8
Cloud-Based Data Analytics for Monitoring Smart Environments

D. Karthika

Department of Computer Science, School of Computing Sciences, Vels Institute of Science, Technology & Advanced Studies (Formerly Vels University), Chennai, Tamil Nadu, India

Abstract

Wireless communication has made tremendous progress. These developments have sparked new wireless connectivity and networking paradigms. For example, the research community is looking at 5G for automated mobile communications in wireless networks. A recent chapter focuses on the concept of Internet of Things (IoT) as a key element of 5G wireless networks, which aims to connect each unit, such as wireless sensor nodes and home appliances, to the internet. As these technologies grow, they are applied to different real-world problems. Distinct from the maximum relevant application areas is the efficient and smarter monitor of populations. The intelligent city is a vision that extracts information from city systems to take management measures. This vision can be realized by using data and communication skills to track and control these processes. The IoT is used because it would integrate all the city’s infrastructure into the internet.

Keywords: Cloud computing, data analytics, IoT, smart cities, wireless technologies

8.1 Introduction

Wireless portfolios can be used as hospitals, highways, trains, and electricity supplies to chart networks like these. To order to pass the readings of the city sensor, local authorities need internet access. The advanced world utilizes knowledge and networking technology for vital data management [1]. Urban grid, intelligent transport, intelligent communication, clever building, intelligent housing, and intelligent services link the whole vision of the urban community. Main structures are interpreted and information collected through improved public management and environmental emissions are evaluated, including alternative knowledge analysis techniques. Sensors over this purpose are built to major urban areas as power grids, water and irrigation systems, and reservoirs for petroleum and gas, trains, roads, schools, hospitals, stations, and airports, etc. Insights are one of the big subjects of today’s infrastructure that contributes itself to the popular trend of social networking and even technological advancement. Apple products also include smartphones, systems, and tablets. Social media, mobile, and laptop developments have as never before changed lifestyles, but insights shift the lives of companies like never. The proliferation of new aspects of the system generated by various types of networks and networks gives businesses useful knowledge and information. The term Business Intelligence (BI), a critical aspect of the results warehouse, was ultimately taken up in its root [2].

All operations—results from integration, pollution, and cleaning, scheduling and data management, tracking, and research unique types, consultancy services, sales and control teams, and system-center users—were still included in BI. A large, diverse, and comprehensive data collection is created for the use of smart technologies and video/audio channels. To track phenomena of interest with various graphical and heterogeneous measurements, sensor devices and networks are used. This also saves, spreads, and analyzes the data for some purposes, including environmental services, the security of air quality, and risk management. For several years, businesses have been accruing data collection, running analyzes of such data in broad data sets, and primarily developing data processing tools [3]. However, the method of disconnecting data production, information management, and technology growth has experienced recent changes, offering companies different market positions. Scalable approaches should integrate the activities of suppliers, producers, service providers, and retailers throughout this scenario. The focus in this section is on data collection and a modern system for the monitoring of urban environmental activities is primarily proposed. The Internet of Things (IoT) is connected to millions of physical sensors and devices to provide heterogeneous, complex, and unstructured data. Many business and scientific projects worldwide focused on IoT data processing to offset costs as well as data maintenance and modeling results.

Powerful database networks infrastructure can effectively handle large programs, and cloud computing can play a crucial role in the IoT paradigm. Cloud storage also provides a broad array of scale-processing and storage features. Therefore, we have built a broad-based data storage cloud architecture that can be implemented and run in different smart environments (e.g., smart cities, home safety, and disaster prevention). Over its associated processes, methodologies, training, and certification, BI as a sector has grown up, is structured, and is now an integral component of this at all large public and private enterprises. Differentiate from documentation and storage BIs and analyses use data mining, analysis, and simulation to provide visibility into the future. It would be inefficient to substitute BI as an umbrella for hypotheses with anything specific to the data. This part grants a detailed step-by-step analysis approach and BI will form the basis. It is thus important that the two meanings be separated and that the two terms are implemented [4].

8.2 Environmental Monitoring for Smart Buildings

Effective protection defines the developments and practices for environmental health identification and monitoring. Advances in technology, engineering, and materials science have opened the way for ever more innovative instruments to be used in environmental monitoring work. Sensor node availability enables fast-to-market speed, which is crucial to the development of new network applications. Embedded sensors remain close to the application-specific hardware from a cost point of view on mobile devices [5]. There are several requirements for the protection of the environment of smart buildings, while there are several limitations and difficulties. The monitoring area and the application scenario both affect requirements and challenges, and the monitoring target, sampling rate, and overall implementation costs are parameters to be considered in the design decisions of the system. At almost the same time, the coexistence of the wireless network and complex changes in the environment are challenges for any proposed solution.

8.2.1 Smart Environments

Each section gives an insight into the main features and facilities in key clever settings like intelligent homes, clever health, and intelligent cities to intelligent factories. The two symmetrical databases are selected for their distinctive scale and complexity, for example, personal-to-business, single user-to-many users, and separate “intelligence” goals. Although there are certainly other types of intelligent environments, we focus as reflective user cases on the four domains. For three main reasons, homes can accommodate intelligent technologies: (i) modern homes already contain many technologies, even though not often linked or interoperable; (ii) controlled environments; and owners that (at least in principle) invest in technology solutions and provide access to them [6].

Schematic illustration of the smart environment information system.

Figure 8.1 Smart environment information system.

Intelligent homes are standard in infrastructure. They usually categorize them as funding and administration systems Figure 8.1. Smart home assistive systems are designed to provide users with practical support for themselves which their everyday lives and responsibilities at home. If, for instance, watching the television or listening to music is a concern for individual users, the intelligent home may assist the preparation and setup of the lighting and interactive computers according to consumers’ needs through preparing noise examples generated by computer automation, such as turning the laundry machine on. Assistance services can do be tailored to the unique needs of the user, particularly when a user is an elderly person, a disabled person, or a simply ill person, such as environmental support services or e-health services. Tech services cover special smart home capabilities.

For instance, the health and protection of the people or energy management in the house, for example, the regulation of ventilation and energy storage solar panels or the control of electrical and lighting equipment, which minimizes energy usage while satisfying the need for the energy supply. While fundamentally specific, intelligent home services are typically built at the application level of context-aware frameworks, which are focused on common functions and mechanisms. Besides, the building blocks of a house these as windows, doors, the electrical system, the air-conditioning system, the energy production network, sensors, appliances, and so on are at the foundation of home control systems. However, more advanced service companies also require data on the customers, which can be accessed by combining environmentally sustainable and compact cameras these as smartphones, usually paired with their users’ devices [7].

Such additional sensors are required to obtain a comprehensive user history that is necessary to transform the intelligent home into an intelligent user-friendly environment. Examples of such details provide the user location (through a localization system), the user’s physical environment (through wearable sensors), or the behavior of the user. Nevertheless, such high-level knowledge cannot be accessed by analyzing raw data from sensors directly. This will then be evaluated by appropriate algorithms based usually on signal processing, machine learning, and/or data analysis. The full spectrum of these networks and applications promotes the implementation of a comprehensive and diverse IoT network from a scheme and new technologies (ICT) perspective at least. All these technologies must cover through to the huge spectrum of device/government data collection, storing, and synthesis and to include adequate support for the configuration, overseeing and governance of the information showcase, and the test results on the mechanical movement at the household [8].

8.3 Smart Health

Over the past 20 years, the uses of smart environments and the Web have rapidly expanded due to significant declines in sensor prices and advances in both signal processing techniques and data integration/quality.

8.3.1 Description of Solutions in General

Under the common scheme described in Figure 8.2 are enabling technologies and their use for health care. Several devices, inserted into and/or worn by the person’s environment, can collect any data on an ongoing or ongoing basis and analyze it so that the first thing you can provide the patient with details or input is to remind medical staff, the family or other designated persons of the condition [9]. These methods can also be used to monitor medical care in specific contexts, e.g., following surgery or to allow individuals to aware of extensive, healthier, and more self-sufficient lives, e.g., for elderly or disabled people.

Schematic illustration of the technologies in healthcare.

Figure 8.2 Technologies in healthcare.

8.3.2 Detection of Distress

Continuous tracking of people through clinical data or using background/ user activities could describe the physical status of the individual and the ability to increase alerts about pain or many dangerous conditions. Another important reason is that, despite its importance in developing countries, the decline in elderly individuals exists separately. This is, for example, one of Germany’s most common sources of emergency and health care in households. Systems need to be developed to analyze, monitor, and consider what kinds of people will be sent for support (medical or ordinary citizen). Drop is one of the main uses of IoT and intelligent communities, including a comprehensive research topic in recent years, for health applications. Many of the most common methods of fall diagnosis is the inertial sensors used by the individual.

With the large variety of devices available and the technology of the last few years, the inertial movement detector has become feasible to detect falling. For instance, the handheld sensors (IMU, GPS, etc.) fuse together to attempt to understand the warning sense of the activity sensor-based algorithm and that false positive. One approach to do then is to attach additional IMU capabilities, such as PIR sensors in the home, to test the person’s behaviors with the IMU over the next few minutes after the incident has been detected. In subsequent years, the usage of recording devices in this context has increased. Deep cameras will offer more information and evaluate moving subjects and stationary artifacts more easily. It can be used for situation analysis, behavior detection, and this irregular drop scenario [8–10].

As for 2D cameras, the city of the graphic arts information is given and the pertinence of the different findings and the actual improvement achieved toward what remains to be accomplished is analyzed. Finally, there are other forms of solutions and, for example, a system based on home integrated radar sensors. Drops tend to be very dangerous and demanding (considering both forms of falls). Another form of the condition that may be observed in cardiac and atrial fibrillation, for example. It can be tracked with a smartphone camera or even now with an intelligent clock. The detection involves one of the world’s most frequent heart attacks and is now a subject of study using telephones and watches. Cases such as regression can also be detected and evaluated in depression. This device can require intervention to avoid a deterioration of the person’s condition as quickly as possible.

8.3.3 Green Protection

As previously highlighted, intelligent homes have been a great challenge and a hot topic for research in recent years [11]. Nevertheless, data from an urban healthcare center could have been used to establish human behavior and to detect changes to the person’s health to identify signs of a protection degradation. In this case, the correct segmentation of data throughout uncertain experiments, big multimodality using very distinct data types, the way we can adapt the structures to the person we are tracking, the issue of inferior actions and high-level data from recognized activities, and the efficiency and capacity of different types of identification are many challenges. The recognition of actions is important for health-related applications in smart homes since it provides the basis for the person’s well-being, the interaction with the effects, and for use of geriatric scales as ADLs.

The issues are rather complex since the first challenge is that the activities performed are indeed very harder to identify and, secondly, the execution not just to comes down to the individual, but rather the relevance for which it is accomplished. It leads to very complex models for designing and analyzing. Attributes are then duplicated from these detections and identification. The first is to assess the status, for example, of the person being watched in the home for a certain type of problem. The assessment shows the evolution of the person’s question to find out whether he or she can no longer live independently. The second type of application is to help individuals carry out these activities with due regard for their disability/illness. It can improve living conditions and relationships at home.

8.3.4 Medical Preventive/Help

All cell sensor with health solutions aims to prevent or help a person who is confronted with a specific condition. Dependence is one of the greatest costs of our healthcare systems and would that the part if it could be modified [12]. A lot of work focuses on taking different types of support into account. Elderly people, for example, provides a system that combines a mobile app that allows the caregiver to provide environmental guidance and a smartphone that supports the person in everyday life. Such technologies require experience so that it can be useful to the customer as rapidly as possible. It typically co-designs for customers. The evaluation or enhancement of the status of individuals with chronic diseases or persistent disorders is indeed one of the major symptoms. In the case of chronic diseases like diabetes, certain electronic devices and applications may be helpful.

The purpose of such systems, which usually rely on measurement equipment and/or mobile devices and many other electrical components, is to help people manage and control the burden of chronic disease or to regulate drug observance. Inside things like the house, tactics can also be applied every day [13]. The objective of this application is to analyze sleep standards so that improvements can be observed and/or quantified and thus abnormalities diagnosed. Sensors and pressure gage are installed into the bed for this purpose. To improve or worsen the living conditions of the person, the long-term development of his needs is necessary. Certain data, for example, can be obtained to long-term track the nature of these data and possibly raise alarms in distress situations, as described previously.

8.4 Digital Network 5G and Broadband Networks

The growing penetration of DERs into distribution networks, such as electric vehicles, photovoltaics, and wind turbines, improves bilateral energy flows and enables customers’ habits to be less predictable [14]. Such observability can be achieved through real-time monitoring of the distribution grid, where networking technologies can turn the infrastructure into an intelligent grid. The system uses sophisticated mid-/low-voltage phasor measurement instruments and secure data sharing via the public transmission network.

8.4.1 IoT-Based Smart Grid Technologies

The IoT, the robust and efficient connectivity system, which includes data recovery, collection, transmission, and storage, is one of the newest digital communication technologies. Modern society defines the quality-of-the-art technology to monitor energy sources, automated vehicles, and home appliances and to manage the consumption of electricity, water, and gas. By comparison to other communication technologies, IoT technology has several benefits. One is that by rising energy consumption and costs, system use can be made more efficient [15]. Also, ICT technology is required to guarantee business continuity for service companies. Recent IoT innovations have led by providers, service suppliers, and entrepreneurs, to IoT technology used by smart grids and other creative areas like smart city development, smart buildings, and smart homes.

Throughout this chapter, the IoT web development and system architecture is thoroughly analyzed [16]. A few innovative solutions are proposed, which include wideband execution besides Lorawan, Internet access, Mobile data-A, and blockchain (NB–IoT), low sustainable energy, and ZigBee and Ethernet (BLE). The lengthy-term communication capacities of unlicensed bands are greatly enhanced by both technologies. The most important LPWAN technologies are LoRa, which it supplies with Semtech’s reinforced, Sig Fox reinforced, ultra-narrow-belt reinforced (UNB), new enhanced weightless communications, the LTE-M machine type (LTE-M), and NB-IoT 3GPPs. The most deployed LPWAN devices are LoRa and UNB since they use unlicensed frequency bands. The LoRaWAN is a separate LoRa strategy that fits into the topology of stars and cells.

8.5 Emergent Smart Cities Communication Networks

In recent years, energy demand has grown rapidly, with electrical equipment volume and scale increasing steadily. Simultaneously, significant changes are happening in the liveliness subdivision, largely due to traditional to wind transformation, sustainable energy strategies, and more efficient renewable micro-generation [17–19].

Effective electricity network operation depends on the coordination of output and use, which is a major challenge to network control. The arrangement of the grid evolves from a hierarchical and central layout in which large manufacturing units are situated on the land of the grid to a more localized one-stop, local processing system. The result is a more robust and less expensive energy supply than intermittent renewable resources. Energy stocks have increased. A broad management framework for realtime measurement, predictions, and surveillance expertise is required to coordinate efficiency and use effectively. One of the most simple and nuanced development settings is intelligent settlements. It covers growing areas such as climate, economy, mobility, electricity, planning, governance, and other sectors.

Schematic illustration of the smart cities communication networks.

Figure 8.3 Smart cities communication networks.

In Figure 8.3, numerous relevant concerns, including a range of administrators, such as city managers, supervisors, service providers, and residents, may be competing priorities. Smart societies are not just a technological challenge but also one where the most complex and heterogeneous development barriers arise and technology tend to address specific needs and expectations. In terms of ICT, the development is divided into 10 areas and issues, covering a variety of intelligent cities spanning from e-tourism, e-culture, e-government, intelligent resources to smart networking, and e-health to well-being. In combination with the political issues, the difficulty of technical and open technologies will hinder the development of intelligent cities. Therefore, it is no surprise, especially for the information sharing of technological systems and knowledge is very important as a technology facilitator for smart cities [20].

The most recent developments in this regard are the implementation of new participatory sensing paradigms, in the (cheap) position of data sensing from cities, including the user himself, mobile applications for personal smartphones. Besides also have the enormous benefit of allowing customers to create intelligent cities, in addition to or entirely remove costs associated with the construction and management of capillary sensory devices in the city. Nonetheless, the relative complexity of these techniques has decided to keep them well apart from standardization, and several independent research schemes in smart cities have been attempted.

8.5.1 RFID Technologies

For even a broad frequency range, communication protocols, and device implementations, RFID is the most widely known concept. Consequently, several international organizations, including ISO, ITU, and IEC as well as regional organizations like DIN (Germany), JIS (Japan), and SINIAV (Brazil), have been adopting RFID technologies. The newly applied labels to eliminate unusual use of alphanumeric models such as MI fare or RAIN RFID are further confusing the complexity of RFID. A common approach is to refer to apps that use the same frequency band amen to provide meaning to navigate across the range of RFID technology [21].

8.5.2 Identifier Schemes

As previously pointed out, the RFID would promote a standard way of interpreting identifiers derived from tags to develop transparent IoT systems that typically involve several stakeholders and promote flexible operations. By contrast, specific labeling schemes for a variety of material objects, positions, and even digital things are already widely used in such a manner that RFID in the IoT field is not feasible from a financial or organizational perspective from a clean slate. The main common usage in the recognition of thousands of already addressed objects is authorized distributor protocols (EPC), Sg-1 and EPCglobal, object identification (OID) under the ISO/ITU Standard, Ubiquitous IDs in broad use in Japan, and many other systems commonly employed in QR code and barcode scanner encoding. Although it is not possible to cover each of these schemes in detail, it finds that all schemes remain methodical giving to a comparable decoration. In specific, a program begins with a prefix that defines the following code, i.e., within an EPC context, the prefix of 00110000 is a 96-bit Serialized Global Trade Object (SGTIN-96) integer. The remainder of the code is then typically hierarchically organized to facilitate the distribution of data sharing between regions and organizations.

8.6 Smart City IoT Platforms Analysis System

Almost all types of sensors now have a WLAN connection, which provides the connectivity of the device, with data transmission and reception capacity, into our everyday life. IoT is a concept that makes it possible, at any time and any place, to interact between items. IoT is widely used in various applications, such as home automation, health care, traffic, and development. Through data from sensors scattered around the cities, it is made available with the advent of IoT technology. Big data technology and machine learning algorithms enable citizens and decision-makers to use these urban data for services and solutions [22].

8.7 Smart Management of Car Parking in Smart Cities

It would be fair to say that people who travel through a community to the other side are only responsible for a small part of urban traffic. Many citizens who want to drive to the other end of a town do not move but instead, go around the area to prevent congestion [23]. Then, it is fair to assume that only a few minutes or even a much longer time will be needed for almost everyone else, including taxis. They could spend 30 seconds to 20 minutes on average, depending on the city, looking for space for parking. We can, therefore, infer that parking contributes significantly to congestion in urban cities. As the population of a city grows, the number of vehicles on roads rises and today our society faces the great challenge of a worldwide gridlock.

8.8 Smart City Systems and Services Securing: A Risk-Based Analytical Approach

The critical infrastructures (CI) are the electricity, service, and transportation components necessary for the survival of vital functions of society. Intelligent infrastructure is also an ecosystem that incorporates smart solutions to manage its resources and improve service quality. Land management includes waste management, water management, and energy. The technology standard includes e-government and public utilities, urban mobility, and networking. This is linked to the objective of encouraging economic growth and improving the quality of life of professional city people. Another of the problems faced by smart cities is the volume of data produced by CI subsystems. Such requirements vary by the type, importance, and flexibility of any cybersecurity strategy for smart cities. Providers and end-users, therefore, need to be willing to use any cybersecurity approach. The usability is defined as the efficiency and efficiency of a consumer’s activities [24]. A working system meets the requirements of the user. In this sense, the implementation of cyber-fusion centers requires a creative strategy to enable users to make an informed decision-making and to provide throughout-depth information on further developments. Records contained in a standard internal file can create logical hypotheses.

8.9 Virtual Integrated Storage System

The collect servers from a variety of heterogeneous monitoring infrastructures (MIs) and disconnect database device functions that manage separate types of information. This covers instances of both DO-SS and OO-SS in the distributed cloud network that are used to provide a flexible and efficient hybrid database system by unique rules. MI access is collected through the storage of GUI processing [25]. It is an interaction between different databases and communication systems. It is the interfaces. The Knowledge Administrator responsible for virtual data processes all collected data improves remote information selects the correct storage device for the different types of network and eventually adds storage to the storage system. Authentication features are implemented by Identity Administrator and Access Manage modules to control user identity and to provide guidance on accessing data and services. The RESTFUL API allows registered users to access the data.

The core feature of the Info Manager is 1) data extraction and 2) data improvement. System: Cloud Server Administrator is responsible for MI data processing. The results abstraction feature of the program manager is essential to address data fragmentation issues. This presents all tracking and sensed data with a succinct logical description. Distributed workers interact and reflect the exercise setting in which all actions (e.g., data processing) are observed by artifacts (e.g., tracking equipment). The OGC Sensor Web Enabling (SWE) project has taken important first measures for web-based discovery, dissemination, and analysis of sensing data. It points out the interoperability specifications for sensor services through uniform system interfaces. The sophistication of the basic sensor network, its contact information, and several hardware components are the protection of SWE resources from devices built at the top [26].

They expand the SWE definition to define the data currently contained in the paper in the cloud. Although SWE is built into a sensing history to clarify findings, they also have a focus on managing objects, automating querying, and recovery. In particular, the OO-SS norm depends on metadata [27–30]. For monitoring purposes, abstract data following SWE requirements are necessary to connect the organization to its context and to provide a seamless query interface for end-users. To this end, the abundance of data allows context-aware metadata, compatible with SWE’s requirements, to extend the information schema for each object. The OGC-SWE criteria for the incorporation of monitoring systems into the cloud and the dissemination of information to deliver the content were closely watched last year.

This post reflects on data storage problems and presents a new approach to data processing and organization. Two different SWE standards refer to Sensor Observation Device (SOS) and Sensor Alert Service (SAS). Specifically, the SOS norm defines experiences for request, filtering, collection of observations, and data in sensor systems, and the SAS standard describes interfaces for sensor publication and subscription observations. The SOS operator and the SAS assistant, respectively, are part of the data manager can be seen in Figure 8.4.

They are designed to meet expectations, the SWE guideline points out. Also, all the functionalities to identify, view, and capture sensors in a well-defined format are supported by the SOS officer. The definitions are then transmitted as per the O&M (Observation and Measurement) requirements of SWE Sensor [31–33]. The sensor provides models and XML schemes for the definition of sensor systems and processes, and O&M supplies XML schemes and models that reflect the estimation of effects and sensing circumstances. A network to meet the needs of cloud users who need advanced environmental information services is the primary task of the SAS Officer. It supplies information according to the subscription publishing model. A publication for the category of observations (characterized by a specific phenomenon found in a well-defined MI) is available to users using a published publication paper SWE-SAS, an XML document, and one or more comments relating to the same publication.

Schematic illustration of the data storage processing

Figure 8.4 Data storage processing.

The constructs cannot be defined in SWE files but can be used to specify the contents of the organization in compliance with SWE specifications only by ordering relevant metadata. This improves the geolocation (e.g., time and place of arrival, user, and expiry time) of the product. The DO-SS is given by the SAS agency for this geolocation information. The object is sent to OO-SS by the information manager portion following the data enrichment process. Therefore, data objects are segregated for the maintenance of storage, database, and recovery processes: the description of the metadata is placed in the DO-SS, while the applicant is retained in the OO-SS. The programs are sent to the DO-SS from the end-users’ point of view. When data are used for monitoring services, requests are found and planned. The user submits his/her application to the system and the DO-SS collects the information concerned. The recovery process will also provide the key to OO-SS entry when the obtained aid is an entity.

8.10 Convolutional Neural Network (CNN)

With respect to other versions of the neural network, such as multi-layer (MLP), CNN is intended to take many arrays as inputs and then use the local field convolution operator to process information by copying the pupils’ image perceptions. It shows excellent productivity in solving computer vision problems, such as picture classification, detection, and interpretation. It is also useful in various ways, including mixed-voice spectral representations, the physical design of VLSI, multi-media encoding compared to conventional DCT transformation and compressive sensing approaches and the identification of cancer from a variety of pictures changing conditions. Moreover, some of the top players recently played a go match in a confusion with AlphaGo, launched by CNN. But CNN’s architecture is becoming broader and more complicated to obtain good statistical performance and to achieve increasingly challenging targets. At the identical time, more pixels are crammed into one image by high-resolution sensors [35–37]. The preparation and production of CNN is very expensive and limited to implementation due to their slow speed.

Although its conception CNN has studied efficiency and competitiveness, it seems to have focused more recently because it has good industrial performance. Any company has introduced profound knowledge, which can be routinely used for CNN. The Tensor Processing Unit (TPU) of Google’s second generation is built for TensorFlow with a peak performance and an on-chip of 28 MB of 92 TFLOPS. It supports both integer and floating-point calculations that make deep learning more effective. Initiated with a free license for data-intensive automotive goods, NVIDIA introduces an open source Project called NVIDIA Deep Learning Accelerator (NVDLA). They share four key features, including weight sharing, local connection, pooling, and multi-layer use between these different structures. There are several widely used layers, including convolution layers, subsampling layers, and completely entangled layers. In general, after the data there is a coevolutionary layer. The layer of the subsample still parallels the coevolutionary layer. In order to maximize the CNN range, this disparity also occurs.

Technologies of CNN Communication

Whereas RFID systems are used primarily for the recognition of an entity, CNNs are used in the environment for physical interaction. Many smaller sensor actuators are computer-like CNN-tools, like RAMs for top-10kB and 8/15-bit processors, and power limitations, many of them running on a limited source of energy, (e.g., coin cell battery) [34]. The principal cable or digital communications technologies in CNN are discussed in each segment. In most instances, PHY and MAC come from many of these technologies, although some are described as part of a larger protocol stack in Table 8.1.

8.10.1 IEEE 802.15.4

The Wireless Personal Area Network (WPAN) is a hardware device family built for the management and regulation of deployment. The first edition was an important milestone in 2003, when open, focused, and economical connectivity with a focus on accessibility and low energy use was initially established. The platform is more common and the foundation for effective network deployment, including IPv6 and non-IP communication approaches, like ZigBee. Nonetheless, the Times Slotted Channel Humping (TSCH) modes were designed to resolve vulnerabilities in industry contexts in specific contexts. Standard protocol stacks typically, though, for industrial applications like ISA 100.11a and Wireless HART TSCH.

8.10.2 BLE

The Bluetooth mainstream, low-energy version was launched in 2010. BLE will partly reuse Bluetooth and BLE can be activated with a system that uses the standard Bluetooth for low additional costs. Therefore, BLE will collect data from or send commands to the captures and actuators around it using its widespread presence on smartphones. The mobile device can also be used as a medium for communication sensors, drives, and the internet. BLE has also become the world’s leading supplier of wearables, tablets, and other consumer electronic devices.

Table 8.1 Technologies in CNN.

Model Layer size Configuration Features Parameter size Applications
LeNet 7 layers 3C-2S-1F-RBF output layer - 60,000 Document recognition
AlexNet 8 layers 5C-3S-3F Local response normalization 60,000,000 Image classification
NIN - 3mlpconvglobal average pooling mlpconv layer: 1C-3MLP; global average pooling - Image classification
VGG 11-19 layers VGG-16: 13C-5S3F Increased depth with stacked 3 x 3 kernels 133,000,000 to 144,000,000 Image classification and localization
ResNet Can be very deep (152 layers) ResNet-152: 151C2S-1F Residual module ResNet-20: 270,000; ResNet-1202: 19,400,000 Image classification; object detection
GoogleNet 22 layers 3C9 Inception-5S1F Inception 6,797,700 Image classification; object detection

8.10.3 ITU-T G.9959 (Z-Wave)

ITU-T G.9959 is a standard that sets the lower layers of Z-Wave. Z-Wave has been designed as a home automation proprietary protocol stack.

8.10.4 NFC

The NFC system is a short-range (e.g., max 0-10 cm) wireless system. It provides inherent safety features as it minimizes unlicensed users’ ability for collecting data. The NFC offers various forms of communication for the payment application, e.g., card emulation, reader mode, or peer contact.

8.10.5 LoRaWAN

LoRaWAN is the new group’s unauthorized cable gear. Wide Area Low Power (LWGN) networks. LoRaWAN uses the physical layer LoRa infrastructure to improve the connectivity over up to 10 km2. The topology-based gateway collecting data for 100,000 devices such as sensors provides low infrastructure costs to benefit from fast transfer speeds and bit rates. The gateway offers low infrastructure costs.

8.10.6 Sigfox

Sigfox also offers a broad range of low infrastructure coverage and, at a high cost, is a leading LPWAN wireless technology. The technology is run by a Sigfox company in unlicensed frequency ranges. It is based on a star topology and 10 sq. connection spectrum as other LPWAN technologies. You can get miles.

8.10.7 NB-IoT

Narrowband IoT (NB-IoT) is often recognized as a new technology in the LPWAN network. It is based on a range of licenses and allows several low-bit devices at a single basic platform. In the Release 13 specification, NB-IoT was defined by 3GPP.

8.10.8 PLC

The Protocol for the Power Line (PLC) defines the technology used by power grid networks as a communication tool. PLC is based on wired networking but interferes with this so that it is resistant to media-like errors sharing. Smart home apps are often the PLC models, such as IEEE 1901.2 or ITU-T G.9903 low-bit rates, and associated applications, such as Smart Grid.

8.10.9 MS/TP

Master-Slave/Token Passing (MS/TP) is a cabled device that belongs to the standard family BACnet building automation. Currently, grid power is given to MS/TP phones. While the functions listed in this article are not as restricted as other overviewed technologies, the MS/TP equipment is limited and a small bit rate is provided based on the RS-485 requirements of the physical layer.

8.11 Challenges and Issues

The following hurdles for IoT solutions in intelligent environments are interoperability and standardization, integration and personalization, and identity recognition and virtualization.

8.11.1 Interoperability and Standardization

The IoT definition is primarily motivated by the development of standards spanning all layers, either de facto or de jure, from physical to application levels. Some are introduced although continue to evolve specifically in smart environments [39]. Interoperability between different machines is important, for example, in the background of smart connections for Industry 4.0 networked performance. To deal with this issue, The UA OPC offers a forum for free, scalable, and transparent information between user and system. OPC UA uses standard transport protocols and encodes to ensure communication, for example, as high-end business-service environment among embedded controllers. It offers users, clients, and servers warning and incident notifications to monitor and check their contact credibility from a security perspective. Usually, certain factory management and automation devices and delivery networks are time-sensitive.

Norms are typically not as mature as the smart sector, and in other fields, such as clever home, smart security, or smart cities, are always plentiful. While the prevalence of specifications attests to the maturity of the technology, on the other side, the extensive variety of requirements and numerous significant overlaps suggest that the market is competitive and is growing rapidly and still finding equilibrium. The product’s ambiguity concerning requirements is positive in the user’s view and reduces the future manufacturer’s lock-in, but the reality that requirements are often not easily interoperable leads to standard lock-in. If protocols in a fast-developing industry are quickly obsolete, lock-ins can be troublesome and can hinder future business reforms [40]. Thus, the interoperability of various IoT standards can become crucial in context (e.g., by identifying appropriate gateways).

8.11.2 Customization and Adaptation

A rising tendency in personalization makes increased demands for accessibility and interoperability, both in leisure and in intelligent market environments [41–43]. Current networking and connectivity issues occur for well-functioned sensor and actuator networks to allow IoT to be directly operated by various stakeholders these as cell devices, fogs, club computing, and technology. One essential issue is how the algorithms will change the data processing and support the user of the solution or anyone who gets the best possible interface from the processing. In the previous sections, we have seen that although this issue is solved with significant effort, there are still several obstacles to enable sensors and actuators to function together in the environment. The availability of reliable and validated knowledge and easy installation, learning, and servicing are two characteristics of the significance of the adoption and long-term usage of modern technologies [44].

Two factors such as mobile house environments, connected houses, and creative manufacture are critical when the device is not to be discarded in the cases above. Devices have been stated to be discarded easily for safety or well-being because the awareness given was originally inspiring people, so that information is not enough to encourage use for a long time. For example, if the actions are not detected by an operation tracking program that of the standard that identifies the operation not meeting your use, things would not be included (e.g., wrong IMU/Heartrate adjust requirements for workout applications). That is why such a device “learns” the way the consumer operates in the first few days. Another sign is the difficulty. For example, a variety of experiments in this area attempt to construct mechanisms for monitoring the activity of an older individual in the home to detect behavior changes.

The best part has been that a new scheme is relatively simple to operate, introduce, and serve. Deployment and maintenance are the key problems of hardware and design. However, embedded algorithms help make it easier to use. There are continuing research efforts at the beginning of this issue. The aim is to use software traces and so-called tacit user feedback to continuously change the behavior of the system. Such suggestions are a comment that the operator creates on the system that the behavior he does not like or that the acts of the user match the behavior of the system. Determining how a person resides in an intelligent environment to detect such variances in the data is still very costly and important for the user experience and reliability of such systems.

8.11.3 Entity Identification and Virtualization

To create an integrated and ever increasingly complex connectivity network, IoT technology unites several elements and functions. Organizational identification is a key factor for meeting that task and ensuring that completely integrated systems are designed and completed and that trust remains in scalable and complex operations. This, also, suggests the necessity for formal IoT identification and the support of the Agency Codes resolution and the associated metadata implementation. Remember that IoT organizations may integrate wide-ranging heterogeneous styles of materials, manufacturing objects and equipment, sites, persons, and other live plants, the atmosphere, and building locations. Although there are several explanations that the usage of contact identities has been made, it is not a universal solution, namely, the generally narrow reach of communication, multihomed relationships, the surgical relationships between individuals. However, agency recognition is a core element in the creation of efficient ways of testing confidence ties and regulating access between IoT companies and apps to confidential information. Finally, the recognition in entities is a crucial feature of organization virtualization in IoT, an integral phase in interoperability as it can be monitored and synchronized with physical and electronic assets.

8.11.4 Big Data Issue in Smart Environments

The cloud storage system we incorporate provides heterogeneous control and connectivity to the servers [45]. This allows users to show their needs for metric scale, time interval, information glocalization, and standard data reception design. To explain our main design approaches, we must first highlight the key issues to be addressed in the data management analysis. Investment management includes various communities worldwide in smart environments. Most models would guide site locators to share their knowledge. For example, tenants have information available for network sensing. The cloud storage provider is interested in integrating this information into the system in this case because at the same time the user also has both a set of resources and a database. The type of agreement between the monitoring system manager and cloud storage provider is beyond its influence, but we want to highlight that in such a complex situation, tracking network data is extremely heterogeneous.

8.12 Future Trends and Research Directions in Big Data Platforms for the Internet of Things

Advancing the internet of products and networking things on earth produces compelling demands in a profoundly complicated and complex setting, subject to real research. Data processing and collaboration between artefacts are constantly expanding for high demand in the network and for data to be stored and transferred. Connect protocols are needed not only to allow high-capacity traffic but also to maintain communication between subjects even when wired or wireless connections are interrupted. New strategies should also be designed to store, scan, and retrieve data used in these settings. IoT creates systems like things, people, climate, networking, and governance. The urban system is packed with qualitative knowledge to rapidly maximize the efficiency of resources through its economic, social, and human activities. Based on the technology it will expand vital community networks such as public safety, transit, government, social security, education, and health services [43].

Determining and implementing resource availability and printing, subscription/notification procedures will also improve the management of complex structures. For rational decision-making of joint initiatives, improved test facilities are needed. Using fractured organic methods would also increase data accuracy. Many of us would have difficulty negotiating with them. Modern solutions must be developed to ensure compatible, autonomous, flexible, reliable, and trustworthy goods. They focus on new general structures, hierarchical, and decentralized. Simultaneously, optimal task distribution is sought across high-powered smart devices and IoT networks. New structures and procedures for solving IoT-level privacy and protection problems, like infrastructure. Improved protection strategies could focus on context-conscious control frameworks. Modern energy-saving, energy-efficient, and self-sustainable approaches are needed. In search of new power-efficient systems and technologies, researchers may explore the potential of analytical structures to derive energy from their environment.

Efficient storage, sorting, positioning, and replicating large data inevitably requires the development of specialized hardware and infrastructure to achieve better real-time data access and improved data infrastructure efficiencies [44]. Big data cloud systems are interesting. This approach is preferable to centralized management, which cannot adapt to complex, complete, and dimensional problems. Self-employment also matters at organizational level. Creating and organizing large environments that promote the IoT is easier if autonomous things respond to events arising from context changes. Product shortages and energy consumption derive from unique conditions. New methods are needed to manage energy efficiently and connect to the network routing level at different levels from design level. One challenge is how to implement strategies and assess the quality of solutions. Finally, to maintain confidentiality, privacy, integrity, and availability, issues need to be resolved.

8.13 Case Study

This chapter provides a case study demonstrating how industry 4.0 IoT, cloud, edge computing, and big data technologies could turn a conventional cooler into an interactive device to drive extra value. This experiment provides technical development based on requirements, design, and test (Table 8.2).

A technology-enabled Smart Products and Services (SPS) network. The aim of using IoT in a smart home is to maximize resource usage, provide stability, enable advanced maintenance, and optimize cost-effective system. The case study focused on mass production in Medellin, Refrigerator Antioquia. This project is part of a portfolio of IoT-enabled projects in a proven factory with over 70 years of op-oration. The minimal impact on current production lines is required to convert ordinary household appliances to smart ones. While this paper focuses on refrigerators as a typical home appliance, the same process can make other appliances like washing machines smart. Because of similarities such as network connectivity, restricted features, data volume, data structure, response time, and the most important possibility of incorporating the custom IoT-enabling board into each home appliance, this industrial case study may be extended to all smart home appliances and other home appliances can execute the suggested implementations. Using a single IoT-middle device and mobile app in home applications. Several user acceptance tests were performed for various devices, but the goal is to provide an optimized smart home solution to all appliance systems. In this segment you can find various aspects of an IoT-end-enabled smart appliance network. Finally, IoT’s simulation and cloud technology is important. The whole section addresses current challenges, PCB design, network connectivity (Wi-Fi and Bluetooth), edge computing, hardware architecture, software engineering, IoT platforms, real-time security, dashboards, applications, and mobile app development.

Table 8.2 Case study of IoT platforms smart products and systems.

IoT platform Functionalities Services Real life products
AWSIoT Connecting things, secure interactions, data process, and evev offline interactions of products, services and systems HTTP Web Sockets MQTT PaaS
SaaS
IaaS
Messages can be routed to AWSendpoints, e.g. Lambda, Kinesis, S3, Machine Learning, Dynamo, DB, CloudWatch, and Elastic search Service with built-in Kibana integration
Azure IoT Suite Easy integration with ERP/SAP, CRM/Sales force and Microsoft Dynamics Remote monitoring, predictive maintenance, connectied factory devices HTTPAM QPMQTT PaaS
SaaS
IaaS
Tetra Pak (keeping food & drink safely), Rock well AUtomation (Smarter industrial machines), ABUS (Safe guard development), Kennametal (Innovation in metal science)
Google Cloud IoT Utilized Google’s backbone and integrated with Google’s web processing, analytics, and machine intelligence MQTTHTTPGCM PaaS
SaaS
IaaS
Com Philips, Spotify, Zulity, Scitis, Airbus, GOJEK (logistics & payment), Oden (IoT manufacturing), Motorola, Ocado (Improved customer care and operations with machine learning)
IBM Watson IoT Machine learning, automated data processing, analyze real-time IoT data, IoT app development supporting Raspberry Pi MQTT PaaS
SaaS
IaaS
(improve health outcome for 33.5M members), KONE (Connects 2 million elevators), ISS (Managing 25,000 buildings worldwide), Teradyne (tracking facility utilization)
Open Source IoT Analytics platform allows aggregating, visualizing and analyzing live data streams in the cloud. HTTPRESt ful MQTTA PaaS WSO2 brings fexibility to mobile projects. It provides manufacturers to develop connected products as well as rich integration and smart analytics capabilities.

8.14 Conclusion

This chapter discussed the challenges of using smart city technology and IoT networks via CR and EH technologies. The smart grid depends on heterogeneous and multi-scale communication networks to incorporate new applications and services. To understand it, smart grid operators need to focus on how best to exploit the benefits of these new digital innovations for their daily networking needs and share their experience in determining the imminent. As an advanced risk-based cybersecurity strategy by computer science, this helped reduce cybersecurity risk exposure through a big data fusion center solution. Popular smart city applications with cloud support, connectivity protocols, security mechanisms, storage systems, and prediction capabilities. Although prediction is important in our daily lives, about half of the frameworks have predictive capabilities. This enables smart city dream to be accomplished as the systems can be understood, evaluated, and implemented using communication technologies. It makes smart and efficient community management in terms of public facilities, services, telecommunications, retail, energy, water, etc.

References

1. Abhishek, R. et al., SPArTaCuS: Service priority adaptiveness for emergency traffic in smart cities using software-defined networking. Proceedings of IEEE International Smart Cities Conference (ISC2), Trento, pp. 1–4, 2016.

2. Alparslan, M. et al., Impacts of microgrids with renewables on secondary distribution networks. Appl. Energy, 201, 308–318, 2017.

3. Aydeger, A., Akkaya, K., Uluagac, A.S., SDN-based Resilience for Smart Grid Communications. Proceedings of IEEE Conference on Network Function Virtualization and SDN, Demo Track, 2015.

4. Aydeger, A., Akkaya, K., Cintuglu, M.H., Uluagac, A.S., Mohammed, O., Software Defined Networking for Resilient Communications in Smart Grid Active Distribution Networks. Proceedings of IEEE ICC SAC Communications for the Smart Grid, 2016.

5. Cai, Z., Liu, F., Xiao, N., Liu, Q., Wang, Z., Virtual network embedding for evolving networks. Proceedings of IEEE GLOBECOM, pp. 1–5, 2010.

6. Cankaya, H.C., SDN as a Next-Generation Software-Centric Approach to Communications Networks. OSP, 33, 2, 1–12, Feb. 2015.

7. Cerroni, W. and Callegati, F., Live migration of virtual network functions in cloud-based edge networks. Proceedings of the IEEE International Conference on Communications (ICC 2014), pp. 2963–2968, 2014.

8. Chiosi, M., Clarke, D., Willis, P., Feger, J., Bugenhagen, M., Khan, W., Fargano, M., Chen, C., Huang, J., Benitez, J., Michel, U., Damker, H., Ogaki, K., Fukui, M., Shimano, K., Delisle, D., Loudier, Q., Kolias, C., Guardini, I., Demaria, E., López, D., Salguero, Ramón, F.J., Ruhl, F., Sen, P., ‘Network functions virtualisation’, introductory white paper. Proceedings of SDN and OpenFlow World Congress, 2012.

9. Chowdhury, N., Rahman, M., Boutaba, R., Virtual network embedding with coordinated node and link mapping. Proceedings of IEEE INFOCOM, pp. 783–791, 2009.

10. Dorsch, N., Kurtz, F., Georg, H., Hagerling, C., Wietfeld, C., Software-defined networking for Smart Grid communications: Applications, challenges and advantages. Proceedings of the 2014 IEEE International Conference on Smart Grid Communications (Smart GridComm), Venice, Italy, 3–6 November 2014, pp. 422–427, 2014.

11. Aguilar, S., Vidal, R., Gomez, C., Opportunistic sensor data collection with bluetooth low energy. Sensors, 17, 1, 159, 2017.

12. Akl, A., Chikhaoui, B., Mattek, N., Kaye, J., Austin, D., Mihailidis, A., Clustering home activity distributions for automatic detection of mild cognitive impairment in older adults 1. J. Ambient Intell. Smart Environ., 8, 4, 437–451, 2016.

13. Al-Fuqaha, A., Guizani, M., Mohammadi, M., Aledhari, M., Ayyash, M., Internet of Things: A survey on enabling technologies, protocols, and applications. IEEE Commun. Surv. Tut., 17, 4, 2347–2376, 2015.

14. Amato, G., Bacciu, D., Broxvall, M., Chessa, S., Coleman, S., Di Rocco, M., Dragone, M., Gallicchio, C., Gennaro, C., Lozano, H., McGinnity, H., Micheli, A., Ray, A.K., Renteria, A., Saffiotti, A., Swords, D., Vairo, C., Vance, P., Robotic ubiquitous cognitive ecology for smart homes. J. Intell. Robot. Syst., 80, 1, 57–81, 2015.

15. Aziz, A.A., Klein, M.C., Treur, J., An integrative ambient agent model for unipolar depression relapse prevention. J. Ambient Intell. Smart Environ., 2, 1, 5–20, 2010.

16. Bacciu, D., Chessa, S., Gallicchio, C., Micheli, A., On the need of machine learning as a service for the Internet of Things, in: ACM International Conference Proceedings Series, ACM, 2017.

17. Baldewijns, G., Claes, V., Debard, G., Mertens, M., Devriendt, E., Milisen, K., Tournoy, J., Croonenborghs, T., Vanrumste, B., Automated in-home gait transfer time analysis using video cameras. J. Ambient Intell. Smart Environ., 8, 3, 273–286, 2016,.

18. Baños-Gonzalez, V., Afaqui, M.S., Lopez-Aguilera, E., Garcia-Villegas, E., IEEE 802.11 ah: A technology to face the IoT challenge. Sensors, 16, 11, 1960, 2016.

19. Baronti, P., Pillai, P., Chook, V.W., Chessa, S., Gotta, A., Hu, Y.F., Wireless sensor networks: A survey on the state of the art and the 802.15. 4 and ZigBee standards. Comput. Commun., 30, 7, 1655–1695, 2007.

20. Bellavista, P., Chessa, S., Foschini, L., Gioia, L., Girolami, M., Human-enabled edge computing: Exploiting the crowd as a dynamic extension of mobile edge computing. IEEE Commun. Mag., 56, 1, 145–155, 2018.

21. Bernardino, S., Freitas Santos, J., Cadima Ribeiro, J., The legacy of European capitals of culture to the “smartness” of cities: The case of Guimarães 2012. J. Conv. Event Tour., 19, 138–166, 2018, Taylor & Francis.

22. Bleser, G., Steffen, D., Weber, M., Hendeby, G., Stricker, D., Fradet, L., Marin, F., Ville, N., Carré, F., A personalized exercise trainer for the elderly. J. Ambient Intell. Smart Environ., 5, 6, 547–562, 2013.

23. Bormann, C., Castellani, A.P., Shelby, Z., CoAP: An application protocol for billions of tiny Internet nodes. IEEE Internet Comput., 16, 2, 62–67, 2012.

24. Callaghan, V. and Hagras, H., Preface, Thematic issue: Smart homes. J. Ambient Intell. Smart Environ., 2, 1, 207–209, 2010.

25. Cardone, G., Cirri, A., Corradi, A., Foschini, L., The participact mobile crowd sensing living lab: The testbed for smart cities. IEEE Commun. Mag., 52, 10, 78–85, 2014.

26. Castro-Jul, F., Díaz-Redondo, R.P., Fernández-Vilas, A., collaboratively assessing urban alerts in ad hoc participatory sensing. Comput. Networks, 131, 129–143, 2018.

27. Cesta, A., Cortellessa, G., Fracasso, F., Orlandini, A., Turno, M., User needs and preferences on AAL systems that support older adults and their carers. J. Ambient Intell. Smart Environ., 10, 1, 49–70, 2018.

28. Chahuara, P., Fleury, A., Portet, F., Vacher, M., On-line human activity recognition from audio and home automation sensors: Comparison of sequential and non-sequential models in realistic smart homes 1. J. Ambient Intell. Smart Environ., 8, 4, 399–422, 2016.

29. Chin, J., Callaghan, V., Ben Allouch, S., The Internet of Things: Reflections on the past, present and future from a user centered and smart environments perspective. J. Ambient Intell. Smart Environ., 11, 1, 45–69, 2019.

30. Potluri, S., Quality of Service based Task Scheduling Algorithms in Cloud Computing. Int. J. Electr. Comput. Eng., 7, 2, 1088–1095, April 2017.

31. Potluri, S., Efficient Hybrid QoS Driven Task Scheduling Algorithm in Cloud Computing Using a Toolkit: Clouds. JARDCS, 12-Special Issue, 8, 1270–1283, 2017.

32. Potluri, S., Optimization model for QoS based task scheduling in cloud computing environment. IJEECS, 18, 2, 1081–1088, 2020.

33. Le, D.N., Kumar, R., Nguyen, G.N., Chatterjee, J.M., Cloud computing and virtualization, John Wiley & Sons, Hoboken, USA, 2018.

34. Moy Chatterjee, J., Fog computing: beginning of a new era in cloud computing. Int. Res. J. Eng. Technol. (IRJET), 4, 05, 735, 2017.

35. Akhare, R., Mangla, M., Deokar, S., Wadhwa, V., Proposed Framework for Fog Computing to Improve Quality-of-Service in IoT Applications, in: Fog Data Analytics for IoT Applications, pp. 123–143, Springer, Singapore, 2020.

36. Chatterjee, J.M., Priyadarshini, I., Le, D.N., Fog Computing and Its security issues, in: Security Designs for the Cloud, Iot, and Social Networking, pp. 59–76, 2019.

37. Kumar, A., Payal, M., Dixit, P., Chatterjee, J.M., Framework for Realization of Green Smart Cities Through the Internet of Things (IoT), in: Trends in Cloud-based IoT, pp. 85–111, Springer, Cham, 2020.

38. Sujatha, R., Nathiya, S., Chatterjee, J.M., Clinical Data Analysis Using IoT Data Analytics Platforms, in: Internet of Things Use Cases for the Healthcare Industry, pp. 271–293, Springer, Cham, 2020.

39. Priya, G., Shri, M.L., GangaDevi, E., Chatterjee, J.M., IoT Use Cases and Applications, in: Internet of Things Use Cases for the Healthcare Industry, pp. 205–220, Springer, Cham, 2020.

40. Almusaylim, Z.A. and Zaman, N., A review on smart home present state and challenges: linked to context-awareness internet of things (IoT). Wirel. Netw., 25, 6, 3193–3204, 2019.

41. Almulhim, M. and Zaman, N., Proposing secure and lightweight authentication scheme for IoT based E-health applications, in: 2018 20th International Conference on Advanced Communication Technology (ICACT), 2018, February, IEEE, pp. 481–487.

42. Almulhim, M., Islam, N., Zaman, N., A Lightweight and Secure Authentication Scheme for IoT Based E-Health Applications. Int. J. Comput. Sci. Netw. Secur., 19, 1, 107–120, 2019.

43. Alshammari, M.O., Almulhem, A.A., Zaman, N., Internet of Things (IoT): Charity Automation. Int. J. Adv. Comput. Sci. Appl. (IJACSA), 8, 2, 166–170, 2017.

44. Potluri, S., An IoT based solution for health monitoring using a body-worn sensor enabled device. JARDCS, 10, 9, 646–651, 2018.

  1. Email: [email protected]
..................Content has been hidden....................

You can't read the all page of ebook, please click here login for view all page.
Reset
44.222.90.25