed using the machine learning and the artificial intelligence in order to enhance efficiency in part manufacturing operations [14].

Conclusion

     Machine learning and artificial intelligence are applied to various industrial applications in order to improve performances of industrial process. To increase accuracy as well as efficiency during CNC machining operations, different applications of the machine learning and artificial intelligence systems are studied in different research works. Reducing machine downtime, optimization of CNC machine tools, cutting tool wear prediction, cutting force model, CNC machine tool maintenance, monitoring of machining operations, surface quality prediction and energy prediction systems are some examples of machine leniting applications in the development of CNC machining operations. ML techniques are recently applied to energy consumption prediction models in order to decrease the energy consumption during CNC machining operations. Accuracy and radiality of energy consumption models are significantly enhanced using the ML methodologies in comparison to the tradition’s methods of energy usage predictions during CNC machining operations.

     The applications of machine learning and artificial intelligence systems in CNC machine tools are examined in this study by analyzing recent achievements from published papers. The main aim of this study is to provide an overview of current studies on machine learning and artificial intelligence techniques in CNC machining operations in order to provide a useful study for the researchers in the interesting field. Network of sensors and cloud data sources can connect CNC machines together can be employed to provide smart CNC machine tools. The machining industry’s efficiency can be increased as it transitions to smart machining techniques, allowing it to achieve self-optimization and adaption to uncontrolled circumstances. However, developing the applications of advanced machine learning systems in CNC machining operations as the combination of physical, computers, and networking process created challenges and difficulties regarding the safety and security of the web of data. In order to provide secure and advanced connections between the different CNC machine tool, the security of networks should be enhanced.

Abstract

The Internet of Things (IoT) is becoming pervasive and with each new installation of IoT platforms new and legacy brokers have to be exploited. New internal brokers are those under the control of the platform, while legacy external brokers are those in place managed by third parties. The solution proposed addressed problems of (a) interoperability to reduce set up time to cope with unknown data structures (devices, entities) distributed via brokers; (b) performance by dimensioning both front-end and back-end processes to reach high rates in a broker-based platform, while preserving full capability features of the data warehouse. Interoperability aspects have been addressed by introducing our concepts and a reasoner into an IoT Directory tool to manage Internal and External brokers, automate device discovery and registration from both standard and customized data models. Despite the managed complexity, a broker-based solution turned out to provide high performance. To this end, a specific assessment and architecture tuning have been performed and reported in the paper to give evidence and validation. The proposed integrated IoT Directory has been developed in the context of the Herit-Data Project, and it is currently used in the whole Snap4City network of 18 tenants and billions of data. Snap4City is an open-source IoT platform for Smart Cities and Industry 4.0, which is an official FIWARE platform and solution, EOSC service and libs of Node-RED.

Introduction

The Internet of Things (IoT) defined a paradigm for the computation and communication amongst things, which is becoming every day more and more pervasive and adopted in many different domains [1,2]. It is partially due to the worldwide intense deployment campaign about Low-Power Wide Area Network technologies [3], as well as to many approaches and protocols for communications amongst devices (e.g., Message Queue Telemetry Transport or MQTT, Next Generation Service Interfaces or NGSI, Advanced Message Queuing Protocol or AMQP, Constrained Application Protocol or COAP). The approach is also covering the cloud and fog infrastructures [4,5]. Thus, IoT network infrastructures are becoming every day more complex to be managed due to the networks’ structure and existence of several protocols, formats, and concepts [1,6]. Hence, the complexity is growing in terms of data management, not only for huge amount of data but also for interoperability and abstraction levels needed to data managing. Relevant aspects to be considered are security and privacy [7] on specific data models and entities. To increase the complexity, there is a range of different owners and managers who may control different parts of such IoT networks [8].

In this context, the concept of Gateway is relevant for any segments of IoT networks connecting, as well as for the Web of Things, WoT [9]. Gateways may be integrated with one or more Brokers to send/receive data to/from Devices/entities. The Gateways and Brokers are typically compliant with a single protocol and may be managed by third parties with respect to data management platform. In some cases, they are provided as public services for several interested customers in the same area (for example: Proximus for LoraWAN services [10]). A Gateway may abstract from the IoT Broker level managing multiple brokers for multiple organizations/tenants (which can be regarded as customers of Gateway services to manage several Devices), via some API and/or Web user interface. Typical IoT Brokers can only manage one organization, and thus are single tenant, meaning they broker messages using topic/entity concepts (which can be regarded as the key for subscription on that specific device/entity) without any internal partition of services, but as a sort of family of devices and subscriptions. Some IoT Brokers can be multi-tenant, such as the FIWARE Orion Broker [11,12], which provides support for partitioning the served devices/entities/topics in groups, and each of them may have a dedicated service/path for a specific scope (or a specific customer). Furthermore, devices of different tenants could exist physically in different places (even having identical identifiers, IDs), and the subscription to the broker’s tenant may imply receiving all messages/services in the partition in push. That is feasible only if the subscriber knows the service/path identifier and, in the event of access control, the subscriber has the grant to access the broker’s services. Different IoT Devices connected to the same broker adopt the same protocol, may use different data structures/models and have the same semantic information.

Conclusions

The proliferation of IoT devices, brokers, networks, data models, operators and tenants, makes the harmonization and management of IoT Platform a hard goal. This paper offers an analysis and a comparison amongst relevant existing platforms, and it points out the basic requirements to achieve such aims. These identified requirements are in most cases not addressed by main platforms which prefer to stay on their own end-to-end solutions with limited interoperability and capacity of exploiting legacy IoT networks in place, in terms of performance. The proposed solution addressed problems of (a) interoperability by reducing set up time to efficiently detect and learn how to process unknown data structures (devices, entities) distributed via brokers; (b) performance by dimensioning the front-end and back-end processes to reach high rates in a broker-based platform, while preserving full capabilities features of data warehouse.

As to interoperability, the main identified and solved problems are those related to a large variety of Data Models coming from non-controllable External Brokers. The issue has been solved by designing and implementing a harvester and reasoner that is capable to automatically recognize/understand and map the new data models/types into those already known by Knowledge Base. This approach, together with the definition of a comprehensive meta model and dictionary, has allowed to speed up the process more than 800 times. The harvesting and comprehension process can be periodically performed to keep the platform updated with any newly defined data models by third party brokers. Furthermore, the process is helped by Km4City ontology and Data Dictionary to recognize the new data types and models according to the semantic domain.

Abstract

The lightweight Unmanned Aerial Vehicle (UAV) flight activities are constrained, particularly in the UAV range or activity span and perseverance, by the strategic correspondence link capabilities. This paper tends to the different overlap issue of trading off a set of mission prerequisites, the UAV execution parameters, and strategic credibility; thus compromising between the communication load characterized by a crucial, communication link transmitting power necessities, power accessibility onboard UAV as a weight-restricted parameter, and the UAV security.

Introduction

In current years, an outcome of decreased sizes and weights concentrated charges and amplified functionalities of various sensors and devices. The Unmanned Aerial Vehicles are attracting progressively general trendy varied variety of internal submissions (atmosphere monitoring, surveillance, photography, search-and-rescue, etc.) [1]. Through the marketplace develops and the diversity of applications increases, additional necessities are present-day being compulsory in spreading the flight variety and growing adaptability toward complex assignments of UAV structures [2].

UAV stands for Unmanned Aerial Vehicle, which is commonly considered to be a drone or an aircraft with no pilot on board. UAVs can be remote-controlled aircraft [3]. The unmanned aerial vehicle contains cameras, sensors, communication belonging as well as other payload devices [4]. It was created for military usage, and civilian usage to protect the border. UAVs are widely used in the military [5]. Unmanned aircraft system manufacturing was started by the United States department of defense (DOD) in 2005 [6]. Presently, leading UAV manufacturing countries are the USA, Israel, China, Iran, and Russia [7]. Rustom series of Indian UAVs are under development [8]. The Building of a UAV is easier, it is less expensive and it can be created at a reasonable price. UAVs are made of essential components.

Conclusion

In this paper, we reviewed UAV systems and communication design, and methods used to connect and send data to the ground station. Further, we discussed data processing via sensors and applications with security. We provide key definitions and concepts of UAV systems and details of the development of communication systems and GCS. We provide reviews and analyses of the security of UAVs and the types of attacks that are launched to disturb the communication of the UAV system. Finally, open research areas are given for further development of the UAV system and communication, which leads to better UAV technology for the industry.

Abstract

In machine learning, two approaches outperform traditional algorithms: ensemble learning and deep learning. The former refers to methods that integrate multiple base models in the same framework to obtain a stronger model that outperforms them. The success of an ensemble method depends on several factors, including how the baseline models are trained and how they are combined. In the literature, there are common approaches to building an ensemble model successfully applied in several domains. On the other hand, deep learning-based models have improved the predictive accuracy of machine learning across a wide range of domains. Despite the diversity of deep learning architectures and their ability to deal with complex problems and the ability to extract features automatically, the main challenge in deep learning is that it requires a lot of expertise and experience to tune the optimal hyper-parameters, which makes it a tedious and time-consuming task. Numerous recent research efforts have been made to approach ensemble learning to deep learning to overcome this challenge. Most of these efforts focus on simple ensemble methods that have some limitations. Hence, this review paper provides comprehensive reviews of the various strategies for ensemble learning, especially in the case of deep learning. Also, it explains in detail the various features or factors that influence the success of ensemble methods. In addition, it presents and accurately categorized several research efforts that used ensemble learning in a wide range of domains.

Introduction

In a world full of diverse and varied data sources. Machine learning has become one of the most important and dominant branches of artificial intelligence methods, which is applied in many fields. There are many different learning algorithms and methods. Each method’s pitfalls and drawbacks are measured in terms of several factors, including performance and scalability. Based on a lot of research in machine learning, two methods dominate learning algorithms; namely deep learning (Deng et al., 2014) and ensemble learning (Polikar, 2012, Sagi and Rokach, 2018, Rokach, 2019). The deep learning techniques can scale and handle complex problems and offer an automatic feature extraction from unstructured data(Kamilaris and Prenafeta-Boldú, 2018). Also, deep learning methods contain several types of network architectures for different tasks, such as feeding forward neural networks (Bebis and Georgiopoulos, 1994), convolutional neural networks (Collobert and Weston, 2008), recurrent neural networks (Yu et al., 2019). Many others (Ain et al., 2017). However, the training process of deep learning models requires a massive effort, and tuning the optimal hyper-parameters requires expertise and extensive trial, which is a tedious and time-consuming task. Also, training more complex deep neural network increases the chance of overfitting.

Conclusion

In machine learning, reducing the bias and the variance of models is one of the key factors determining the success of the learning process. In the literature, it has been proven that merging the output of different classification algorithms might decrease the generalization error without increasing the variance of the model. The previous is the key essence of the so-called ensemble learning. Numerous research efforts have preferred ensemble learning over single-model learning in various domains. The main advantage of ensemble learning is combining several individual models to improve prediction performance and obtain a stronger model that outperforms them. In the literature, there are several ensemble techniques to boost classification algorithms. The main difference between any two ensemble methods is training the baseline models and how to combine them. Several research efforts introduced ensemble learning into deep learning models to remedy the problems appearing during the learning process of deep learning models. Usually, the main challenge of deep learning models is that they need a lot of knowledge and experience to tune the optimal hyperparameters aiming at reaching a global minimum error. However, finding the optimal hyperparameters requires an exhausting technique in the search space, which in turn becomes a tedious and time-consuming task. Thus, several research efforts have applied deep ensemble learning in many fields, and most of these efforts are articulated around simple ensemble methods. This paper provided a comprehensive review of the various strategies for ensemble learning, especially in the case of deep learning. The paper also illustrated the recent trends in ensemble learning using quantitative analysis of several research papers. Moreover, the paper offered various factors that influence ensemble methods’ success, including sampling the training data, training the baseline models, and the fusion techniques of the baseline models. Also, the papers discussed the pros and cons of each ensemble method. Additionally, the paper extensively introduced and presented several research efforts that used ensemble learning in a wide range of domains and categorized these efforts into either traditional machine or deep learning models as baseline classifiers. It is worth noting that an ensemble of deep learning models using simple averaging methods is not a smart choice and is very sensitive to biased baseline models. On the other hand, Injecting diversity in ensemble deep learning can become robust to the biased baseline models. The diversity can be achieved by training different baseline deep learning architectures over several data samples. The diversity, however, is limited by the computation cost and the availability of suitable data to be sampled.

Abstract

     Economic development and the comfort-loving nature of human beings in recent years have resulted in increased energy demand. Since energy resources are scarce and should be preserved for future generations, optimizing energy systems is ideal. Still, due to the complexity of integrated energy systems, such a feat is by no means easy. Here is where computer-aided decision-making can be very game-changing in determining the optimum point for supply and demand. The concept of artificial intelligence (AI) and machine learning (ML) was born in the twentieth century to enable computers to simulate humans’ learning and decision-making capabilities. Since then, data mining and artificial intelligence have become increasingly essential areas in many different research fields. Naturally, the energy section is one area where artificial intelligence and machine learning can be very beneficial. This paper uses the VOSviewer software to investigate and review the usage of artificial intelligence and machine learning in the energy field and proposes promising yet neglected or unexplored areas in which these concepts can be used. To achieve this, the 2000 most recent papers in addition to the 2000 most cited ones in different energy-related keywords were studied and their relationship to AI- and ML-related keywords was visualized. The results revealed different research trends in recent years from the basic to more cutting-edge topics and revealed many promising areas that are yet to be explored. Results also showed that from the commercial aspect, patents submitted for artificial intelligence and machine learning in energy-related areas had a sharp increase.

Introduction

     Economic development and increasing welfare are always entangled with the rising consumption of energy resources. Increasing energy generation as the default answer to how to cope with this additive energy consumption may not be the best answer. From an energy justice perspective, it’s not acceptable to deplete energy reservoirs that belong to the next generations [1]. Although retrofitting existing equipment and minimizing energy usage by combining (or cascading) multiple systems for increased efficiency may be an answer to this challenge, increasing efficiency may not be the ideal solution. It’s valid that efficiency leads to lower consumption, but it should be noted that many of these systems might also be able to be turned off or operate on lower loads. So, a more dynamic approach may be a better solution.

     Artificial intelligence and machine learning are relatively new concepts in energy that can be promising tools to operate systems by implementing past and predicted futures to increase the effectiveness of systems.

Conclusion

     The concept of artificial intelligence (AI) and machine learning (ML) is for computers to simulate humans’ learning and decision-making capabilities. With advances in computing systems, AI and ML have become increasingly important areas in many different branches of science and industry. The energy section is also one of the areas that can benefit from AI and ML. To investigate the current standing point of AI and ML in energy-related areas, we used VOSviewer software to investigate and review the relatively new usage of AI and ML in the energy field and propose promising or neglected areas in which these concepts can be used.

     The results showed that from 2000 AI gains an increasing focus, especially after 2014 when the number of articles in ML skyrocketed. And in 6 years become 10 times more compared to 2014. Although there have been many papers in different fields of energy that introduced new usages of AI and ML in that section, due to the vastness of energy usage and respected fields of it, obviously, in no way the available articles could cover all these areas.

Abstract

     The latest 5G technology is being introduced the Internet of Things (IoT) Era. The study aims to focus the 5G technology and the current healthcare challenges as well as to highlight 5G based solutions that can handle the COVID-19 issues in different arenas. This paper provides a comprehensive review of 5G technology with the integration of other digital technologies (like AI and machine learning, IoT objects, big data analytics, cloud computing, robotic technology, and other digital platforms) in emerging healthcare applications. From the literature, it is clear that the promising aspects of 5G (such as super-high speed, high throughput, low latency) have a prospect in healthcare advancement. Now healthcare is being adopted 5G-based technologies to aid improved health services, more effective medical research, enhanced quality of life, better experiences of medical professionals and patients in anywhere–anytime. This paper emphasizes the evolving roles of 5G technology for handling the epidemiological challenges. The study also discusses various technological challenges and prospective for developing 5G powered healthcare solutions. Further works will incorporate more studies on how to expand 5G-based digital society as well as to resolve the issues of safety–security–privacy and availability–accessibility–integrity in future health crises.

Introduction

     The latest wireless mobile phone technology, Fifth Generation (5G), was first widely deployed in 2019. While 5G network is at developing stage, some countries including China, South Korea, the UK and the US commercially deployed 5G networks, and some other developed countries are being expecting to put forward commercial 5G networks by 2025 [1]. Comparing to the existing wireless networks, 5G provides high data rate, lower latency, high volume of devices connectivity with energy efficiency, high reliability and mobility support [2]. Thus, the services offered by 5G mobile networks are generally categorized into three group, namely (i) “enhanced mobile broadband (eMBB)”, (ii) “ultra-reliable low latency communications (URLLC)”, and (iii) “massive machine-type communications (mMTC)” [3]. In 2019, 67% of the world people have subscribed to cellar devices, of which 65% have used smart phones [4], and 204 billion apps have been downloaded [5]. About 3.8 billion public have eagerly used social media reported in January 2020 [6]. As the number of digital devices connectivity with 5G is dramatically increasing, the extent of variation in exposure to radiofrequency arenas is still under investigation.

Conclusion

     The use of 5G network in integrating other digital technologies (like AI and machine learning, IoMTs, big data analytics, cloud computing, etc.) is now a reality in healthcare. From the literature, it is clear that 5G enabled technologies as well as intelligent systems have a prospect to improve healthcare services, such as medical diagnosis, health monitoring, treatment, management, etc. As the 5G poses non-ionizing radiation, there is no significant health risk of 5G RF wave below the recommended safety level. There is no evidence that 5G signal causes the spread of the coronavirus. Moreover, 5G enabled digital technologies have been used for limiting the COVID-19 outbreak, and enhancing the public-health strategies in 2020. Some advanced technology giants are conducting their researches on 5G related applications that will combat against the health risks of any undesirable epidemics. The successful utilization of 5G technology to tackle healthcare challenges will enhance the acceptance of such technologies for other sectors in the future.

     As it is a new field of research, some issues are to be considered in deployment of 5G network in healthcare including infrastructure development, technical standards, effective regulations and policy, privacy protection and security, as well as availability of research data. Further works will incorporate more study on how to expand 5G based digital society as well as to resolve some issues, such as safety–security–privacy, availability–accessibility–integrity, and to augment resilience to future health crises. It is also required to compile laws and prepare proper guidelines to deal with ethical and legal counteractions in managing the online services during the emergency situations. With the new technological innovations, 5G network will provide a complete pathway for a smart society with various promising healthcare applications.

Abstract

     With the rapid increase in the volume of data on the aquatic environment, machine learning has become an important tool for data analysis, classification, and prediction. Unlike traditional models used in water-related research, data-driven models based on machine learning can efficiently solve more complex nonlinear problems. In water environment research, models and conclusions derived from machine learning have been applied to the construction, monitoring, simulation, evaluation, and optimization of various water treatment and management systems. Additionally, machine learning can provide solutions for water pollution control, water quality improvement, and watershed ecosystem security management. In this review, we describe the cases in which machine learning algorithms have been applied to evaluate the water quality in different water environments, such as surface water, groundwater, drinking water, sewage, and seawater. Furthermore, we propose possible future applications of machine learning approaches to water environments.

Introduction

     With rapid economic development, wastewater containing various pollutants is generated, posing serious threats to natural water environments. Thus, various water pollution control measures have been developed. To a large extent, water quality analysis and evaluation have substantially improved the efficiency of water pollution control [1]. To date, many methods have been developed to monitor and assess water quality worldwide, such as the multivariate statistical method [2], fuzzy inference [3], and the water quality index (WQI) [4]. For evaluating water quality, although most water quality parameters can be monitored according to the procedures defined in the relevant standards, the final water quality evaluation results may widely vary owing to the choice of parameters [5]. Considering all water quality parameters is unrealistic because it is not only expensive and technically difficult but also fails to deal with the variability in water quality [6]. However, in recent years, with the advances in machine learning methods, an increasing number of researchers believe that vast amounts of data can be successfully captured and analyzed to meet the complex and large-scale water quality evaluation requirements.

Concluding remarks

     Machine learning has been widely used as a powerful tool to solve problems in the water environment because it can be applied to predict water quality, optimize water resource allocation, manage water resource shortages, etc. Despite this, several challenges remain in fully applying machine learning approaches in this field to evaluate water quality: (1) Machine learning is usually dependent on large amounts of high-quality data. Obtaining sufficient data with high accuracy in water treatment and management systems is often difficult owing to the cost or technology limitations. (2) As the conditions in real water treatment and management systems can be extremely complex, the current algorithms may only be applied to specific systems, which hinders the wide application of machine learning approaches. (3) The implementation of machine learning algorithms in practical applications requires researchers to have certain professional background knowledge.

Abstract

     Distributed model predictive control (DMPC) schemes have become a popular choice for networked control problems. Under this approach, local controllers use a model to predict its subsystem behavior during a certain horizon so as to find the sequence of inputs that optimizes its evolution according to a given criterion. Some convenient features of this method are the explicit handling of constraints and the exchange of information between controllers to coordinate their actuation and minimize undesired mutual interactions. However, we find that schemes have been developed naively, presenting flaws and vulnerabilities that malicious entities can exploit to gain leverage in cyber-attacks. The goal of this work is to raise awareness about this issue by reviewing the vulnerabilities of DMPC methods and surveying defense mechanisms. Finally, several examples are given to indicate how these defense mechanisms can be implemented in DMPC controllers.

Introduction

     The last years have witnessed a growing interest in distributed control methods due to their superior scalability in large-scale applications such as smart grids (Qi et al., 2011, Yazdanian and Mehrizi-Sani, 2014), water systems (Negenborn, van Overloop, Keviczky, & De Schutter, 2009), and traffic control (De Oliveira & Camponogara, 2010). This approach considers the overall system as an aggregation of smaller pieces, i.e., subsystems, which are locally managed by control units referred to as agents (Kordestani et al., 2021, Scattolini, 2009), whose combined decisions determine the overall performance due to the subsystems’ coupling, e.g., in the control objectives and the system constraints. Moreover, this decomposition may be the only choice regarding the control architecture in applications where the implementation of a centralized controller becomes unfeasible due to the problem size or the existence of multiple independent decision-making entities.

Conclusions

     The evolution of technology is leading towards a world of pervasive connectivity where control systems are expected to play a major role. In this context, DMPC algorithms are likely to become essential because they enhance their capabilities with the progress of technology and they offer a means to coordinate control actions in order to attain optimal performance in a scalable manner. However, an interconnected world will also offer significant opportunities for cyber-attacks, which may have devastating consequences if they affect critical infrastructures. Even nowadays we can find every now and then headlines in major newspapers that show how severe cyber-threats can become, with some notorious attacks disrupting applications such as nuclear plants, power grids, smart buildings and autonomous cars, to make a few examples.

     In this article we have reviewed the most vulnerable spots in the control infrastructure that can be exploited to attack DMPC methods. In addition, we have seen that the algorithms have inherent vulnerabilities because their design is usually based on the assumption that every controller in the network will be compliant with the algorithm employed. To deal with these issues, we have presented detection and mitigation mechanisms that can be used to make these schemes resilient. In particular, we have seen that learning methods can make a difference in defensive tasks because of their superior flexibility to adapt to the nominal operation conditions.