**2. Challenges of IoT**

the ability to autonomously obtain and apply knowledge, and the term environment refers to the surroundings. Therefore, a smart environment can be adapted by obtaining knowledge and applying it according to its users' requirements to improve their experience of that environment. In addition, the interconnection among different smart objects can enhance their functional capabilities [1]. In this context, IPv6 plays a vital role because of several features, including scalability in the case of billions of connected devices, better security mechanisms, and the elimination of network address translation (NAT) barriers. The "Internet of Things" (IoT) concept was first

Nowadays, IoT is receiving attention in many fields such as transport, agriculture, industry, and healthcare [2, 3]. Cisco reports that 50 billion devices and objects will be connected to the Internet by 2020. Also, the Internet of Things (IoT) will contribute \$117 billion to the IoT-based healthcare industry and \$1.9 trillion to the global economy according to Gartner and Forbes. In addition, according to Automotive News, the number of cars connected to the Internet worldwide will increase from 23 million in 2013 to 152 million in 2020. According to another report from Navigant Research, the number of installed smart meters around the world will grow to 1.1 billion by 2022. The prediction of such significant growth shows that IoT will become the umbrella of modern societies to realize the vision of smart environments. A lot of research efforts have been developed to integrate IoT with smart environments. To enable the user for monitoring the environment remotely or from remote sites, the integration of IoT with a smart environment is needed to extend the capabilities of smart objects. Based on the application requirements, IoT can be integrated with different smart environments. So, IoT-based smart environments can generally be classified into the following areas: (a) smart homes, (b) smart buildings, (c) smart cities, (d) smart grid, (e) smart health, (f) smart transportation, (g) smart industry, and (h) smart

coined by Kevin Ashton, where smart objects are connected with the Internet.

106 Telecommunication Networks - Trends and Developments

agriculture. **Figure 1** illustrates the IoT-based smart environments.

**Figure 1.** IoT-based smart environments.

There are many open challenges that have been described by various researchers including those related to power supply, enabling a complex sensing environment, evolving architecture, multiple connectivity options, complexity of IoT, security of information exchange within IoT, and privacy [4–6]. Due to the lack of a clear and widely accepted business model that can engage investments to encourage the deployment of these technologies, there is difficulty in the adoption of the IoT paradigm [3].

To a certain extent, the above-mentioned challenges can be met, with the aid of a variety of wireless and wired connectivity options, such as radio frequency identification (RFID), near-field communication (NFC), Bluetooth, and Wi-Fi. These connectivity options are categorized into three broad types considering their geographical area coverage, that is, personal area network (PAN), local area network (LAN), and wide area network (WAN) [7]. **Figure 2** shows this categorization. The existing Wi-Fi networks should be modified to attain a wider coverage and to support mesh networks [8]. In addition, the confirmation on communication pathway of IoT is very important to understand the information exchange within IoT. It uses various standards, techniques, and protocols to disseminate information.

It is essential to support device-to-device (D2D), device-to-server (D2S), and server-to-server communications (S2S) to facilitate information sharing within the IoT [7, 9]. There are multiple standards and protocols involved with IoT communication. Some of these standards and

**Figure 2.** IoT communication technologies [7].

protocols take a higher priority, such as Internet Protocol version 4 (IPv4), Internet Protocol version 6 (IPv6), IPv6 over Low-Power Wireless Personal Area Network (6LoWPAN), User Datagram Protocol (UDP) Constrained Application Protocol (CoAP), and Transmission Control Protocol (TCP). However, UDP is advantageous and cost-effective, due to its smaller size and performance according to constrained device developers [10]. To find a model for arranging these protocols into constrained and unconstrained stacks according to the TCP/ IP network layer architecture, some efforts were made. The unconstrained stack contains Hypertext Transfer Protocol (HTTP), common standards Extensible Markup Language (XML), and IPv4, whereas the constrained stack contains Efficient XML Interchange (EXI), CoAP, and 6LoWPAN which are protocols with similar functionality but the complexity is reduced significantly [3]. In real life, the IoT has been rapidly developed and deployed with the enormous contribution from companies and research centers [11]. However, IEEE 802.11, IEEE 802.3, and IEEE 802.15.4 are the most common standards related to IoT [10], and the Internet Engineering Task Force (IETF) protocol suite has a vital contribution toward IoT for determining the challenges for IoT [12]. So, recently IoTs are widely accepted for using in practical application scenarios. Matrices are available to measure the cost, processing speed, and communication speed. However, there are few studies on application layer protocols and performance of 6LoWPAN [13, 14], IPv6 routing protocol for low power and lossy networks (RPL) [15–17], and IEEE 802.15.4 [18]; a complete evaluation of IoT has not taken place until now. Hence, this gap needs to be filled up in near future, considering a holistic view of IoT.

orchestration techniques, dynamic resource management, and dynamically offloading from clients/hosts to cloud are new challenges while overcoming existing individual challenges of

Benefits and Challenges of Internet of Things for Telecommunication Networks

http://dx.doi.org/10.5772/intechopen.81891

109

In critical applications, reliability is very important [23]. Reliability is not just sending reliable information, but being able to adapt to changing environmental conditions, be resistant to long-term usability and security problems [24]. In all aspects of software and hardware of IoT, reliability requires to be guaranteed. Attempts were made to explain clearly with the architecture considerations, the reliability consideration for transport, link, and application layers together [24]. Moreover, to describe and analyze reliability and cost-related properties

Security and privacy are an essential requirement of most of the applications. In IoT, memory cards of a device have a limited storage capacity. So, only small amounts of data can be stored in them, and some of the data will be stored in other sites remotely. For these remote data, users do not want to disclose their information to others, so these data need high security and privacy. In terms of security, privacy, and governance rules, new technology is required to give users the ability to verify whether the company satisfies their service level agreement or not, dynamically. Therefore, they should adapt pertinent mechanisms for IoT, to meet the expected security level of a user. Privacy, communication, trusted sensing, computation, and digital forging are rarely addressed tasks in terms of security scope [26]. IoT does not adhere to common security standards and architecture; however, security has become a very important issue [27]. Traditional security architectures cannot fully satisfy the security requests of IoT, because of the existence of a huge number of heterogeneous devices that are connected together. As result, there is a large number of malware entry points, which increases vulnerability. By applying a biological immunology approach, a scheme has been proposed based on a dynamic defense security mechanism to alleviate these security issues in an IoT architecture [28]. In [26], attempts were done to secure IoT communications by ensuring the security of IoT devices. As the first step, computer-aided design (CAD) techniques have been proposed to design IoT devices, which are highly optimized in both energy and security. Importantly, compared to the expensive hardware-securing concepts proposed, CAD techniques can be used to implement strong and ample security with low cost. However, it is practically not in use until date.

In literature, there are several approaches for tackling the security issues in the current IoT paradigm. However, the authentication of devices and securing links in a dynamic mobility environment are still unresolved challenges. Thus, the authentication of IoT devices in real-world scenario still has unresolved issues. Researchers have warned of a realistic threat to the IoT community in the future in industries called "smart home hacking" to meet these

The increase in the number of smart devices and the advances of embedded technologies have increased the devices-to-person ratio up to 1.84 in 2010 [29]. In addition, the requirements from applications by a client increase over the time. So, the scalability of IoT, which is the ability to add more devices and services to IoT without degrading the Quality of service (QoS), must be considered. Due to the heterogeneity of devices and underlying technologies, scalability becomes a critical issue in IoT. To enable unified addition of new devices via a layered

of the service composition in IoT, a probabilistic approach was proposed [25].

IoT and cloud [22].

challenges.

In this section, common major challenges of IoT and its future directions will be introduced. These challenges are *performance, availability*, *reliability, security and privacy*, *scalability*, *precision, interoperability*, *compatibility, Big IoT Data*, *mobility,* and *investment*. IoT is used to facilitate information and data anywhere, at any time for any person based on his request [19]. So, to realize IoT, availability is a highly critical issue. The IoT network requires the high availability guarantee of physical devices as well as IoT applications for achieving high availability. The feasible solution to this issue is using redundant maintenance of programs and hardware devices, so that the program or redundant device can be used to perform load balancing when the failure exists [20]. There are situations where simplicity is disclosed to achieve availability, even though redundancy increases complexity. Thus, to achieve availability, the feasible solution is redundant hardware components. In [20], two redundancy models are proposed: passive and active redundancy models. The active redundancy model performed bad compared to the passive redundancy model. Also, in the passive model, spare components are activated only when the primary component fails and these components will be at sleep mode or partially loaded during the other times. The reference provided claims a mathematical model based on Markov chain, which estimates availability and reliability. Since IoT depends on components and performance of involving technologies, its performance cannot be evaluated using a simple mechanism. Moreover, the other factors that influence the performance of IoT are network traffic, huge amounts of data, and heavy reliance on the cloud [21]. Cloud facilitates resource sharing, which is a vital requirement of IoT environments. In addition, users are enabled access to the services irrespective of the location via an Internet connection with the convergence of cloud and IoT. The convergence of cloud and IoT follows IoT-centric cloud approach or cloud-based IoT approach. In either way, orchestration techniques, dynamic resource management, and dynamically offloading from clients/hosts to cloud are new challenges while overcoming existing individual challenges of IoT and cloud [22].

protocols take a higher priority, such as Internet Protocol version 4 (IPv4), Internet Protocol version 6 (IPv6), IPv6 over Low-Power Wireless Personal Area Network (6LoWPAN), User Datagram Protocol (UDP) Constrained Application Protocol (CoAP), and Transmission Control Protocol (TCP). However, UDP is advantageous and cost-effective, due to its smaller size and performance according to constrained device developers [10]. To find a model for arranging these protocols into constrained and unconstrained stacks according to the TCP/ IP network layer architecture, some efforts were made. The unconstrained stack contains Hypertext Transfer Protocol (HTTP), common standards Extensible Markup Language (XML), and IPv4, whereas the constrained stack contains Efficient XML Interchange (EXI), CoAP, and 6LoWPAN which are protocols with similar functionality but the complexity is reduced significantly [3]. In real life, the IoT has been rapidly developed and deployed with the enormous contribution from companies and research centers [11]. However, IEEE 802.11, IEEE 802.3, and IEEE 802.15.4 are the most common standards related to IoT [10], and the Internet Engineering Task Force (IETF) protocol suite has a vital contribution toward IoT for determining the challenges for IoT [12]. So, recently IoTs are widely accepted for using in practical application scenarios. Matrices are available to measure the cost, processing speed, and communication speed. However, there are few studies on application layer protocols and performance of 6LoWPAN [13, 14], IPv6 routing protocol for low power and lossy networks (RPL) [15–17], and IEEE 802.15.4 [18]; a complete evaluation of IoT has not taken place until now. Hence, this gap needs to be filled up in near future, considering a holistic view of IoT.

108 Telecommunication Networks - Trends and Developments

In this section, common major challenges of IoT and its future directions will be introduced. These challenges are *performance, availability*, *reliability, security and privacy*, *scalability*, *precision, interoperability*, *compatibility, Big IoT Data*, *mobility,* and *investment*. IoT is used to facilitate information and data anywhere, at any time for any person based on his request [19]. So, to realize IoT, availability is a highly critical issue. The IoT network requires the high availability guarantee of physical devices as well as IoT applications for achieving high availability. The feasible solution to this issue is using redundant maintenance of programs and hardware devices, so that the program or redundant device can be used to perform load balancing when the failure exists [20]. There are situations where simplicity is disclosed to achieve availability, even though redundancy increases complexity. Thus, to achieve availability, the feasible solution is redundant hardware components. In [20], two redundancy models are proposed: passive and active redundancy models. The active redundancy model performed bad compared to the passive redundancy model. Also, in the passive model, spare components are activated only when the primary component fails and these components will be at sleep mode or partially loaded during the other times. The reference provided claims a mathematical model based on Markov chain, which estimates availability and reliability. Since IoT depends on components and performance of involving technologies, its performance cannot be evaluated using a simple mechanism. Moreover, the other factors that influence the performance of IoT are network traffic, huge amounts of data, and heavy reliance on the cloud [21]. Cloud facilitates resource sharing, which is a vital requirement of IoT environments. In addition, users are enabled access to the services irrespective of the location via an Internet connection with the convergence of cloud and IoT. The convergence of cloud and IoT follows IoT-centric cloud approach or cloud-based IoT approach. In either way, In critical applications, reliability is very important [23]. Reliability is not just sending reliable information, but being able to adapt to changing environmental conditions, be resistant to long-term usability and security problems [24]. In all aspects of software and hardware of IoT, reliability requires to be guaranteed. Attempts were made to explain clearly with the architecture considerations, the reliability consideration for transport, link, and application layers together [24]. Moreover, to describe and analyze reliability and cost-related properties of the service composition in IoT, a probabilistic approach was proposed [25].

Security and privacy are an essential requirement of most of the applications. In IoT, memory cards of a device have a limited storage capacity. So, only small amounts of data can be stored in them, and some of the data will be stored in other sites remotely. For these remote data, users do not want to disclose their information to others, so these data need high security and privacy. In terms of security, privacy, and governance rules, new technology is required to give users the ability to verify whether the company satisfies their service level agreement or not, dynamically. Therefore, they should adapt pertinent mechanisms for IoT, to meet the expected security level of a user. Privacy, communication, trusted sensing, computation, and digital forging are rarely addressed tasks in terms of security scope [26]. IoT does not adhere to common security standards and architecture; however, security has become a very important issue [27]. Traditional security architectures cannot fully satisfy the security requests of IoT, because of the existence of a huge number of heterogeneous devices that are connected together. As result, there is a large number of malware entry points, which increases vulnerability. By applying a biological immunology approach, a scheme has been proposed based on a dynamic defense security mechanism to alleviate these security issues in an IoT architecture [28]. In [26], attempts were done to secure IoT communications by ensuring the security of IoT devices. As the first step, computer-aided design (CAD) techniques have been proposed to design IoT devices, which are highly optimized in both energy and security. Importantly, compared to the expensive hardware-securing concepts proposed, CAD techniques can be used to implement strong and ample security with low cost. However, it is practically not in use until date.

In literature, there are several approaches for tackling the security issues in the current IoT paradigm. However, the authentication of devices and securing links in a dynamic mobility environment are still unresolved challenges. Thus, the authentication of IoT devices in real-world scenario still has unresolved issues. Researchers have warned of a realistic threat to the IoT community in the future in industries called "smart home hacking" to meet these challenges.

The increase in the number of smart devices and the advances of embedded technologies have increased the devices-to-person ratio up to 1.84 in 2010 [29]. In addition, the requirements from applications by a client increase over the time. So, the scalability of IoT, which is the ability to add more devices and services to IoT without degrading the Quality of service (QoS), must be considered. Due to the heterogeneity of devices and underlying technologies, scalability becomes a critical issue in IoT. To enable unified addition of new devices via a layered architecture, a distributed, interoperable architecture was proposed for IoT to address the scalability issues without degrading the QoS for the realization of IoT notion [30]. In [30], the authors propose three layers of IoT infrastructure: (1) virtual object layer (VOL), (2) composite virtual object layer (CVOL), and (3) service layer (SL). The base structure "IoT daemon" of the distributed architecture consists of the functionalities of the three layers which are object virtualization, service composition and execution, and service creation and management. Based on the processing power and memory, every object hosts its own IoT daemon. Various applications are unified by using the three layers of IoT daemon. VOL digitally represents the properties and functionalities of each object. However, to perform a task, multiple objects work in collaboration. Thus, during runtime, composite virtual object (CVO) is created as a mash-up of VOs corresponding to the task. To create a mash-up, potential VOs should be identified, which is done at the CVOL. With the aid of uniform representation of objects (virtual object (VO)), addition of new objects to the IoT network does not degrade QoS because all the devices are connected with distributed architecture. Also, there are scalability issues due to the increase of network elements (NEs) in the Internet. Compensating the scalability issues with a service-oriented path computation element (S-PCE) instead of conventional host-oriented PCE was proposed by Barbosa C. Souza et al. in [31]. The performance evaluation confirmed that the proposed model supports more network elements than host-oriented PCE by comparing results obtained and the logs of DNS servers [31].

when deploying IoT in a smart environment, these parameters need to be considered. For example, longer network latencies can cause delays in applying car brakes and be very dangerous in the case of vehicle-to-vehicle communication in smart transportation environments. Successful IoT deployment in smart environments can be achieved by designing and

Benefits and Challenges of Internet of Things for Telecommunication Networks

http://dx.doi.org/10.5772/intechopen.81891

111

Big data are another challenge in IoT because IoT is one of the largest sources of collecting large amounts of data. As mentioned earlier, by 2020 more than 50 billion devices will be connected with each other, which can lead to big data production. The performance of most IoT applications is based on the data management services. Therefore, due to big data generated by devices forming a smart IoT environment, managing the big IoT data in terms of processing, access, and storage requires highly scalable computing platforms that do not affect the

Compatibility is another challenge in an IoT-based smart environment, where various products are connected with each other. Due to the unavailability of a universal language, most of the products are unable to connect with each other and lead to compatibility issues. To connect devices with each other, collaboration among enterprises, such as LG, Philips, and Samsung, is required. People will be frustrated if these companies are not collaborated and they are only capable of using one brand, in this case. Therefore, the collaboration among these companies is demanded to obtain the infrastructure information of each product and design a universal coding language accordingly by developers. To ensure the success of IoT,

Massive investment in IoT scenario is required for the investment decision to deploy an industrial IoT environment. In IoT, there is a difficulty for industries to adopt this technology where things are not open and interoperable in terms of hardware and software. Therefore, open and integrated hardware and software-based IoT solutions should be built for deployment in industries. In addition, instead of replacing these deployments with new systems, the solutions should be flexible enough for enabling industries to evolve and adapt to their changes. Expertise and investment are required for generating innovation within existing

In this section, the state-of-the-art IoT-based smart systems are presented and categorized and classified according to application domain. The main categories are as follows: (a) smart homes, (b) smart building, (c) smart cities, (d) smart grid, (e) smart health, (f) smart transport,

In [35], for detecting a fault in the software defined network (SDN)-based smart home environment, a cloud-based home solution was proposed. To find the faulty location in an

developing high-precision systems.

performance of the application.

a solution to compatibility issues is demanded.

hardware and software architectures.

**3. IoT-based smart environments**

(g) smart industry, and (h) smart agriculture.

**3.1. Smart homes**

Interoperability is another major concern with regard to IoT, since various types of devices are connected to each other via IoT. Hence, IoT should facilitate services to all these devices regardless of the type, as interoperability is a necessity. By adhering to standardized protocols, this can be achieved to a certain level at the network and application levels. Due to ambiguous interpretations of the same protocol, achieving interoperability is challenging. So, by avoiding such ambiguities, interoperability of IoT would become more realistic. In [32], a solution to address IoT resources using Web protocols via IoT hubs has been proposed. Thus, the interoperability challenges are reduced to data formats and presenting hub catalogues.

In IoT, most of the devices are mobile devices, which make the IoT scenario more complex. So, IoT applications need to deliver services by considering the mobility factor as well. There are available standard management protocols, that is Mobile IPv6 (network layer) and TCP migrate (transport layer), to facilitate mobility issues in IoT. However, these standards are too complex to be used in IoT nodes. For constrained devices in IoT, a CoAP-based mobility protocol (CoMP) was proposed [33]. Moreover, to ensure mobility, a group mobility management (GMM) mechanism is shown to be promising [34]. In this context, the leader machine does mobility management for the group of machines that are grouped according to mobility patterns.

Precision is another one of the most important challenges that need to be addressed in many smart IoT environments such as transportation, healthcare, and unmanned aerial vehicular networks, where devices and systems are connected globally. Compliance with stringent requirements becomes central to the health and safety of the machine operators, machines, and related businesses when dealing with precision machines that can fail if the timing is 1 ms. Available bandwidth and network latency are the key factors that can affect the precision of distributed IoT delay-sensitive mission-critical environments. Therefore, when deploying IoT in a smart environment, these parameters need to be considered. For example, longer network latencies can cause delays in applying car brakes and be very dangerous in the case of vehicle-to-vehicle communication in smart transportation environments. Successful IoT deployment in smart environments can be achieved by designing and developing high-precision systems.

Big data are another challenge in IoT because IoT is one of the largest sources of collecting large amounts of data. As mentioned earlier, by 2020 more than 50 billion devices will be connected with each other, which can lead to big data production. The performance of most IoT applications is based on the data management services. Therefore, due to big data generated by devices forming a smart IoT environment, managing the big IoT data in terms of processing, access, and storage requires highly scalable computing platforms that do not affect the performance of the application.

Compatibility is another challenge in an IoT-based smart environment, where various products are connected with each other. Due to the unavailability of a universal language, most of the products are unable to connect with each other and lead to compatibility issues. To connect devices with each other, collaboration among enterprises, such as LG, Philips, and Samsung, is required. People will be frustrated if these companies are not collaborated and they are only capable of using one brand, in this case. Therefore, the collaboration among these companies is demanded to obtain the infrastructure information of each product and design a universal coding language accordingly by developers. To ensure the success of IoT, a solution to compatibility issues is demanded.

Massive investment in IoT scenario is required for the investment decision to deploy an industrial IoT environment. In IoT, there is a difficulty for industries to adopt this technology where things are not open and interoperable in terms of hardware and software. Therefore, open and integrated hardware and software-based IoT solutions should be built for deployment in industries. In addition, instead of replacing these deployments with new systems, the solutions should be flexible enough for enabling industries to evolve and adapt to their changes. Expertise and investment are required for generating innovation within existing hardware and software architectures.
