Discussion Board Topic 1

Following the reading of the articles provided the discussion board, answer the following questions:

1a. Describe in your own words the idea of Internet of things (IoT). What are the three examples of IoT applications used in your workplace, or environment that you are familiar with?

WE WRITE ESSAYS FOR STUDENTS

Tell us about your assignment and we will find the best writer for your paper.

Write My Essay For Me

1b. Explain in your own words the key issues and challenges.

1c. Describe the potential utilization of IoT in a market sector of your choice (i.e. healthcare, transportation, utilities, etc.)

1d. Briefly describe your opinion about the top challenges of IoT security, provide examples to support your answer.

* Provide academic or industry source to support your answers.

Must be in APA format and all sources within the last three years.

Computer Networks 144 (2018) 17–39
Contents lists available at ScienceDirect
Computer Networks
journal homepage: www.elsevier.com/locate/comnet
Review article
Internet of Things (IoT): A review of enabling technologies, challenges,
and open research issues
Alem Colakovi ˇ c´ a,∗
, Mesud Hadžialic´ b
a University of Sarajevo, Faculty of Traffic and Communications, Zmaja od Bosne 8, Sarajevo, Bosnia and Herzegovina b University of Sarajevo, Faculty of Electrical Engineering, Zmaja od Bosne 8, Sarajevo, Bosnia and Herzegovina
a r t i c l e i n f o
Article history:
Received 15 December 2017
Revised 18 May 2018
Accepted 9 July 2018
Available online 20 July 2018
Keywords:
IoT (Internet of Things)
IoT vision
IoT features
IoT enabling technologies
Open issues and challenges
Future research direction
a b s t r a c t
IoT (Internet of Things) is a new paradigm which provides a set of new services for the next wave of
technological innovations. IoT applications are nearly limitless while enabling seamless integration of the
cyber-world with the physical world. However, despite the enormous efforts of standardization bodies,
alliances, industries, researchers and others, there are still numerous problems to deal with in order to
reach the full potential of IoT. These issues should be considered from various aspects such as enabling
technologies, applications, business models, social and environmental impacts. In focus of this paper are
open issues and challenges considered from the technological perspective. Just for clarification, we put
in light different visions that stand behind this paradigm in order to facilitate a better understanding
of the IoT’s features. Furthermore, this exhaustive survey provides insights into the state-of-the-art of IoT
enabling and emerging technologies. The most relevant among them are addressed with some details. The
main scope is to deliver a comprehensive overview of open issues and challenges to be tackled by future
research. We provide some insights into specific emerging ideas in order to facilitate future research.
Also, this paper brings order in the existing literature by classifying contributions according to different
research topics.
© 2018 Elsevier B.V. All rights reserved.
1. Introduction
IoT (Internet of Things) is based on integrations of various processes such as identifying, sensing, networking, and computation.
It enables large-scale of technological innovations and value-added
services which personalize users’ interaction with various “things”.
There are numerous IoT applications that can be grouped into various domains such as health, traffic, logistics, retail, agriculture,
smart cities, smart metering, remote monitoring, process automation, etc. Despite the enormous progress in different research fields
including architectures, standardization, emerging technologies, security, etc. we consider that IoT is still in the nascent stage of its
development. This article provides reports on the state of art, current trends and open issues according to the main IoT visions and
enabling technologies. We conducted a review of up-to-date reference literature including journal papers, conference’s papers, standards, project reports, white papers and reports from industries.
Thus, we provide useful guidelines for readers to understand IoT
paradigm and open issues in order to provide perspectives for future research and development.
∗ Corresponding author.
E-mail address: alem.colakovic@gmail.com (A. Colakovi ˇ c). ´
To discuss research trends, we need to understand what IoT really means and to examine its impact on everyday life, industry,
and new business models. We are still in nascent stages where everybody is trying to interpret IoT according to their visions and
needs [1]. Because of that, there is no universal definition of IoT
and existing visions are fuzzy. Also, there are some other terms
related to IoT such as IoE (Internet of Everything), WoT (Web of
Things), CoT (Cloud of Things), M2M (Machine to Machine), etc.
Some authors consider these terms have the same meaning but
the majority of authors distinguish these meanings according to
particular vision [2]. We do not include deep analysis of different
visions but we present some definitions so that the reader can gain
an overall view of IoT.
IoT is based on integration of various standards and enabling
technologies with different sensing, connectivity, storage, computational, and other capabilities. However, the fragmentation of standards and diversities in deployed technologies produce significant
challenges in providing full connectivity of everything [3–5]. This
causes complex integration challenges [6] as one of the major challenges of IoT development. Numerous standardization organizations, alliances, academics, and industries make an effort on IoT
developments, innovation, and standardization but there is still
lack of a comprehensive framework with integrated standards unhttps://doi.org/10.1016/j.comnet.2018.07.017
1389-1286/© 2018 Elsevier B.V. All rights reserved.
18 A. Colakovi ˇ ´c, M. Hadžiali´c / Computer Networks 144 (2018) 17–39
der one IoT vision [7,8]. This causes many challenges that have
been identified and discussed in this paper.
Some of the most important challenges that IoT faces are related to traffic loads and various traffic models [9]. Every day more
and more devices (things) are being connected to Internet and devices are becoming the major producers and consumers of traffic
[10]. This is the reason why traffic requirements arise and we need
new traffic models, protocols, network capabilities, security mechanisms, etc. There is a need of simplification and adoption of the
current IP (Internet Protocol) architecture in order to enable seamless connectivity and effective management in HetNet (Heterogeneous Networks) environment [6,11,12]. Some other challenges related to development of IoT include devices identification, addressing, interoperability, mobility, massive scaling, management, energy efficiency, security, privacy, etc. Also, future deployments of
IoT need to fulfill a sustainable smart world with the focus on
green IoT enabling technologies which is another major issue [13].
In order to identify and discuss technology-based issues, this
paper presents a classification of IoT enabling technologies according to their functionalities. This approach has its value to the research community because it can be used as a starting point for
future research. The main objectives are summarized as follows:
– To clarify IoT vision and definitions as well as to provide a comprehensive overview of IoT features.
– To provide an overview of common IoT enabling and emerging
technologies according to their functionalities.
– To provide a discussion about IoT open issues and challenges to
be tackled by future research.
The remaining of this paper is organized as follows.
Section 2 introduces to IoT visions and features according to
various perspectives. Section 3 provides a comprehensive overview
of IoT enabling technologies which are grouped into four main
domains according to their functionalities. Open issues and challenges related to IoT are addressed and discussed in Section 4.
Furthermore, we present a literature pool and give some future
research hints. Section 5 focus on specific emerging ideas and
issues such as modeling and mathematical formulation of IoT
systems. The final Section 6 contains concluding remarks.
2. Internet of Things (IoT): visions and features
There are various definitions, outlooks, and visions that stand
behind IoT. The early stages of Internet were characterized by
WWW (World Wide Web) with linked static HTML (Hyper Text
Markup Language) documents. This concept evolved to Web 2.0
which enabled user interaction through social networks, forums,
blogs, e-learning platforms, CMS (Content Management Systems),
etc. The next step in evolving of Internet is referred as Web 3.0
or Semantic Web. The main goal of Web 3.0 is to make web content and services understandable by devices without human involvement. IoT takes Web 3.0 to a new level by enabling seamless
connectivity anytime and anywhere by anyone and anything. It enables to create novel value-added services by dynamically assembling different types of capabilities (sensing, communication, data
processing, actuation, etc.) [14].
IoT (Internet of Things), IoE (Internet of Everything), M2M (Machine to Machine), CoT (Cloud of Things), WoT (Web of Things), are
related terms that have been used by various authors, standardization bodies (ITU, ETSI, IETF, OneM2M, OASIS, W3C, NIST, etc.),
alliances (IERC, IoT-i, IoT-SRA, MCMC, UK FISG, etc.), projects (IoTA, iCore, CASAGRAS, ETP EPoSS, CERP, etc.), industries (e.g. CISCO,
IBM, Gartner, IDC, Bosch, etc.) with the same or different meaning. For example, there is a vision that IoE has more comprehensive meaning than IoT such as CISCO’s vision of IoE which is built
upon the “four pillars” (building blocks) including people, things,
data and process while IoT is only composed of “things”. WoT is
mostly considered with the similar meaning as IoT while M2M
refers to direct communication (without human intervention) between objects (devices). Most of the approaches are based on a
technological aspect of IoT paradigm because the same technologies are deployed in most of the cases independently of different
visions. However, we have to highlight and other aspects such as
social, business, and environment, that effect on IoT development
in the future. This multidisciplinary approach is required to overcome any biases of different perspectives of IoT.
Differences in the IoT visions are the result of various approaches to this issue. The most of authors define IoT according to particular aspect and specific interests [1,2,15–17]. We need
to address this challenge to understand IoT concept which will
help to facilitate further research. In recent literature, IoT is considered from two [17] or three [16] main perspectives (visions)
such as “Internet” oriented, “Thing” oriented and “Semantic” oriented perspective. We consider that the basic concept of Internet
is not changing much and it is all about things which are changing and becoming smart. Deep research of IoT requires a comprehensive approach and the best way is to put together various perspectives and visions because IoT is not an individual system but
integrates several subsystems and technologies. For example, IoT
includes sensors, network infrastructure, data analytic tools, etc.
upon which various applications and services can be run [18]. All
this can be considered according to ITU vision of IoT that enables
connectivity with 3A concept: anytime, anywhere, by anyone and
anything [19].
According to Internet-oriented vision, IoT is considered as a
global infrastructure that enables connectivity between both virtual and physical objects. ITU-T Y.2060 Recommendation defines
IoT as a global infrastructure for the information society enabling
advanced services by interconnecting (physical and virtual) things
based on, existing and evolving, interoperable information and
communication technologies [19]. Also, ITU described some technologies deployed in the development of the IoT such as RFID, sensors, smart technologies, nano technology, etc. The similar definition has been provided by ISO/IEC JTC where IoT is defined as an
infrastructure of interconnected objects, people, systems and information resources together with intelligent services to allow them
to process information of the physical and the virtual world and
react [20].
There are some definitions that include only physical aspect
(things-oriented vision) of IoT. “Things” can be “real world entities”
or “virtual entities”. For example, Al-Fuqaha et al. [5] consider IoT
as a technology that enables physical objects to see, hear, think,
share information, coordinate decisions and perform jobs. In special IEEE report IoT is defined as “a network of items – each embedded with sensors – which are connected to the Internet.” The
similar approach is used by OASIS which describes IoT as “a system
where the Internet is connected to the physical world via ubiquitous sensors”. The most of definitions related to this perspective
refers to term M2M. ETSI defines concept named M2M communications as the communication between two or more entities that
do not necessarily need any direct human intervention. Also, ETSI
provides an architectural model for M2M.
Some approaches include a combination of two or all three perspectives. IETF describes the term IoT along with definition for “Internet” and “things”. IAB (The Internet Architecture Board) defines
IoT in RFC 7452 [21] as a trend where many embedded devices
employ communication services offered by the Internet protocols.
If we use a definition of “Internet” as world-wide network based
on TCP/IP, and if we consider “things” as a semantically various objects, then IoT means a world-wide network of interconnected objects uniquely addressable, based on standard communication protocols [22]. From this perspective, WoT is the term used by W3C to
A. Colakovi ˇ
´c, M. Hadžiali
´c
/ Computer Networks 144 (2018) 17–39 19
define IoT from the application and Web technologies perspective:
“The Web of Things is essentially about the role of Web technologies to facilitate the development of applications and services for
the Internet of Things. This includes sensors, actuators, physical objects tagged with a bar code or NFC as well as their virtual representation. An example of relevant Web technology is HTTP which
is used for accessing RESTful services and for naming objects as a basis for linked data and rich descriptions. Also, JavaScript APIs
(Application Programming Interfaces) is web technology used for
virtual objects to act as proxies for real-world objects”. This perspective of IoT is used by Chen et al. [23] who define IoT as an
intelligent network which connects all things to the Internet for the
purpose of exchanging information and communicating with the
information sensing devices in accordance with agreed protocols.
Whitmore et al. [24] define the IoT as
a paradigm where everyday
objects can be equipped with identifying, sensing, networking and
processing capabilities that will allow them to communicate with
one another and with other devices and services over the Internet to accomplish some objective. In Oxford Dictionary IoT is defined as the interconnection via the Internet of computing devices
embedded in everyday objects, enabling them to send and receive
data [25]
.
Instead of including
a specific definition, some definition approaches include descriptions and list of requirements. For example, OneM2M present an exhaustive list of M2M/IoT system requirements as well as functional roles and standards related to
architecture, interfaces, security, communication, etc. The National
Institute of Standards and Technology (NIST) give
a description
of IoT rather than
a formal definition. There are many other approaches to define IoT but some basic characteristics of these definitions are common. IoT require a new “technology stack” that includes various hardware and software embedded in the objects.
Also, understanding of this paradigm requires considering many
other aspects such as applications, social impact, business models, environmental issues, etc. We can signify that a comprehensive
perspective is needed to understand IoT paradigm. For providing
our definition we introduce
a service-oriented vision which put IoT
services in the focus while it includes all perspectives mentioned
above
(Fig.
1). Based on these considerations we highlight common characteristics of IoT (Fig. 2) and we can define Internet of
Things (IoT) as the inter-networking paradigm enabled by technology stack which provides a seamless connectivity between physical
and virtual objects to facilitate the development of intelligent services and applications with self-configuring capabilities. The technology stack is a combination of various technologies that enable
these processes as well as to provide
a seamless connectivity anytime and anywhere by anyone and anything.
3. IoT enabling technologies
IoT systems are comprised of functional blocks to facilitate various utilities to the system such as sensing identification, actuation, communication, and management [26]. Therefore, IoT enabling technologies can be summarized into several categories
such as: sensing technologies, identification and recognition technologies, hardware, software and cloud platforms, communication
technologies and networks, software and algorithms, positioning
technologies, data processing solutions, power and energy storage,
security mechanisms, etc. For the purpose of this paper, we present
IoT enabling technologies according to functional blocks composed
of four main domains as shown on Table
1
. This classification is
used to provide an overview of open issues and challenges clearly
and concisely. All these domains (system layers) include various
hardware, software, and technologies with specific functionalities
and capabilities. Incorporating these domains into the IP architecture enables full deployment of the IoT technologies. IoT platforms
Table 1
IoT enabling technologies and functional blocks.
20 A. Colakovi ˇ ´c, M. Hadžiali´c / Computer Networks 144 (2018) 17–39
Fig. 1. Towards IoT visions.
Fig. 2. Towards IoT features.
enable easy integration of various IoT enabling technologies. These
platforms can be defined as an intelligent layer that connects the
things to the network and abstract applications from the things
with the goal to enable the development of services [27]. They provide a framework for connecting things to the various networks
and applications. There are various IoT platforms such as hardware
and cloud platforms which can be used to increase productivity,
usability, and flexibility of IoT applications as well as to decrease
time and cost of the developing process.
IoT enabling technologies can be considered from various aspects (utility factors) which we used to provide insights into the
open issues and challenges.
3.1. Application domain
IoT has a huge potential for developing intelligent applications
in almost every vertical market such as smart home, smart healthcare, smart transportation, etc. For example, there are many IoT
applications that have been already successfully implemented for
smart traffic systems, fleet tracking solutions, control of logistics
chain, smart cities, smart metering, industrial automation, collision avoidance systems in cars, energy efficiency, smart buildings/homes/offices, environment monitoring, etc. IoT applications
provide a set of functionalities and capabilities which can be
grouped according to domain of utilization into four areas: monitoring (devices condition, environment state, notifications, alert,
etc.), control (control of devices functions), optimization (device
performances, diagnostics, repair, etc.) and autonomy (autonomous
operations). A survey by Gluhak et al. [28] presents facilities for
practical aspects of IoT.
Application domain manages application services that are usually provided through the IoT middleware layer. Therefore, software
and APIs can be mapped to the application or middleware domain.
However, we grouped IoT enabling technologies according to their
functionalities and based on this consideration we included software (including OS) and APIs to this layer. For enabling functionalities of application domain there are some common embedded
operating systems in use such as TinyOS, Contiki, LiteOS, Android,
Riot OS, etc. These systems support low-power Internet communication and they require very few kilobytes of RAM. The SDKs (Software Development Kit) for these operating systems provides a software framework for various microcontroller firmware’s to be run
on IoT devices. SDK framework supports application programming
by using various programming languages such as c, C++, C#, Java,
etc. Another important building block of IoT systems are software
platforms that enable integration of IoT objects with network technologies by using various communication protocols. These platforms provide an appropriate APIs or they are used in various environments for monitoring and controlling purposes [29]. Also, there
are some platforms that provide other functionalities such as a
development of services independently of hardware, data storage,
and analytics, etc. Examples of these Cloud-based platforms are
AWS IoT Platform, IBM Bluemix Platform, Microsoft Azure, Google
Cloud, Platform, ThingWorx, Xively, etc. (Table 1).
Key issues in developing IoT applications are related to deployment in various use cases, availability, management, reliability, interoperability, scalability (large-scale deployment and integration),
security (authentication, access control, configuration management,
antivirus protection, cryptography, etc.), and privacy. Research challenges include creating algorithms and schemes to present, analyze
and process data collected by sensors. One of the major challenges
related to incorporating IoT objects on the IoT web-based architecture is difficulty in extending existing approaches such as SOA
(Service Oriented Architecture) [30]. Many IoT applications rely on
REST (Representational State Transfer) or RESTful to provide interoperability but still, there are several issues to overcome. Some of
technological developments and issues in the IoT application domain context have been envisioned in [17].
3.2. Middleware domain
We consider IoT middleware as a system constrained by software and infrastructure which is designed to be the intermediary between IoT objects and application layer. In this domain, we
A. Colakovi ˇ ´c, M. Hadžiali´c / Computer Networks 144 (2018) 17–39 21
grouped technologies that provide functionalities such as aggregating, filtering and processing received data from the IoT devices,
information discovery, machine learning, predictive modeling and
providing access control to the devices for applications. Some literature uses other meanings of IoT middleware such as an approach
that include applications in this domain. As we mentioned before,
some IoT middleware provide OS and API management while enabling IoT applications communicating over heterogeneous interfaces.
The number of connected objects increases as well as raw data
(unstructured data) that needs to be managed and processed. Huge
amounts of data, including real-time data, are collected from heterogeneous sources such as sensors, smart objects as well as social
networks and the web. Data and services are fragmented across
many entities including data storage and processing units. Semantic web technologies such as RDF (Resource Description Framework) are developed as a model for data interchange on the Web
to facilitate data merging. Message formats can be classified into
text-based and binary-based. After conversion data into internal
object structure, it is then ready for processing and visualization
[31]. The most used text-based encodings are XML and JSON and
these formats provide human-readable data. Binary messaging formats such as PBF (Protocol Buffers), Colfer, Protostaff Java object
serialization, AMF (Adobe’s Action Message Format) and Kryo reduce message size and because of that are mostly used to encode
object graphs. Also, there are some encoding techniques to convert text-based messages to binary-based messages such as EXI
(Efficient XML Interchange), BSON, BJSON, UBJSON. These encoding techniques provide message size reduction and improve their
processing.
The most common solutions with IoT middleware functionalities rely on Cloud computing. Cloud platforms enable IoT services
development and data processing independently of the hardware
platform. There are many commercial and open source platforms
(Table 1) that can be evaluated according to gateway support, application protocol support, programming language, etc. Platforms
such as Apache Hadoop, Apache Spark, Apache Kafka, Apache
Storm, Apache Ambari, Apache HBase, Spark Streaming, Druid,
Open TSDB, etc. can be deployed to build an efficient and scalable
IoT platform. In paper [32] some survey has been done with comparisons of these platforms.
It is very important to ensure the high level of security and privacy that include mechanisms such as authentication, access control, configuration management, antivirus protection, cryptography,
etc. The most recent research is focused on cross-domain solutions
to improve security and data quality levels [192]. Also, in the focus
of research community are paradigms known as Fog computing,
MCC (Mobile Edge Computing), MCC (Mobile Cloud Computing),
and Cloudlet. These computing systems based on these paradigms
distributes some resources, processes, and services to data centers
which are closer to the edge of the network to improve IoT system
performances (e.g. response time, throughput, energy efficiency) as
well as to provide better security and privacy. They include several mechanisms similar to Cloud computing but deployed at edge
nodes that are located between IoT devices and Cloud infrastructure. According to this, they can be considered as a part of IoT
middleware.
3.3. Network domain
Network domain includes hardware, software, technologies and
protocols that enable connectivity between objects, and between
objects and global infrastructure (e.g. Internet). IETF RFC 7452
[21] outlines the framework of IoT communication models with
following variations: Device-to-Device Communications, Device-toCloud Communications, Device-to-Gateway Model, Back-End DataFig. 3. IoT protocol stack.
Sharing Model. Recommendation ITU-R M.2083-0 [33] highlights
the importance of key capabilities in different usage scenarios. IoT
systems use an extreme diversity of communication technologies
that need to be interoperable in order to meet IoT requirements.
Also, it is more and more difficult to meet traffic requirements
while the volume of traffic increases. ITU-T Y.2060 Recommendation
[19] highlights some high-level requirements for networks to support IoT such as identification-based connectivity, autonomic networking, autonomic services provisioning, location-based capabilities, security, privacy, quality, plug and play, manageability. Other
open issues related to this domain are interoperability, scalability,
reliability, mobility management, routing, coverage, resource usage control and management, self-configuration, energy efficiency,
spectrum flexibility, bandwidth, latency, etc. These issues will be
described in the following Section of this paper.
IoT protocol stack based on TCP/IP reference model with the
most common protocols is presented in Fig. 3. Various protocols
need to interoperate thus we need to use appropriate communication system architecture. Several IoT communication architectures
have been proposed by authors and projects [34]. However, there
is still problem of interoperability between various network technologies. Palattella et al. [35] consider that the standardized approach based on latest developments is the only way to the future
development of IoT. Developing new IoT-based protocols and architectures will play an important role in the following years.
Application layer protocols used in traditional Internet services
are not the corresponding option for IoT due to constraints of
LLNs (Low Power and Lossy Networks). Because of that, there have
been created protocols utilized to connect to things as well as
to end-user applications. The most used application layer protocols are CoAP (Constrained Application Protocol), MQTT (Message
Queuing Telemetry Transport), MQTT-SN (MQTT For Sensor Networks), AMQP (Advanced Message Queuing Protocol), XMPP (Extensible Messaging and Presence Protocol), DDS (Data Distribution Service)- Some basic information about these protocols are
presented in Table 2. There are some other projects that define
other protocols and groups of protocols to be used in IoT solutions
such as Mihini/M3DA, LLAP (Lightweight Local Automation Protocol), LWM2M (Lightweight M2M). Also, some IoT applications may
use other protocols such as HTTP, SSH, etc. Service Discovery Protocols such as DNS-SD (DNS based service discovery), SSDP (Simple
Service Discovery Protocol), SLP (Service Location Protocol), UDDI
(Universal Description, Discovery, and Integration), mDNS (Multicast
DNS), Lightweight Directory Access Protocol (LDAP), APIPA (Automatic
Private IP Addressing), Physical Web, HyperCat, UPnP (Universal Plug
and Play) are used to enable seamless and efficient discovery functions. These protocols can be grouped by function: discovery (DNSSD, SLP, UDDI), naming (mDNS, LDAP) and addressing (APIPA) protocols.
22 A. Colakovi ˇ ´c, M. Hadžiali´c / Computer Networks 144 (2018) 17–39
Table 2
Comparison between the IoT application protocols.
Application protocol Standard RESTful support Transport protocol Security QoS support
CoAP [36] IETF RFC 7252 Yes UDP DTLS Yes
MQTT [37] OASIS Standard No TCP TLS/SSL Yes
MQTT-SN IBM Zurich Research website No TCP TLS/SSL Yes
XMPP [38,39] IETF RFC 6120, 6121 No TCP TLS/SSL No
AMQP [40] ISO and IEC No TCP TLS/SSL Yes
DDS [41] OMG (Object Management Group) No UDP DTLS Yes
HTTP IETF (RFC 2068, RFC 2616, RFC 7230), W3C Yes TCP SSL No
WebSocket [42] IETF Internet Draft Yes TCP TLS/SSL No
Fig. 4. Wireless communication technologies for Internet of Things.
Beside TCP (Transmission Control Protocol) [43] and UDP (User
Datagram Protocol) [44] at transport layer, there are some other
transport protocols used in experimental phases such as QUIC
(Quick UDP Internet Connections) [195] and NanoIP (Nano Internet
Protocol) [196]. QUIC is designed by Google to support a set of
multiplexed connections between two endpoints over UDP. This
protocol was designed to provide security protection equivalent to
TLS/SSL, flow control equivalent to HTTP/2, and to reduce connection and transport latency as well as avoiding congestion by using mechanism of congestion control equivalent to TCP. NanoIP is
a concept based on two transport techniques: nanoUDP for an unreliable and simple transport and nanoTCP which provides retransmissions and flow control. This project provides an alternative networking stack for control, automation and sensor networks without the overhead of TCP/IP.
IPv6 (Internet Protocol version 6) [45] is one of the key enablers
for IoT [46] because IPv4 have gotten exhausted. It is a protocol
for packet-switched networks that provides end-to-end datagram
transmission across multiple IP networks. RPL (Routing Protocol for
Low Power and Lossy Networks) [47] is mostly used routing protocol
for IoT-based applications. It was created by IETF working group
ROLL with the aim to support minimal routing requirements with
simple and complex traffic models while enabling robust topology
over lossy links with updating the routing information.
The most of IoT services rely on wireless technologies because
of availability and mobility requirements. In Fig. 4 we present
the most common wireless communication technologies used for
IoT. Network technologies can be grouped into ANs (Access Networks) and CNs (Core Networks). Access networks such as WiFi, WiMAX, GERAN, UTRAN, eUTRAN, satellite communication, etc.
provide connectivity between IoT objects and CNs. Core networks
such as 3GPP CNs, ETSI TISPAN CN, etc. provide as interconnection
with other networks as well as roaming. There are various efforts
to adapt to LLNs environment and to support lightweight traffic
(low overhead) requirements. 6LoWPAN (IPv6 over Low Power Wireless Personal Area Networks) was created by the 6LoWPAN IETF WG
[51–54] as a convergence layer to adapt link layer to IPv6 packets
over IEEE 802.15.4 networks. It integrates IPv6 based infrastructure
and WSNs to support low bandwidth, header compression, fragmentation, limited packet size (max. packet size to be transmitted
on PHY layer is 127 bytes), multi-hop delivery and various address
lengths described in [35]. It does not provide any routing capabilities and this task is provided by upper layers (e.g. RPL).
Some of the key enabling technologies for IoT are RFID (Radio
Frequency Identification) and NFC (Near Field Communication) [48–
50]. RFID is short range communication technology that uses an
electromagnetic field to automatically identify and track tags attached to objects. RFID systems must not interfere with other systems such as radio emergency services or broadcasters of television
signals, etc. Integrating sensor technologies and RFID enable a lot
of new possibilities in IoT paradigm while enabling sensing, computing and connectivity capabilities in passive systems. The benefit of RFID technology has extended with tracking capabilities and
making data accessible through the Internet. NFC is based on the
ISO/IEC 18092:2004 standard and this technology is created on the
RFID to enable a short-range communication. Each NFC tag has a
unique identifier UID (Unique Identification). When it has Internet
connectivity there is a possibility of data exchange with online services to extend its benefits. Atzori et al. [16] have presented some
basic comparison between RFID, WSN, and RFID sensor networks.
IEEE 802.15.4 (Low-Rate Wireless Personal Area Networks – LRWPANs) [55] was developed as a sub-layer for MAC (Medium Access Control) and PHY (Physical Layer). It is utilized for IoT due
to reliable communication, low-power consumption, low data rate,
low cost, high message throughput, security, encryption, authentication, and support for a large number of nodes [5]. This standard
is the basis for several specifications such as ZigBee, ISA100.11a,
MiWi, WirelessHART. The most popular specification based on
this standard is ZigBee (IEEE 802.15.4-2006) that operates in the
2.4 GHz frequency range with 250 kbps and 1024 as the maximum
A. Colakovi ˇ ´c, M. Hadžiali´c / Computer Networks 144 (2018) 17–39 23
number of nodes. It is a low-power, low-cost wireless technology
(standard) that is deployed in many WSNs but its single channel nature makes it unreliable [6]. Also, it is not energy efficient
while it requires router nodes to always be active [35]. This technology can be used with 6LoWPAN and traditional Internet protocols that empower its capabilities. However, it does not support
QoS and this is very challenging issue. For example, interesting research area is the ZigBee evaluation to support QoS by exploiting
a mix of IP multicasting, queuing management and traffic analytics
techniques [56].
IEEE 802.15.1 (Bluetooth) operates in the 2.4 GHz worldwide
available ISM (Industrial, Scientific and Medical) band and is one
of the key enabling technologies for short-range IoT applications.
Bluetooth SIG (Special Interest Group) proposed BLE (Bluetooth
Low-Energy) in the Bluetooth 4.0 specification and Bluetooth 5
[57] as the latest version of the Bluetooth core specification to enable collecting and aggregating data from devices (sensors) which
generate data at a very low rate. It is designed for short range
(up to 50 m) which is suitable for control and monitoring applications. BLE is also known as Bluetooth Smart protocol for shortrange communication with low power consumption. Previous studies such as [58–62] have presented some of BLE functionalities
with the conclusion of being a good option for some IoT case studies. IETF 6LoWPAN WG developed specification that enables transmission IPv6 packets over BLE [63] that empowered the IoT capabilities of this technology. New version Bluetooth 5 focuses on
improvement of speed, range, security, energy efficiency, locationbased functionalities, interoperability and coexistence with other
technologies. It brings some major advances to the technology to
make it a key enabler of IoT. Also, there are some other efforts
to improve Bluetooth 5 such as Eddystone protocol [197]. It is released by Google in 2015 to define a BLE message format for proximity beacon messages.
IEEE 802.11 is set of MAC and PHY specifications for WLAN
(Wireless Local Area Network) mostly known as Wi-Fi (Wireless
Fidelity). The main issue of this technology is large energy consumption compared to Bluetooth and Zigbee [6]. Some improvements are required to overcome this limitation as well as to improve mobility, roaming, and QoS performances. There are some
improvements such as IEEE 802.11ah (Low-Power Wi-Fi) [198] supports a wide range of IoT applications while being able to provide
more energy efficiency, QoS, scalability (a large number of devices)
and cost-effective solutions [64,65].
There are technologies that are primarily designed for IoT/M2M
applications that need a wide area coverage, long battery lifetime, low bandwidth, low-cost devices. These technologies are
known as LPWAN (Low Power Wide Area Network) [66] such as:
Weightless [199], NB-IoT (Narrow-Band IoT) [200], LoRa WAN (Low
Power Wide Area Network) [203], Sigfox [204], RPMA (Random
Phase Multiple Access) [205], Wi-Fi HaLow [206], etc. In many
cases, they operate in an unlicensed spectrum which is one of
the main reasons leading to downsides such as scalability issues
for a large-scale of devices due to spectrum congestion. Also,
many IoT applications rely on data transfer over cellular technologies such as 2G (GSM, D-AMPS, PDC), 2.5G (GPRS), 2.75G (EDGE),
3G (UMTS/WCDMA, HSPA, HSUPA, EvDO), 4G (i.e. LTE, LTE-A), 5G.
M2M (Machine-to-Machine) connectivity is referred within the cellular context or MTC (Machine-type Communication) within 3GPP
(3rd Generation Partnership Project). 3G and 4G technologies such
as 3GPP LTE are enabling technologies that offer wide area coverage, QoS support, mobility and roaming support, scalability, billing,
high level of security, the simplicity of management as well as connectivity of sensors through a standardized API [6]. LTE-A (Long
Term Evolution – Advanced) and Mobile WiMAX Release 2 (Wireless MAN – Advanced or IEEE 802.16 m) enabling higher speeds,
more scalability, and low costs. In order to respond to the emerging IoT market needs and to avoid technology fragmentation, 3GPP
has made major efforts in Release-13 and Release-14. For this
purpose, 3GPP specified technologies such as eMTC (enhanced
Machine-Type Communication), NB-IoT, and EC-GSM-IoT. The eMTC
brings some LTE enhancements for MTC such as a new Power Save
Mode (PSM). Release 14 brings new eMTC feature enhancements
such as support for positioning and multicast, mobility for interfrequency measurements, and higher data rates [202]. It brings
enhancements such as lower costs, reduced data rate/bandwidth,
and some other protocol optimizations. Also, Release-14 delivers
new enhancements for the NB-IoT technology such as support for
multicast, power consumption and latency reduction, mobility and
service continuity enhancements, etc. EC-GSM-IoT delivered EGPRS
enhancements, which in combination with PSM makes GSM/EDGE
systems IoT ready. This technology brings improvements such as
extended coverage, support for massive number of devices: at least
50.000 per cell, improved security compared to GSM/EDGE, etc.
Summary for eMTC, NB-IoT, and EC-GSM-IoT has been delivered in
the report on progress on 3GPP IoT [201]. All these technologies
meet some of the requirements for IoT with but some other challenges occur. For example, QoS and network congestion are very
challenging issues due to a huge number of deployed nodes (devices) [67].
A global perspective of 5G (5th generation mobile networks
or 5th generation wireless systems) considers capabilities from
ITU-R M.2083-0 and relates them to following use cases: mobile
broadband, massive-machine communication, and mission-critical
communication that should provide the possibility of full deployment of IoT solutions. 3GPP Releases 15 and 16 focuses to deliver the first set of 5G standards as well as the maturing of
the LTE-Advanced Pro specifications. First functional specifications
are expected in second half of 2018 which will include a performance evaluation against mMTC (massive Machine Type Communications) requirements, specifications for eMBB (enhanced Mobile
Broadband) and URLLC (Ultra-Reliable and Low Latency Communications), etc. A detailed survey of the 5G cellular network architecture and some key emerging technologies such as interference
management, spectrum sharing with cognitive radio, cloud computing, SDN, etc. has been presented in paper [4]. Some authors
consider that there is a significant overlap among IoT and 5G. 5G
design efforts are in place to support large-scale deployment of
devices to enable global IoT as well as lower energy consumption
with lower costs. An overview of unique characteristics and some
comparisons of these technologies are compared in paper [191].
According to previous considerations, 3GPP has been working to
support M2M application but still there are numerous challenges
to overcome such as issues related to the energy efficiency, battery lifetime, network coverage, user identification, security, QoS,
complexity, variety of IoT applications, etc. There are numerous
open issues in 5G network design that need to be surveyed including advantages and shortcomings of backhaul solutions [69].
Enabling D2D (Device-to-Device) communication (exchange data
without the involvement of base station or with its partial aid)
is one of the turning points in a cellular system [68]. Also,
very interesting research areas related to deploying 3GPP technologies for IoT are Fog computing, context-aware services, QoS
management, etc.
3.4. Object domain
Object domain presents endpoint layer that includes physical
things (real world entities) and virtual things (virtual entities).
These objects have various capabilities such as sensing, actuation,
identifying, data storage and processing, connecting with other objects, integration into communication networks, etc. IoT objects include embedded software (operating system, onboard application,
24 A. Colakovi ˇ ´c, M. Hadžiali´c / Computer Networks 144 (2018) 17–39
etc.) and hardware (electrical and mechanical components with
embedded sensors, processors, connectivity antennas, etc.).
Sensors are objects that detect and measure some events or
changes in its environment such as temperature, air pressure, acceleration, light, movement, etc. They perform various actions to
provide an output for future processing. Various sensors are embedded in many objects (e.g. smartphones) to enable value-added
services based on IoT. All devices should be identified by unique
ID and connected with other objects and/or with IoT middleware.
There are many identification methods including EPC (Electronic
Product Code), uCode (Ubiquitous codes), QR (Quick Response) or
matrix barcodes, etc. RFID identification has the similar function
as Bar Code but is more advanced because it doesn’t require the
reader to be physically placed in front of it to have optical vision.
RFID can be used as an actuator to trigger different events and
even has modification abilities which Barcodes clearly do not have
[70]. RFID tags (active or passive) have a unique identifier and the
most commonly used is EPC. Active tags have a battery attached to
the object and have continuously transmitted signals while passive
tags emit signals only when it is triggered. As the unique identifier can be used the uID (Unique/Universal/Ubiquitous Identifier)
architecture. Another technology with similar identification management is NFC.
Connectivity components enable wireless or wired connections
by using different communication technologies which allow an exchange of information between different objects. SN (Sensor Networks) is a collection of sensors which communicates between
each other or/and transmits data to some other infrastructure (e.g.
Fog or/and Cloud). SN consists of the sensors, actuators, firmware
and a thin layer of software framework. All these capabilities enable objects to be aware of their environment and to exchange data
which is one of the goals of IoT. The most of IoT products use WSN
(Wireless Sensor Networks) solutions which are mostly based on
the IEEE 802.15.4 standard. This standard defines the MAC and PHY
layers for low power as well as low bit rate communications in
WPAN (Wireless Personal Area Network). IoT devices may contain
gateways that collect data from sensors and send it over the Internet to other infrastructure (e.g. Cloud). They may be connected
to other objects or networks via multiple gateways that can also
act as a proxy between devices and networks. It is not strongly
standardized how IoT devices are connected to the Internet apart
from networking protocols [2]. IoT hardware platforms can facilitate communication, data flow as well as device, data and application management.
IoT constrained devices such as sensors have many challenges
due to the requirements for identity management, processing,
memory, connectivity, and energy capabilities. Many IoT projects
were failed because of these issues [71]. Also, sensors have limited
processing abilities and because of that they usually do not process data and instead they forward data to another infrastructure
(data storage) across the network. Another major research field is
to enhance sensors in sense of energy efficiency. Sometimes it is
not possible to replace the sensor batteries due to dynamic environments; therefore, the design of low power sensors and sensors
that not need any change of battery due the lifetime is a very attractive topic for future research.
4. IoT key issues and challenges
IoT based systems are usually complex due to a tremendous
impact on all aspects of human lives as well as its various technologies deployed to enable autonomous data exchange between
embedded devices. Development of IoT has an impact on various
aspects of human lives (e.g. security, safety, health, mobility, energy efficiency, environmental sustainability, etc.). Therefore, IoT
related issues and challenges need to be considered from various aspects such as enabling technologies, services and applications, business models, social and environmental impacts (Fig. 5).
Analysis of recent contributions and research papers show that the
most of the open issues arise due to an increasing number of connected devices which causes increased traffic demands with new
traffic models. Other issues are related to the integration of various technologies, heterogeneous environment (e.g. various devices,
data types, and network technologies), increased data storage and
processing demands, privacy and security risks, etc. Therefore, in
our focus are issues considered from technology-based perspective.
We performed a classification of previous research and present a
literature pool to bring order in the literature by classifying existing works according to different research areas (Table 3).
We consider that full potential of IoT can be achieved only by
deploying the corresponding architectures and enabling technologies. Because of that, there is a need for continuous improvement
considering all aspects mentioned before. Therefore, it is very important to identify major issues and challenges related to the IoT
development. This paper provides the insight into current research
trends and provide some future research directions. The literature
used in this paper is grouped according to research areas to facilitate future research.
4.1. Standardization
Diversities in technologies and standards are identified as one
of the major challenges in the development of IoT applications
[5,72]. Standardization of IoT architecture and communication
technologies is considered as a backbone for the IoT development
in the future [5,35]. These facts imply that open standards are one
of the key factors for the successful deployment of IoT. This type of
standards is an important facilitator for innovation because of their
availability to the public. They are being developed, approved, and
maintained by a collaborative consensus-based decision-making
process to provide better interoperability for systems using different technologies. Also, by using open standards there is less chance
of being limited to a specific vendor or technology which is very
important factor for IoT development.
The main standardization bodies such as ITU, ETSI, IETF, IEEE,
W3C, OneM2M, OASIS, NIST, etc. are involved in the effort to make
a framework for IoT standards. The scopes of standardization activities are various in order to provide open standards and architectures, seamless connectivity, interoperability, etc. But despite enormous efforts from standardization bodies and alliances, there is
still no reference standard for IoT platform [10]. There is a problem to integrate various standards and contributions to be consistent and coherent. ETSI White Paper No.3 [73] highlighted several
problems and typical consequences of non-coherent contributions
such as incompleteness, inadequate interfaces (reference points),
poor handling of options, lack of clarity and poor maintenance. All
these open issues and challenges need to be considered in future
to enable seamless connectivity as well as the integration and interoperability among various IoT enabling technologies.
4.2. Architecture
IoT systems should be framed within open IoT architecture
and set of standards to enable integration of various technologies
and to support full interoperability. IoT architecture needs to
provide interoperability and to support full mobility to ensure
service continuity (without interruption). Accordingly, one of the
main challenges for IoT systems is to use an open, integrated
and standardized architecture with separated application logic
and hardware infrastructure. The corresponding architectures
need to support heterogeneous nature of things, networks, data
A. Colakovi ˇ ´c, M. Hadžiali´c / Computer Networks 144 (2018) 17–39 25
Fig. 5. Various aspects related to IoT.
and applications to support full interoperability. Key design requirements for IoT architecture are scalability, interoperability,
openness, and modularity in a heterogeneous environment. IoT architecture should enable multi-systems integration, cross-domain
interactions, simple and scalable management functionalities, data
analytics and user-friendly applications as well as the possibility to
include the intelligence and automation across the IoT system. It
may be treated as a system or paradigm which may consist physical objects (e.g. sensors, actuators), virtual objects (e.g. fog/cloud
services, communication layers and protocols) or a hybrid of these
two perspectives [74]. According to this perspective, IoT architecture can be classified into four groups: general/system-level,
hardware/network, software and process [24].
General architecture considers a conceptual model to meet IoT
requirements. There are numerous proposed IoT reference models
but still, there is no general architecture which provides full interoperability. The existing approaches and efforts to solve this issue
are based on layered frameworks and architectures [7]. In Table 4
there is an overview of some existing general architectures and
conceptual models.
Hardware/Network architectures have to enable interoperability
among various networks and communication technologies to provide full connectivity between IoT objects. IoT devices can be
grouped into two groups based on TCP/IP protocol suite support. In order to solve this heterogeneity related issue, there have
been proposed several approaches such as APIs, gateway solutions,
SDN-based solutions, NFV (Network Functions Virtualization), CCN
(Content-Centric Networking), etc. Also, there are some other proposals such as a wearable IoT architecture with the ability to offer traceability of streamed data from source and the devices engaged with [188]. Other challenges are related to various technologies deployed in different networks including communication interfaces and access controls. All these issues must be considered
in hardware and network architectures. Reference model and architecture based on IoT communication stack is presented in EU
(European Union) project IoT-A [34]. This model is a good option
for various cases but for some application domains other architectures should be investigated such as cloud centric architecture for
cost efficiency services [17]. Examples of hardware/network-based
architectures which support IoT paradigm are: EPCglobal based on
RFID and EPC technologies [82,83], Sensor and WSN based architecture [84,85], peer-to-peer [86] and autonomic [87]. Despite the
enormous efforts of standardization bodies and alliances, industries, academic and others there are still numerous open issues related to these architectures. For example, many recent researches
focus on the questions such as “which computing paradigm (MCC,
MEC, cloudlet, fog and cloud computing) to deploy in specific use
case” or “when and where to deploy a specific computing system
to facilitate the communication between IoT devices and application” [193].
Software architectures should provide a common set of services to enable processing (aggregation, computation, etc.) large
amounts of data for service composition. IoT software architecture
and framework need to be used to overcome the complexity of
systems and to provide an environment for IoT services composition. IoT software platform should be created as OAP (Open Application Platform) to enable modular design as well as providing an open APIs (Application Programming Interface) to sensors
and other devices. For example, JMS (Java Message Service) is a
Java Message Oriented Middleware (MOM) API for sending messages between two or more clients. Also, there must be an integrated development environment such as Java and HTML5. Current
IoT applications are mostly domain-specific with fragmented architectures that cannot integrate the data from different sources [23].
In order to enable integration of various services with other infrastructure (e.g. Fog and/or Cloud) it can be used APIs [5], gateways [56,88], virtualization [89], SDN (Software Defined Networking) [90], cloud centric architecture [17,91], etc. There are several
approaches to provide application framework and another set of
services for IoT such as SOA, RESTful, architectures based on fog
and cloud computing, web application framework based on Google
Toolkit [92], etc. These architectures cover Operating systems, IoT
middleware, APIs, Data management, Big data, etc. Examples of
Operating Systems (OS) for IoT support are Contiki, Riot OS, Android, TinyOS, LiteOS, etc. Future researches need to focus on improvements of RTOS (Real-Time Operating Systems), APIs and other
components within IoT architectures to adapt these systems to IoT.
CORE (IETF Constrained RESTful Environments) working group
has defined the subset of the RESTful specification [93] with CoAP
to meet requirements of IoT applications such as interoperability
with HTTP (HyperText Transfer Protocol), low overhead, multicast.
RESTful principles are investigated in researches such as [84,94,95].
Used content type is XML (eXtensible Markup Language) or JSON
(JavaScript Object Notation) which depends on HTTP server. This
architecture can be applied in smartphone applications as it only
requires an HTTP library but straightforward implementation of
RESTful architecture is not possible due to requirements of LLNs as
described in [35]. ETSI SmartM2M and oneM2M are projects based
on RESTful design with their aim to resolve various fragmentation
issues and to enable interoperability but there are still some open
issues such as scalability and mobility as well as integrating RESTful based services into business process. Although these issues are
highly scalable and distributed RESTful compliant architecture is
considered as one of the best solutions today because some features such as authentication, caching and others can be utilized as
well as all cloud platforms having the support for it.
SOA architecture is recognized as a good solution for the IoT
middleware but we need to emphasize that sometimes (in some
cases) application layer is not considered as a part of the IoT middleware. A design of SOA for IoT is a big challenge while it needs
26 A. Colakovi ˇ ´c, M. Hadžiali´c / Computer Networks 144 (2018) 17–39
Table 3
Research and literature pool addressing IoT open issues and challenges[126,178,179,180,181,182,183,184,185].
to handle many devices connected to the system which phrases
scalability issues [74] and the difficulty in extending SOA [30]. Examples of implementation of SOA approach are CORBA, Jini, UPnP,
OPC-UA, etc. Some specific IoT case studies use SOA architecture
in different ways. For example, middleware layer can have a purpose to develop and handle an infrastructure for data processing
and transmitting data to a gateway or actuating node while other
functionalities from SOA can be distributed on other layers and
components. Some surveys of IoT middleware have been done in
[96,97] while overviews of SOA architecture are given in numerous
researches such as [98–100].
Process architecture should incorporate business processes into
IoT paradigm [101,102]. IoT enables new business opportunities and
brings possibilities for new business models and value chains. The
major challenge is to structure workflows to support IoT environment. Conventional conceptual frameworks need to be extended in
order to enable process-aware IoT [8]. There are some novel architectural paradigms for the future IoT which are based on allocation
A. Colakovi ˇ ´c, M. Hadžiali´c / Computer Networks 144 (2018) 17–39 27
Table 4
Examples of IoT conceptual models.
Author Description
ITU-T Rec. Y.2060 [19] Reference model based on four layers: Device layer, Network layer, Service and Application support, Application layer.
ETSI TS 102 690 [75] A functional architecture framework and high-level architecture for M2M where logical entities comprise this architecture.
ETSI TS 102 689 [76] M2M high level system overview.
IEEE P2413 [77] Three-tier IoT architecture.
CCSA [23] General IoT architecture with functional platforms: Sensing and Gateway, Resource and Administration, Open Application.
GISFI IoT WG [78] IoT Reference Architecture consist 5 layers: Sensor / Devices, Gateway, Core Network, Service Platform, Applications.
H2020 UNIFY-IoT [27] IoT platforms with in-depth analysis as well as components across the IoT architectural layers.
IoT-A [34] Reference models and architectures from different perspectives.
Alhamedi [79] Internet of things communication reference model.
Vermesan and Friess [80] The initial IoT6 architecture design approach.
Kaiwartya et al [81] Five layered architecture for IoV (Internet of Vehicles) with description of functionalities for each layer.
Díaz et al. [32] A generic architecture of IoT web applications.
Kim et al. [8] Advanced conceptual architecture for organizing IoT-based pervasive communities and societies.
Rizzardi, et al. [190] IoT cross-domain modular architecture communicating by means of RESTful services.
of resources and processes between data sources and other infrastructure such as edge computing. This is very interesting research
area with many open issues such as the question “which functions
to allocate to the things, edge and cloud and how to enable interaction between these architectural levels.”
4.3. Interoperability and integration
Interoperability is the ability of multiple devices and systems to
interoperate regardless of deployed hardware and software. A variety of standards and technologies used for IoT development as well
as various solutions from different vendors leads to massive heterogeneity which causes interoperability issues. IoT interoperability
issue is considered across all layers. To overcome this issue, it need
to be used a layered framework with standardized architecture. According to this approach, ETSI White Paper No. 3 [73] presents four
interoperability levels technical, syntactical, semantic, and organizational.
Technical interoperability is usually associated with communication infrastructure and protocols. IoT systems need to provide
interoperability over heterogeneous devices, networks and a variety of communication protocols such as IPv6, IPv4, 6LoWPAN/RPL,
CoAP/CoRE, ZigBee, GSM/GPRS, Wi-Fi, Bluetooth, RFID, etc. The
existing Internet architecture doesn’t support full connectivity of
the heterogeneous devices and this is still a challenging task that
causes complex integration issues [6,103]. Significant challenge is
the integration of different subsystems regardless of their native
communication protocols. This should enable the co-existing of
used protocols and seamless communication even beyond IP because some technologies deployed in IoT systems are not based
on IP. Therefore, in order to enable the ubiquitous connectivity, it
must be developed HetNet (Heterogeneous Networking) paradigm
that supports different MAC/PHY. An abstraction layer is necessary to abstract heterogeneity [31]. Also, there is a requirement
for mechanisms for management and coordination as referred to
[6,12,104]. IEEE 1905.1 standard [105] was designed for interoperability support and provides a common interface (abstraction
layer to hide diversity of MAC) widely deployed in home networking technologies. It provides interoperability as well as secure
connections, facilities network management, path selection, auto
configuration, extends network coverage, and supports end-to-end
QoS. Solutions for interoperability problems are usually based on
APIs [5], implementation of gateways [56,88], integrating smart
resource-constrained objects into the Internet using virtual networks [89], the approach based on SDN [90], etc.
Gateways provide several solutions such as protocol conversion or centralizing remote connectivity depending on purpose and
layer where they are implemented. Interoperable gateways at object layer support multiple interfaces to enable devices to be connected through a different kind of ANs (Access Networks) while
their implementation at the network layer enables connectivity between various network technologies including ANs and CNs. According to a study [106] avoiding gateways as architectural elements at object layer can support full connectivity and peer-topeer networking solutions while gateway roles can be addressed
in smart objects such as smartphones. This solution enables devices to be directly connected to any other device or point on the
Internet which is a true IoT vision. Also, this can solve the potential problem of implementing gateways such as slowing down a
full end-to-end connectivity but the challenge is to enhance devices capabilities such as computing, storage, communication interfaces, low power consumption, etc. The solution for this problem can be smartphone-based mobile gateways as presented in
[10,107,108] while everyday smartphones capabilities are improved
[109]. There are several challenges related to this solution such as
development of mobile gateway software architecture which supports full interoperability, limited storage and computation capabilities, efficiency in battery consumption especially when concurrent
communication interfaces are deployed.
API based solutions for interoperability can be used for the automatic conversion between various application protocols. According to used protocols, IoT devices can be grouped into two groups
based on support for TCP/IP protocol suite. IoT applications that
use CoAP, MQTT, MQTT-SN, AMQP, REST and some other support
TCP/IP but also there are applications without this support and because of that interoperability issue occurs. Al-Fuqaha et al. [5] emphasize the need for new protocols for communication compatibility in a heterogeneous environment. APIs can be implemented
in different software and cloud platforms. Some operating systems
like Contiki, Riot OS, Android, TinyOS, LiteOS, etc. enable modular design and provide open APIs which are necessary for application interoperability. There are several interesting research topics
related to APIs interoperability such as to provide a trustworthy interoperability by using a proxy framework for API interoperability
in the IoT.
IoT platforms improve interoperability in a heterogeneous environment and provide functionalities such as connectivity of various
objects using communication technologies, device management,
data processing, data visualization, etc. According to their functionalities, IoT platforms can be grouped as hardware platforms, software platforms, and cloud platforms. These platforms can include
various solutions such as different network interfaces, gateways,
APIs, etc. IoT hardware platforms enable devices connectivity as well
as data processing outside the data center. They provide gateways
to enable connectivity of devices with various network technologies. IoT hardware platforms provide technical interoperability by
enabling development of IoT products which deploy various de-
28 A. Colakovi ˇ ´c, M. Hadžiali´c / Computer Networks 144 (2018) 17–39
vices and networks. Examples of hardware platforms are Arduino,
Raspberry Pi, Gadgeteer, BeagleBoard, pcDuino, etc. However, there
is still the challenge to develop a platform able to interact with
all technologies deployed in IoT. Another major issue is to enable
interconnection between various platforms to empower their capabilities. Other issues are related to power supply, energy efficiency,
security, seamless connectivity, mobility of platforms, etc. IoT software platforms enable IoT objects to integrate (sensors/actuators)
with network technologies by using various communication protocols. These platforms integrate development environment (HTML5,
Java, etc.) and provide APIs to enable developers to communicate
with IoT devices over various network technologies. Cloud platforms
enable to develop IoT solutions based on three models of cloud service: Software as a Service (SaaS), IaaS (Infrastructure as a Service)
and PaaS (Platform as a Service). Some analysis of the most used
and leading IoT platforms on today’s market have been done in
H2020 – UNIFY-IoT Project [27]. Key issues for all IoT platforms are
related to supporting technologies such as processors and semiconductors, sensors, communication hardware and protocols, operating systems, developer tools, analytic tools, etc. Another key issue
for these platforms is related to security (authentication, authorization, access control, intrusion detection, recovery mechanisms,
etc.).
There are some other solutions for technical interoperability issues at different layers which provide dynamic, flexible and automated management and reconfiguration of the network [6]. The
aim is to simplify network design and management by resolving
some technical interoperability issues. In paper [110], authors proposed a cognitive management framework for IoT to solve heterogeneity problems among devices and associated services. Another
interesting solution is a hub-based approach that provides scalable
and reliable communication and improve some interoperability issues [111,112]. Also, software architectures such as SOA reduce the
system integration problems and improve interoperability among
heterogeneous IoT devices in physical networks [113]. It is done by
providing a powerful framework which support connectivity and
component integration in IoT systems. The primarily goal of this
architecture is to enhance IoT application interoperability and extendibility at the service and application layers. However, despite
the flexibility offered by this framework, there are some issues related to SOA architecture such as the lack of an intelligent and
connection-aware framework to support interoperability.
Syntactical interoperability is associated with understanding content (information) and refers to data formats, syntaxes, and coding
such as XML and HTML. IoT applications need to integrate the data
from different sources [23]. The new software architecture must
enable searching, aggregating and processing the data generated by
heterogeneous devices [24]. For that purpose, it has to be used a
standardized data formats, syntaxes, and coding. IoT middleware
needs to include mechanisms such as APIs solution for support
interoperability within the diverse applications, services, and data
format.
Semantic interoperability enables interpretation of content (the
meaning of information) to be shared by communicating parties
[114]. The term “semantic” in the IoT refers to the possibility of
extracting knowledge from raw data collected from sensors. This
“knowledge” enables to provide useful services and reports based
on analyzed data [115]. The evolution of “Semantic Technologies”
enables some level of data interoperability as well as advanced
decision-making. Semantic interoperability enables IoT objects to
learn, think and understand social and physical worlds. A common high-level framework with adequate architecture as well as
new techniques such as data mining are needed to enable extraction meta-information (convert raw data to knowledge). For example, semantic level interoperability architecture for pervasive computing and IoT is presented in [114]. Some of the semantic technologies for the IoT including JSON, W3C, OWL (Web Ontology
Language), RDF (Resource Description Framework), EXI (Efficient
XML Interchange), and WSDL (Web Services Description Language)
have been presented in [116]. A variety of these technologies leads
to heterogeneity at a semantic level which is significant research
field. One of the possible solutions is to use the semantic model
with XML and ontology [117,118]. Also, there are some useful data
standards such as IOTDB, SensorML, Semantic Sensor Net Ontology
– W3C, Wolfram Language, RAML (RESTful API Modeling Language),
SENML (Media Types for Sensor Markup Language), LsDL (Lemonbeat smart Device Language), etc. However, it is still open issue to
design a semantic IoT framework and open data standards to support full interoperability.
Organizational interoperability is usually associated with the
ability of data exchange even though using different information
systems and infrastructure. There is wide area of research topics
related to IoT interoperability issues such as: developing new APIs,
enhancement existing gateways or creating new gateway, improving devices capabilities to enable seamless connectivity and interoperability, improvement of solutions based on integrating smart
resource-constrained objects into the Internet using virtual networks, using concept based on SDN, etc. SDN is very interesting
research area in recent years because it is considered as emerging
networking technology which provide the ability to establish and
manage virtualized resources as well as to provide some level of
interoperability without deploying a new hardware. All solutions
mentioned above require the corresponding architecture to enable
full interoperability.
4.4. Availability and reliability
Availability of services is one of the key issues to be addressed to properly manage the dynamics of IoT systems. Availability means that IoT applications should be available anywhere
and anytime for every authorized object. The objects that are going to be connected should be adaptive and intelligent to support
seamless connectivity and desired availability.
Availability of network and its coverage area must enable the
continuity of the services use regardless of mobility, dynamic
change of network topology or currently used technologies. All this
requires mechanisms for interoperability, handover, and recovery
in case of some unattended operations. An appropriate monitoring system, protocols, and self-healing mechanisms need to be deployed to enable robustness of system [119]. Some communication
technologies suffer from intermittent availability which can cause
service interruption. For example, some IoT applications can rely
only on satellite communication which fluctuating quality. Therefore, it must be enabled computing (data collecting, processing,
control, etc.) which is independent of sending data over the Internet or other networks to the computing infrastructure. MCC, MEC,
Cloudlet, and Fog computing are paradigms proposed to overcome
some of these issues. However, there are some new challenges as
described before in this paper.
Mobility is another major challenge in IoT systems where services are provided to mobile users. IoT devices may be moved and
frequent topology changes can occur. The goal is to create a robust system in spite of these dynamic changes. Because of that,
there is a requirement for efficient mobility management mechanisms [120]. MIPv6 (Mobile IPv6) is a protocol developed to support mobility in IPv6 networks. In addition, IPSec (Internet Protocol
Security) is mandatory for MIPv6 in order to support trust between
home agent and mobile device. Also, some deployments of IoT systems imply that devices need to know their locations and to be
aware of its environment (location of neighbor devices). This is another major challenge for future research especially when developing some real-time applications that require a location-awareness.
A. Colakovi ˇ ´c, M. Hadžiali´c / Computer Networks 144 (2018) 17–39 29
Routing issues are very important for reliability due to scheduling and routing in the multi-hop mesh topologies. Routing processes need to support dynamic topology changes, multihop routing, scalability, context awareness security mechanisms, QoS, etc.
Also, routing protocols need to be context-aware and energy-aware
(green routing protocols). Some surveys for multicast routing in
the IoT environment have been done in [121] where authors propose algorithms for the establishing a multicast routing tree. RSVP
(Resource Reservation Protocol) and MPLS (Multi-Protocol Label
Switching) solve several routing problems. Also, deploying RPL
routing protocol provide some solutions for routing issues while
connecting LLNs to the Internet [122]. This protocol enables topology adoption and provides efficient routing functionalities. However, it has some weaknesses and limits such as high overhead,
high packet loss, and latency in mobile conditions [123]. Another
challenge is to provide a trade-off between reliability and energy
consumption. This is a reason to use UDP as a transport protocol
while retransmission control mechanisms could be implemented at
the application layer [35]. Therefore, it must be deployed efficient
upper-layer protocols (transport and application) that provide endto-end reliability. Another solution for these problems is to develop
a new protocol extension such as a new mobility support layer
(MoMoRo) for Low Power Wireless Sensor Networks [194].
The IEEE 802.15.4e enables high reliability while it deploys
TSCH (Time Synchronized Channel Hopping) as its part [35]. It
defines a simple scheme for shared cells to solve a problem of
collision as well as synchronization (acknowledgment-based and
frame-based). However, this issue is still open to further improvement.
4.5. Data storage, processing and visualization
With a huge increase of connected objects and data traffic volume, there is a requirement for new calibration and analytic techniques. IoT systems need a common analytic platform to support
a big data which need to be delivered as a service to IoT applications. Various data mining methods such as AI (Artificial Intelligence), machine learning and others intelligent decision-making
algorithms enable computational processes to discover patterns in
large datasets. These techniques can be used to organize raw data
as well as to extract usable information and knowledge from it but
costs are a limitation.
To handle continuously increased amount of data it seems
that only Cloud technology can effectively meet these requirements. There are several Cloud platforms that exist on the market (Table 1) with different storage and computing capabilities,
supporting application protocols, interoperability, gateway support,
billing models, etc. However, there are some issues that need to be
considered. Transferring big data from the edge devices (e.g. sensors, smartphones, etc.) to the cloud infrastructure brings the following issues: network performances (e.g. delays, bandwidth, congestion, reliability, availability, etc.), costs of moving data through
the Internet (it is very expensive), cost of storing data on cloud
servers, security of data transmission and storing, privacy issues,
etc. In paper [5], authors identified several other challenges related to deploying Cloud computing for IoT such as synchronization and standardization between different cloud vendors, balancing between IoT requirements and cloud services environments, reliability in sense of security between devices and cloud platforms,
management of cloud computing and IoT, etc. The interesting challenge is to develop a scalable and high performances hybrid cloud
platform. Also, there is a need for new algorithms for raw data filtering, selection, abstraction, and aggregation. Surveys on state of
the art for Cloud computing and IoT integration has been done in
[1,31,124,129].
Cloud computing is not always necessary such as in the cases
when local infrastructure resources are sufficient for data storage
and processing. Processing raw data at locally deployed nodes can
reduce the amount of data needed to be transferred through the
Internet. This reduces data congestion, latency, costs, and improves
some other performances. New computing paradigms such as MCC,
MEC, Cloudlet, and Fog computing deploy rule-based system applied to a local infrastructure in order to provide an extension of
Cloud computing. These systems improve some IoT performances
such as QoS, reliability, mobility, security, and privacy. For example, fog computing enables to deploy smart devices like smartphones and home gateways for preprocessing data and contextaware computing. However, these platforms are not tailor-made
in all use cases due to lack of capabilities for complex analysis
and storing huge amount of data. These functionalities are complemented with Cloud computing. Therefore, only some basic computation processes can be done at edge layer. In order to overcome resource limitations of local infrastructure, data need to be
forwarded to a Cloud. This issue is related to the question when
and where to deploy some computing system. Also, integration of
IoT and edge computing systems have to face with static and dynamic IoT devices in order to provide mobility support. This leads
to services migration challenge and computational issue due to low
power requirements. Another important issue related to mobility is
collaboration and synchronization between edge nodes. A mechanism for aggregating data could be implemented at the gateway
to manage data flows but this solution implies deployment of adequate software. Therefore, integration of IoT and edge computing
doesn’t solve issues such as mobility, complex analysis, security
and privacy of data on edge devices [127,128]. Also, interesting research topic is context-aware computing that was identified as one
of the key factors for IoT development [31,125].
There are a few current and forthcoming problems and some
specific challenges related to data visualization such as to provide
tools for the interaction of the user with the IoT environment, visualization of the raw data in a meaningful way and according to
the end-user needs [17], etc. IoT applications require connectivity to infrastructure (e.g. cloud and network) and they must support user-friendly interfaces to enable secure and remote control
over various devices. Eliminating the need for physical controls in
the object itself by using new digital user interfaces is less costly
and has more customizes capabilities. IoT applications can be autonomous or controlled by humans (directly, passively or hybrid).
This is very interesting research area due to several challenges such
as understanding different types of human-in-the-loops controls,
how to implement feedback control, etc. Visualization of data gathered from sensors includes charts, animations, maps, tracking location on maps, etc. GUI (Graphic User Interface) provides visualization of performed measurements and enables to perform different
control actions. Also, there are some threats while extracting information from raw data such as the oversimplification of data or
overreliance on visuals. These issues imply requirements for methods of data visualization which enable reducing or illustrating data
in simplified ways. However, this simplification of big data could
lead to unfounded conclusions. Web technologies such as HTML
5 provide solutions for some of these issues. It specifies canvas
elements for dynamic rendering of the 2D graphic while WebGL
(Web Graphics Library) is JavaScript API running on HTML 5 canvas element for advanced 2D and 3D graphics application development [32]. Emerging technologies such as touchscreen technologies, 3D screens, and others can provide an efficient way to navigate the data as well as to extract useful information from raw
data. Although the evolution of visualization technologies from CRT
to Plasma, LCD, LED and beyond this is still open research area. For
example, a very interesting and emerging research topic is the representation of sensors in the 3D landscape.
30 A. Colakovi ˇ ´c, M. Hadžiali´c / Computer Networks 144 (2018) 17–39
4.6. Scalability
Scalability is the ability to add new devices and services to
IoT system without degradation of existing service performances.
A key challenge related to scalability is to support a large number of various devices with memory, processing, bandwidth and
other resource constraints [107]. Scalable mechanisms must be deployed for efficient discovery of devices but also to enable their interoperability. To enable scalability as well as interoperability there
must be used a layered framework and architecture [7]. Design of
IoT architectures that support scalability is a big challenge when
it comes to the future development of IoT systems. These architectures need to handle numerous devices connected to the system which phrases scalability issues [74]. One of the possible solutions is to use highly scalable cloud-based platforms with the possibility to store a huge amount of collected data. The Therefore,
a Cloud of Things [124,129] can be used as a global architecture
that scales up cloud computing. Another solution are edge computing systems that extends Cloud services to edge devices. This
technology provides storage, computing, and some networking services between devices and Cloud infrastructure. The problem is
that edge computing systems cannot provide functionalities such
as complex analysis, data access to a large number of users and
storing historical data which are complemented with Cloud computing [31]. Another challenge is to enable context-aware computing with scalability support. This problem emphasizes issues of object performances including storage and processing capabilities including the power consumption. A major scalability related problem is to provide a seamless connectivity to make it easy when
adding new components and objects to IoT system as well as to
support topology change. To solve these problems in the context of
distribution scalability, mobility and security there is a CCN vision
of the next generation of network architecture [130]. It enables automatic and application-neutral caching in memory wherever it is
located in the network. This paradigm is still in its nascent stage
and it is a very attractive research topic.
4.7. Management and self-configuration
Managing IoT applications and devices is a very critical factor for successful IoT deployments [131]. Management functionalities such as monitoring, control, and configuration are a big challenge due to IoT complexity, heterogeneity, a huge number of deployed devices and traffic demands. IoT software must be able to
identify various smart objects and interact with them to provide
efficient management and self-configuration functionalities. Selfconfiguration means IoT system has capabilities of the dynamical
adoption of changes in its environment. For example, if devices
could switch off when there is no activity it will provide more efficiency in energy consumption.
Data management mechanisms need to provide various functionalities such as raw data aggregation, data analytics, data recovery, and security. They need to enable a different kind of reports
that include: descriptive (e.g. products’ condition), diagnostic (e.g.
causes of failure), predictive (e.g. expected events), etc. IoT data
management framework presented in [132] includes applications
(analysis) layer, query layer, federation layer, a source layer, communication layer, things layer. Also, these mechanisms need to be
adaptive, scalable, and trustworthy [113]. This implies the usage
of new approaches to data aggregation and complex computations
to provide efficient and real-time decision-making [133]. Another
challenge is to provide automatic decisions and self-configuring
operations in complex, integrated and open IoT systems. Objects
need to gain some knowledge from collected data and according
to that performing some context-aware actions. A very interesting
research topic is to automatically allocate tasks between layers of
system architecture. For example, it is challenging task to determine which functions should be allocated to Fog nodes rather than
Cloud infrastructure.
Network management functionalities need to provide efficiency
in network topology management, devices synchronization as well
as traffic and congestion control management. A new network’s design needs to deploy efficient management mechanisms to manage the large-scale of connected devices, an enormous amount of
data (traffic loads) and various services with different QoS requirements. Monitoring network infrastructure enables detection of any
changes and events that effect on network resource usage and security. There is a need for dynamic resource management solution
with a resource allocation scheme which will be effective under
the uncertainty nature of IoT environment. Various protocols have
been developed to monitor and control network elements such
as devices, gateways, terminal servers, etc. LNMP (LoWPAN Network Management Protocol) and SNMP (Simple Network Management Protocol) are existing management protocols for IPv6 based
networks [15] that perform some network management functions.
Also, TSMP (Time Synchronized Mesh Protocol) is a communication protocol that enables synchronization of devices (motes) in
self-organizing wireless networks. There have been proposed various novel concepts and technologies for the efficient management of networks. For example, Rajan et al. [131] proposed a novel
heterogeneous and self-optimizing SN management with a flexible auto-configuration of WSNs. RANaaS (Radio Access Network as
a Service) is a concept that has been developed to manage network resources and to enable flexible management [134]. SDN is
an enabling technology for 5G systems [135] that has been developed to provide dynamic, flexible and automated management
and reconfiguration of a network [6] as well as simplifying network design and management [136]. It enables a cost-effectively
scaling necessary for IoT services. SDN and NFV (Network Function Virtualization) enable a new way of network management by
providing a virtualization of some network functions to be managed through software (e.g. servers as computing platforms). Also,
these paradigms provide functionalities to manage heterogeneous
devices with various deployments and use-cases [137]. Although
these paradigms bring new efficient network management functions there are still some open issues [138]. For example, fragmentation of APIs and controller functions in SDN is a very challenging
problem which causes standardization issue. Also, it is a challenge
to determine how often QoS signaling data should be sent from
network to a controller component.
Devices management mechanisms need to provide monitoring
and remote-control functionalities including remote devices’ activation or deactivation, firmware update, etc. Some device management functionalities can be included in SBC (Single Board
Computer) but for remote control, there must be deployed other
mechanisms including devices and services management protocols. Managing devices and enabling seamless integration in various networks are challenges due to a deployment of various hardware and software while providing operations such as addressing and optimization at the architectural and protocol levels [139].
One of the major issues in IoT systems is the identity management of devices as well as ensuring trusted environment [140]. The
OMA (Open Mobile Alliance) and its Device Management Working Group specify some protocols and mechanisms for devices and
services management in resource-constrained environments such
as LWM2M (The Light-weight M2M) [141]. There are some other
light protocols for devices management including NETCONF Light
protocol [142]. The challenge of device management is especially
pronounced because of the heterogeneity among devices and associated services. There are several efforts to solve this problem
such as cognitive management framework solution [110] and hardware platforms which enable integration of objects and networks
A. Colakovi ˇ ´c, M. Hadžiali´c / Computer Networks 144 (2018) 17–39 31
with a few management functionalities. Some other open issues
are related to the development of lightweight and secure IoT device management framework which provide functionalities such as
location awareness, mobility, low power consumption, support for
various mobile OS, etc.
4.8. Network performances and QoS
The ability to handle numerous connected devices with various services and processes depends on IoT system performances.
Major issues are related to traffic loads and various traffic models [9] which have a dramatic impact on a networks’ performance
and QoS (Quality of Service). Some significant issues related to
QoS include bandwidth, throughput, latency, etc. Also, the connected devices’ growth and increased data rates in wireless networks have caused radio spectrum usage to be a critical issue
[143]. Some countries might have a problem to find an additional
spectrum as a spectral efficiency of radio networks is reaching
its physical limits. All these facts influence on performances of
IoT systems which are considered as extremely significant in most
applications [144].
QoS in IoT system depends on deployed technologies, protocols,
traffic demands, etc. Several studies have been done for the evaluation of IoT service performances but this is still an open issue
[145]. QoE (Quality of Experience) is related to user’s perception
and need to be considered in some application scenarios but not
in all cases. In IoT applications where QoE is not of interest than
QoS – based approach could be used [146]. The most of the previous researches considered QoS parameters for IoT services while
only a few studies considered QoE [147]. Due to the different traffic loads and characteristics of multiple traffic types, it is necessary to develop new models for forecasting traffic load. For example, it is needed a novel traffic model for a new-generation of SN
[186]. Human-based communications (H2H and H2M) and traffic
models are difficult to apply directly to IoT. Laya et al. [148] have
presented some differs between M2M (IoT) and human-based traffic characteristics. In order to simplify traffic management, there
is a need for data (traffic) aggregation science a huge diversity of
traffic types and sources. Some impact of data aggregation in IoT
(M2M) on QoS parameters such as delay and throughput has been
surveyed in [149]. A challenge that occurs due to new traffic requirements is a need for new QoS specifications. There is a requirement for a new controlled and optimal approach for different traffic types and applications [17] as well as efficient congestion control, dynamic scheduling, resource allocation algorithms,
buffering mechanisms, etc. QoS-aware scheduling mechanisms are
required to support various traffic characteristics and QoS specifications [150]. Another challenge is related to QoS management
while IoT devices operate under bandwidth, battery, and processing constraints [151]. Also, QoS-based service selection is the key
to large-scale service-oriented IoT [152]. This is one of the major
issues in the future development of IoT services due to the increasing number of services with various QoS requirements.
IoT performances depend on all components and communication protocols deployed in a system. One of the major challenges
is to select the corresponding protocols for a specific use case (according to various traffic requirements) while there is no evaluation that includes all protocols and scenarios [153]. The existing
surveys are mostly focused on the comparative analysis of various
protocols with an aim of evaluating their performances in specific
IoT scenarios. Sheng et al. [154] have done a survey on protocols
and technologies according to standards, challenges, and opportunities. CoAP, MQTT, XMPP, AMQP, REST and DDS protocols have
been addressed as a common application layer protocols used in
IoT systems [5,153,155]. In [32] were analyzed MQTT, AMQP, XMPP,
and DDS by measuring latency and throughput rate by running on
three different web platforms: HTML 5, Adobe Flash and Microsoft
Silverlight. The survey [32] concluded that HTML 5 is a much more
mature platform for real-time IoT applications although Adobe
Flash has the best performances while Microsoft Silverlight has
the poorest performances. CoAP and HTTP response time and energy consumption were analyzed by Colitti et al. [156]. The performance evaluation of MQTT and CoAP in different use cases have
been done in [107,157]. Some performances comparisons between
AMQP and RESTful have been presented by Fernandes et al. [158].
Some characteristics of WebSocket and HTTP while delivering realtime messages have been described in [159]. The most of the application layer protocol use some of existing transport protocols
which cause several open issues such as fairness problem [189].
TCP protocol was not originally developed for wireless networks
and it exhibits serious network performance degradation in these
networks [160]. UDP is a lightweight protocol and is better for
real-time applications but it is not reliable. Also, routing protocols need to be adopted for effective communication in IoT systems to ensure reliability and to reduce delay. Some of existing
routing algorithms used in IoT systems are: RPL (Routing Protocol
for Low Power and Lossy Networks), AOMDV-IoT (Ad-hoc on Demand Multipath Distance Vector), EARA (Energy Aware Ant Routing Algorithm), PAIR (Pruned Adaptive IoT Routing), SMRP (Secure
Multihop Routing Protocol), REL (Routing protocol based on Energy
and Link Quality), etc. IoT routing protocols can be evaluated by
using many parameters such as multihop routing support, energy
aware, context-aware, security support, dynamic topology support,
heterogeneity support, link quality, reliability, latency, data redundancy, load balancing, scalability, routing information maintaining,
etc. Routing protocols affect reliability and other performances of
IoT systems as we described before. Therefore, optimizations of
routing techniques, energy efficient routing protocols and developing more intelligent routing algorithms (context-aware algorithms)
are very interesting research areas.
QoS performances such as latency, jitter, packet loss, and reliability have to be considered at all layers of network architecture. The corresponding option for system-level architecture and
optimization of resources allocation leads to QoS improvement.
Many IoT applications rely on the Cloud infrastructure to handle
a huge amount of data that need to be stored and processed.
However, transferring big data from the edge devices (e.g. sensors, smartphones, etc.) to the cloud infrastructure brings the following network issues: network performances (e.g. delays, bandwidth, congestion, reliability), costs of moving data through the Internet, etc. These are very challenging issues especially for latencysensitive applications because many IoT applications require very
low end-to-end latency (e.g. within a few milliseconds) and jitter.
Also, Cloud computing is not sufficient to handle mobility support,
geo-distribution, and real-time location-awareness. The concept of
edge computing is proposed to overcome some of these challenges
and limitations. It is a novel system-level architecture that distributes some resources, processes, and services to data centers
which are closer to the edge of the network. For example, fog computing can provide small and deterministic latency for IoT applications which is crucial for some real-time systems. It provides
better QoS in terms of better response time, jitter and throughput. Edge computing technologies enable QoS awareness by providing dynamically adaption to the available resources. However,
edge computing is not tailor-made in all use cases due to computational limitations. To overcome the computational limitation,
it is proposed IFCIoT (Integrated Fog Cloud IoT) architecture. Integration of IoT and Fog/Cloud computing causes some new open
issues such as the metric of computation (e.g. resource and functions allocation) and development of dynamic resource allocation
algorithms.
32 A. Colakovi ˇ ´c, M. Hadžiali´c / Computer Networks 144 (2018) 17–39
4.9. Modeling and simulation
Major challenges in developing IoT services are due to its complexity and heterogeneity in all parts of system architecture. Heterogeneous nature of applications, devices, interfaces, radio technologies, etc. leads to the IoT system modeling issue. There is
no standard methodology available for modeling such real-world
complex IoT-based systems [214]. Therefore, IoT system modeling
for finding eligible deployments is a very challenging issue [208].
Some contributions in theoretical modeling of IoT based on edge
computing was presented in [209–211]. However, there is lack of
mathematical formulation for systems based on the integration of
IoT, computing systems, and other parts of IoT system architecture.
This formulation could help to design IoT system with the most appropriate technologies according to specific requirements. IoT devices are usually resource-constrained in sense of identifying, sensing, networking, computation and other capabilities. Therefore, one
of the key issues is a QoS-aware deployment of IoT through the
corresponding computational offloading system. Because of that,
the future of IoT depends on decentralizing networks. There are
different edge computing models which help to offload IoT devices
and Cloud infrastructure. The most of IoT applications rely on the
support of some computing systems which are the part of IoT system architecture. Even though cloud computing supports the development of IoT applications this model is facing many challenges
such as network delays, cost of transferring raw data through the
Internet, etc. Therefore, some IoT applications require a new computing model to support the ubiquitous deployment of sensors and
other devices. These systems are part of paradigms known as Fog
computing, MCC, MCC, and Cloudlet. They are based on an allocation of resources between data sources and cloud infrastructure. However, there is a challenge of determining which functions
should be allocated to edge nodes rather than Cloud [128]. The allocation of tasks between layers of system architecture needs to
be enabled dynamically based on the system resource state [207].
Using an appropriate model of integration of IoT and computing
systems can reduce the amount of data that need to be transmitted to the cloud infrastructure over the Internet. This improves
QoS performances as well as reduces resource consumption and
costs.
Simulation tools such as Opnet, NS-3, Cloudsim, and others can
be used for understanding and modeling the IoT system. However, the complexity and heterogeneity of IoT scenarios complicate
these processes [211]. This imposes the use of sophisticated, hybrid and multi-level modeling and simulation techniques [212]. An
overview of some other modeling and simulation challenges have
been presented in paper [213]. For example, one of the major issues in existing simulation tools is lack of integrated options to
simulate network and cloud infrastructure to obtain overall performance of IoT system. Also, there is a problem of simulating various protocols, security attacks, computing and other IoT processes
in order to obtain different results such as network performances,
energy consumption, etc. Another important issue is enabling simulation of IoT scenarios which include a large number of heterogeneous devices with various traffic loads and types. This implies
that the problem of simulating IoT scenarios is not only related
to software tools but also hardware performances have to provide
enormous resources such as CPU, RAM, etc. According to previous
considerations, a new enhancement of simulation tools should improve the ability to simulate small, medium, and large-scale IoT
scenarios. Simulation and modeling tools need to support a dynamic nature of IoT, real-time requirements and increasing processing requirements. These scenarios include deployment of heterogeneous technologies. To address these shortcomings there is a
need for the continuous enhancement of simulation and modeling
tools.
4.10. Unique identification
Each IoT object needs to have a unique identifier such as IP address or URI (Uniform Resource Identifier) and this is considered as
one of the most critical factors for IoT success [17]. Appropriate
identity management with unique identifiers and efficient key distribution schemes are issues highlighted in [140,161]. If each object
has a unique identifier and connection to the Internet than objects
can be monitored, controlled and managed throughout the entire
lifecycle. In that context, it must be differentiated objects ID and
its network address (IPv4 or IPv6). Also, some objects need to have
multiple identifiers while sensors, actuators, and other parts are
attached to them. The most critical features related to addressing
schemes such as uniqueness, reliability, persistence and scalability
are described in paper [17].
There are many identification methods and technologies for IoT
objects such as EPC (Electronic Product Code), uCode (Ubiquitous
codes), QR (Quick Response) or matrix barcodes, etc. EPCglobal
is an organization which efforts are in the field of standardization and integration EPC with RFID technology. Registration, unique
identification and discovery in a context-aware way of IoT devices
are issues that require future researches.
IPv4 have got exhausted and it is challenge to provide a unique
network address to all objects. The solution for this problem is
IPv6 that uses 128-bit addresses which provide a huge amount of
addresses. IPv6 is considered as one of the key enablers for IoT.
Integration of IPv6 and its related protocols into constrained capabilities of WSN is identified as one of its major challenges [15].
Because of scalability issues, a manual and static management
of system resources is not an appropriate solution. To solve this
problem, it can be used some of service discovery protocols such
as DNS-SD, SSDP, SLP, mDNS, APIPA, etc. These protocols can be
grouped by their function as we described before. However, there
is a challenge to adapt these protocols for IoT services [162] because of requirement for autonomous registration. Discovery functionality need be dynamically adapted with the inclusion of new
IoT devices in the network. Therefore, IoT architecture should enable to devices to join or leave IoT platform without affect to all
IoT system and this is another open issue for network architecture
which need to support IoT services.
4.11. Power and energy consumption
Power and energy storage technologies need to meet various IoT
requirements such as providing energy sources for small embedded devices. Supplying reliable power to the sensors and devices
as well as developing low-power chipsets is highlighted as a very
challenging problem [2]. Wireless power technologies can transmit
power at some distance and this is promising solution but we are
being still in the early phase of development of these technologies. Another significant challenge is to integrate large computation with small embedded devices that have low power consumption particularly in the case of image and video processing-based
applications. Lanzisera et al. [163] highlighted the challenge of producing low-cost terminal with low active power. One of the major issues is improving device’s capabilities (e.g. computation and
networking) while reducing both costs of devices and power consumption. Also, a low power communication stack is identified as
one of the core requirements related to power and energy efficiency [35].
Existing communication protocols such as HTTP and TCP are not
optimized for IoT as not considered to be deployed in low-power
systems. Energy efficient MAC protocol and appropriate routing
protocol are identified as critical factors for the system to be efficient [17]. Standardization bodies such as ITU-T, IETF, ISO, IEEE, etc.
have defined a several protocols and interfaces on the low com-
A. Colakovi ˇ ´c, M. Hadžiali´c / Computer Networks 144 (2018) 17–39 33
Fig. 6. IoT security and privacy issues.
munication stack layer but the actual deployments of such features are still not satisfactory. One of the proposed solutions is
IEEE 802.15.4 [55] and its improvements such as IEEE 802.15.4e.
This standard defines PHY and MAC layers for the low-power but
it is still ill-suited for low-power multi-hop networks as explained
in [35]. This is another open issue where future effort should improve the capability of IEEE 802.15.4 low-power radio technology.
Another research area includes optimization of routing techniques
to reduce energy requirements and optimize energy consumption.
For example, elimination of data redundancy will reduce energy
requirements for data routing. There are some new technologies
such as networking paradigm known as IoNT (Internet of NanoThings) that can decrease the power consumption but it is still in
nascent phase of development.
4.12. Security and privacy
Security and privacy issues are identified as key challenges in
deployment of IoT solutions [31] because there are numerous examples of threats, vulnerabilities and risks [164]. Several security models and threats taxonomy models for the IoT systems
have been proposed [165,166]. According to Hewlett Packard Enterprise Research study [167], most of the devices privacy concerns
raised due to: insufficient authentication and authorization, lack of
transport encryption, insecure web interface, insecure software and
firmware, etc. In Fig. 6 we present the most common IoT security
and privacy related issues.
To provide confidence in IoT systems security and privacy need
to be considered from various aspects including legal, social and
cultural point of view [168]. Security functionalities need to be embedded at every level of IoT architecture and efficient trust management [169,170] must be deployed. This is the reason why IoT
security architecture is still evolving [171,192] as well as various
mechanisms being developed to improve security and privacy. Security mechanisms should provide authentication, access control,
data integrity and privacy, encryption and other capabilities while
enabling automatic data processing based on the policies and rules
configured by users. These mechanisms must operate in real-time
and they need to be cost effective and scalable to minimize the
complexity and maximize usability. One key issue is lacking a solution for IoT IITP (Identity Theft Prevention) [171] consequently
authorization scheme for IoT which needs to be context-aware. For
example, IoT objects should be aware of location in order to provide robust security. Also, security issues are more emphasized in
a heterogeneous environment and because of the standardization
perspective of data exchange [172]. The most of security issues are
related to communication threats such as: Malicious Code Injection, Sniffing Attack, Spear-Phishing Attack, DoS (Denial-of-Service
Attack), Sybil Attack, Proxy Attack, Sleep Deprivation attack, etc.
Because of these attacks various mechanisms must be deployed
such as: authorization, authentication, encryption, anti-virus protection, etc. Although these mechanisms improve security level in
IoT systems there are many issues that still need to be considered.
For example, proxy attack or man-in-the-middle attack can occur regardless that the transmitted signal is encrypted or not.
IoT systems require reliable and secure communication protocols at all protocol stack layers. There are three main solutions
to provide security at application layer. The best solution to provide superior security properties is to develop a custom application
layer security protocol but design of such protocol is very complex.
Some standards and protocols such as OTrP (Open Trust Protocol)
[173] are used by applications to install, update, and delete applications as well as to manage security configuration. Another solution for enhancement of security is to use IPSec (Internet Protocol Security) but it is not suitable for all IoT applications and
it is much easier to run the protocol over TLS (Transport Layer
Security) [174] which provide a transparent connection-oriented
channel. Some IoT application protocols use other specific methods to enhance security but the most of security solutions rely
on cryptography protocols such as SSL (Secure Sockets Layer) and
DTLS (Datagram Transport Layer Security). These protocols are implemented between the application layer and the transport layer
of TCP/IP protocol stack. TLS must run over a reliable transport
channel (typically TCP) but some IoT applications prefer to use
UDP. Thus, a datagram-compatible variant of TLS is required. DTLS
[175] is a protocol based on TLS which provides equivalent security
guarantees for datagram protocols. Security protocols use various
mechanisms and standards such as X.509 which is used to manage
digital certificates and public-key encryption in TLS. CoAP use DTLS
while a compressed version of DTLS is used in Lightweight Secure
CoAP for the IoT. XMPP and AMQP for security purpose use TLS
and SASL (Simple Authentication and Security Layer). MQTT application protocol is mostly based on leveraging TLS/SSL [5] but there
are some other solutions such as OASIS MQTT that uses Cyber security framework or a new secure MQTT mechanism named AUPS
(AUthenticated Publish & Subscribe) [190].
Security risk emphasizes more when IoT system uses wireless
communications technologies as well because of the system openness, physical accessibility to some components such as sensors,
etc. There should be deployed mechanisms for malicious activities’
detection and mechanisms for recovery (self-healing). IPSec provide end-to-end security at the network layer and it can be used
with various transport protocols. Link layer security is limited to
provide secure communication between devices [171] while effective algorithm for encryption must be used. Encryption is one of
the key elements of ensuring information security but is a very
challenging issue to encrypt large volumes of real-time that data
needs to be transferred. Encryption algorithm needs to be efficient
in power consumption. It is a very challenging task to implement
complex schemes for improving security in the environment where
IoT components should deploy a low energy. Very interesting research area includes interoperable lightweight protocols and encryption algorithms for better security in IoT environment.
Beside various mechanisms, it is needed appropriate policies to
protect privacy and make sure all users feel comfortable using IoT
34 A. Colakovi ˇ ´c, M. Hadžiali´c / Computer Networks 144 (2018) 17–39
solutions. Privacy policies need to allow some dynamic changes
due to changes in the IoT environment. One of the key challenges
is due to an openness of and interoperability of system with others while each system has its own privacy policies. Each object in
an IoT system should be able to check the other’s privacy policies
for compatibility before sharing data [161]. The privacy policies for
infrastructure and applications must be specified by users (human
entities – data owners or physical entities – things). Also, one of
crucial factors is a management policy to guarantee a good level of
security [190].
4.13. Environmental issues
Internet of Things has both positive and negative impacts on
the environment. Every day there is more and more devices that
are being deployed thus “environmental friendliness” is a topic
that should be given more attention in future research. Environmental sustainability is one of the greatest concern due to increasing energy demands and electronic waste. Examples of interesting research topics are: reduce energy consumption, use renewable sources of energy, reduce a size of devices to decrease
the amount of no degradable materials, various effects on human
health, etc. The Internet consumes up to 5% of the total energy
[17] with more and more demands and this is another issue to be
considered in the future development of IoT-based systems. New
green ICT enabling technologies with general green ICT principles
must be deployed in the development of IoT systems [13,176,177].
However, IoT is providing some new ways to solve many environmental problems. One of the most common use of IoT technologies is to provide an innovative way to address these issues. IoT
provide a great potential for developing new solutions for environmental sustainability such as IoT monitoring systems [187]. This is
research field unprecedented opportunities for future development.
5. Future research
The main scope of this paper is to provide a comprehensive
overview of IoT open issues and challenges. In this Section, we
provide some insights into specific emerging issues and ideas to
be tackled by the future research. Analysis of recent contributions
and research papers show that the most of challenges arise due to
increased traffic demands with various traffic types, greater variance in data formats, variety of IoT devices, heterogeneous networks, etc. All these indicators have a dramatic impact on IoT system performances. There are specific application demands in sense
of computing and communication capabilities, QoS, mobility, reliability, privacy and security, etc. To meet these requirements, IoT
applications rely on other systems such as Cloud computing. There
are some other systems based on Cloud technologies such as fog
computing, cloudlets, MEC, MCC, that can be considered as a part
of IoT system architecture. The most of current IoT applications
and services are cloud-based where smart devices exchange data
with Cloud infrastructure to provide services. This model is facing
many challenges such as technical limitations (e.g. network delays,
throughput, reliability), economic considerations (e.g. cost of transferring data to the Cloud), and some other challenges originated
from social and administrative aspects. Therefore, a new computing paradigm is needed to support emerging IoT applications. During the last year’s there have been proposed several architectures
and computing paradigms which deploy infrastructure closer to
the source of data such as fog computing, cloudlets, MEC, MCC, etc.
The distinction between these paradigms lies in the “infrastructure
distance” from the IoT objects, type of infrastructure where functions are allocated, etc. Cloud computing capabilities outclass the
storage and computing capabilities of the infrastructure at the edge
of the network. It can be used for long-term operations such as
data storage and pattern analysis. However, the integration of IoT
and Cloud computing is not efficient for all IoT applications due
to inherent problems such as unacceptable delay, lack of mobility
support, location-awareness, security and privacy issues, etc. Deploying edge computing systems helps to overcome some issues.
For example, they help to reduce costs of data transferring as well
as to improve system performances such as QoS. However, there
are some bottlenecks for each model of integration and using the
corresponding architecture of integrating IoT and computing systems can alleviate some of the major limitations of IoT systems.
The key questions are “which model of integration to choose and
which functions to allocate to the IoT devices, edge, and cloud systems”, “how to enable interaction between these subsystems”, and
“which technologies to deploy for the development of the IoT system”? These questions put in light IoT system modeling issue.
In order to select the corresponding model of IoT and computing systems integration, there is a need for a modeling methodology. However, there is a lack of mathematical formulation and
evaluation methods which include multiple metrics. Contribution
in this emerging research field could help in finding eligible deployments for any part of IoT system architecture. Therefore, it
is required to propose a comprehensive IoT model which include
all possible architectures, technologies, and integration possibilities. Also, there is a requirement for a quantitative method of evaluating IoT system performances. This method could be used for
selecting the corresponding integration model and technologies as
well as for creating performance-based profiles of IoT applications.
There are many possible metrics for quantitative analysis which
can be used for creating IoT application profiles including power
consumption, computation metrics including CPU metrics, RAM usage, network-related metrics including network latency, throughput, packet loss, etc. Most of the previous studies are based on
analysis of individual metric observed on the specific use case.
There is lack of approaches which integrate more than one metric
in order to allow more comprehensive evaluation of IoT systems.
This is another challenge that needs to be addressed for future research. Therefore, the evaluation method should include multiple
metrics to enable a resource optimization as well as finding eligible and QoS-aware deployment of various architectures, interaction models, and technologies. It could be based on an algorithm
which will enable to optimize the load distribution between IoT
subsystems. By using this kind of evaluation, it could be possible
to propose a classification scheme for IoT application according to
their performance profile (e.g. QoS requirements). Also, this evaluation method could enable evaluation of different application designs and resource management policies. Also, this method could
be used for QoS and QoE evaluation as well as for IoT system modeling. Adequate mathematical formulation of IoT system and evaluation methods can break through the future development.
6. Conclusion
IoT needs to enable a seamless connectivity anytime, anywhere
by anyone and anything to provide intelligent services including
identifying, sensing, networking, processing and visualization capabilities. This concept brought many new possibilities for largescale services and products development which caused a massive
wave of innovations and new business opportunities. Various visions and approaches, as well as the lack of coordination between
standards and technologies, lead to the fragmentation of IoT industries which cause a set of new challenges to be tackled by future
research. IoT-based solutions have become more advanced and sophisticated while there is no a comprehensive framework with integrated all standards and technologies. This open several issues
which arise due to an increasing number of connected devices, integration of various technologies, increased traffic demands with
A. Colakovi ˇ ´c, M. Hadžiali´c / Computer Networks 144 (2018) 17–39 35
new traffic models, raw data storage and processing demands, privacy and security risks, etc. This paper clarified different IoT visions and definitions on the basis of referenced literature. Also, the
paper provides insights into the IoT enabling technologies by presenting the functional domains with key utility factors. Besides existing solutions, there are many emerging technologies which empowers IoT systems with numerous of new capabilities and functionalities. This paper summarized the current state-of-the-art of
IoT enabling and emerging technologies in order to provide a comprehensive list of open issues with some details. Furthermore, this
is used to provide some future research directions. We performed
the classification of previous research and present a literature pool
to bring order in the literature by classifying existing works according to different research areas. Therefore, this paper is mostly
focused on issues related to the IoT enabling and emerging technologies.
New IoT applications and emerging technologies bring new
challenges that require attention from the research community.
We have summarized and discussed these issues and outlined the
main challenges by envisioning perspectives for the most attractive future research topics and developments. One of the most attractive future research topics is related to the integration of IoT
with emerging technologies such as hybrid cloud platforms, MCC,
MEC, fog, and cloudlet, nano technologies, etc. These solutions can
breakthrough future development of IoT with a new architecture
known as Cloud of Things or even beyond in a paradigm known as
IoNT (Internet of Nano-Things). Combining these technologies by
enabling cross-domain integration with existing web technologies
in a concept called WoT (Web of Things) provides an even wider
spectrum of new capabilities and functionalities. However, despite
the enormous research efforts many open issues and challenges remain. We highlight these issues such as IoT system modeling and
requirement for a quantitative method of evaluating system performances. A brief overview of interesting research topics, a classification of open issues according to proposed functional domains and
insights into specific emerging issues and ideas facilitate the future
research. Therefore, this paper is valuable to the research community as it can be used as a starting point for future research.
References
[1] D. Singh, G. Tripathi, A.J. Jara, A survey of internet-of-things: future vision,
architecture, challenges and services, in: Proc. IEEE World Forum on Internet
of Things 2014, At Seoul, 2014, pp. 287–292.
[2] M.H. Miraz, M. Ali, A review on Internet of Things (loT), Internet of Everything (IoE) and Internet of Nano Things (IoNT), in: Proc. Internet Technologies
and Applications (ITA), 2015, pp. 219–224.
[3] M. Condoluci, G. Araniti, T. Mahmoodi, M. Dohler, Enabling the IoT machine
age with 5G: machine-type multicast services for innovative real-time applications, IEEE Access 4 (May) (2016) 5555–5569.
[4] A. Gupta, R.K. Jha, A survey of 5G network: architecture and emerging technologies, IEEE Access 3 (July) (2015) 1206–1232.
[5] A. Al-Fuqaha, M. Guizani, M. Mohammadi, M. Aledhari, M. Ayyash, Internet of
Things: a survey on enabling technologies, protocols, and applications, IEEE
Commun. Surv. Tutorials 17 (June (4)) (2015) 2347–2376.
[6] M.R. Palattella, M. Dohler, A. Grieco, G. Rizzo, J. Torsner, T. Engel, L. Ladid,
Internet of Things in the 5G era: enablers, architecture, and business models,
IEEE J. Sel. Areas Commun. 34 (February (3)) (2016) 510–527.
[7] C. Sarkar, S.N Akshay Uttama Nambi, R.V. Prasad, A. Rahim, R. Neisse, G. Baldini, DIAT: a scalable distributed architecture for IoT, IEEE Internet Things J. 2
(December (3)) (2014) 230–239.
[8] M. Kim, H. Ahn, K.P. Kim, Process-aware Internet of Things: a conceptual extension of the Internet of Things framework and architecture, KSII Trans. Internet Inf. Syst. 10 (August (8)) (2016).
[9] E. Soltanmohammadi, K. Ghavami, M. Naraghi-Pour, A survey of traffic issues
in machine-to-machine communications over LTE, IEEE Internet Things J. 3
(February (6)) (2016) 865–884.
[10] G. Aloi, G. Caliciuri, G. Fortino, R. Gravina, P. Pace, W. Russo, C. Savaglio, Enabling IoT interoperability through opportunistic smarthpone based mobile
gateways, J. Netw. Comput. Appl. 81 (March) (2016) 74–84.
[11] B. Soret, K.I. Pedersen, N.T.K. Jorgensen, V. Fernández-López, Interference coordination for dense wireless networks, IEEE Commun. Mag. 53 (January (1))
(2015) 102–109.
[12] J.G. Andrews, Seven ways that HetNets are a cellular paradigm shift, IEEE
Commun. Mag. 51 (March (3)) (2013) 136–144.
[13] C. Zhu, V.C.M. Leung, L. Shu, E.C.-H. Ngai, Green Internet of Things for smart
world, IEEE Access 3 (November) (2015) 2151–2162.
[14] F. Paganelli, D. Parlanti, A DHT-based discovery service for the Internet of
Things, J. Comput. Netw. Commun. 2012 (September) (2012).
[15] N. Benamar, A. Jara, L. Ladid, D.E. Ouadghiri, Challenges of the Internet of
Things: IPv6 and network management, in: Proc. Eighth International Conference on Innovative Mobile and Internet Services in Ubiquitous Computing,
2014, pp. 328–333.
[16] L. Atzori, A. Iera, G. Morabito, The Internet of Things: a survey, Comput. Netw.
54 (October (15)) (2010) 2787–2805.
[17] J. Gubbi, R. Buyya, S. Marusic, M. Palaniswami, Internet of Things (IoT): a
vision, architectural elements, and future directions, Future Gener. Comput.
Syst. (February) (2013) 1645–1660.
[18] J.A. Stankovic, Research directions for the Internet of Things, IEEE Internet
Things J. 1 (March (1)) (2014) 3–9.
[19] Global Information Infrastructure, Internet protocol aspects and next-generation networks, Next Generation Networks–Frameworks and Functional Architecture Models: Overview of the Internet of Things, 2012 ITU-T Recommendation Y.2060 Series Y.
[20] Internet of Things, (IoT), ISO/IEC JTC 1 Information Technology, 2014 Preliminary Report.
[21] Architectural Considerations in Smart Object Networking, IETF RFC 7452 (March)
(2015).
[22] Internet of Things in 2020, A roadmap for the future, INFSO D.4 networked enterprise & RFID INFSO G.2 micro & nanosystems, Co-operation with the RFID
Working Group of the ETP EPOSS Version 1.1, 27 May, 2008.
[23] S. Chen, H. Xu, D. Liu, B. Hu, H. Wang, A vision of IoT: applications, challenges,
and opportunities with China perspective, IEEE Internet Things J. 1 (July (4))
(2014) 349–359.
[24] A. Whitmore, A. Agarwal, L.D. Xu, The Internet of Things — a survey of topics
and trends, Inf. Syst. Front. (March) (2014) 261–274.
[25] Oxford Dictionaries, 2017. January[Online]. Available http://www.
oxforddictionaries.com/us/definition/american_english/Internet-of-things.
[26] S. Sebastian, P.P. Ray, Development of IoT invasive architecture for complying
with health of home, in: Proc International Conference on Computing and
Communication Systems (I3CS ’15), Shillong, India, April 2015, pp. 79–83.
[27] H2020 – UNIFY-IoT Project, Supporting Internet of Things Activities on
Innovation Ecosystems, 2016 Report on IoT platform activities, October[Online]. Available: http://www.unify-iot.eu/wp-content/uploads/2016/10/
D03_01_WP02_H2020_UNIFY-IoT_Final.pdf.
[28] A. Gluhak, S. Krco, M. Nati, D. Pfisterer, N. Mitton, T. Razafindralambo, A survey on facilities for experimental Internet of Things research, IEEE Commun.
Mag. 49 (November (11)) (2011) 58–67.
[29] K. Gama, R. Wanderley, D. Maranhao, V.C. Garcia, A web-based platform for
scavenger hunt games using the Internet of Things, in: Proc 12th Annual
IEEE Consumer Communications and Networking Conference (CCNC), 2015,
pp. 597–598.
[30] H-C. Hsieh, K-D. Chang, L-F. Wang, J-L. Chen, H-C. Chao, ScriptIoT: a script
framework for and Internet-of-Things applications, IEEE Internet Things J. 3
(September (4)) (2015) 628–636.
[31] Z.B. Babovic, J. Protic, V. Milutinovic, Web performance evaluation for Internet
of Things applications, IEEE Access 4 (October) (2016) 6974–6992.
[32] M. Díaz, C. Martín, B. Rubio, State-of-the-art, challenges, and open issues in
the integration of Internet of Things and cloud computing, J. Netw. Comput.
Appl. 67 (May (C)) (2016) 99–117.
[33] MT Vision, Framework and overall objectives of the future development of
IMT for 2020 and beyond, ITU-R Recommendation M.2083-0 (Sep. 2015).
[34] IoT-A (Internet-of-Things Architecture), Initial Architectural Reference Model
for IoT, 2011 Project Deliverable D1.2, June.
[35] M.R. Palattella, N. Accettura, X. Vilajosana, T. Watteyne, L.A. Grieco, G. Boggia,
M. Dohler, Standardized protocol stack for the Internet of (Important) Things,
IEEE Commun. Surv. Tutorials 15 (December (3)) (2012) 1389–1406.
[36] Z. Shelby, K. Hartke, C. Bormann, The constrained application protocol (CoAP),
IETF RFC 7252 (June) (2014). [Online]. Available https://tools.ietf.org/html/
rfc7252.
[37] MQTT, (Message Queue Telemetry Transport), OASIS Standard, October 2014 [Online]. Available http://mqtt.org/2014/11/
mqtt-v3-1-1-now-an-oasis-standard.
[38] P. Saint-Andre, Extensible messaging and presence protocol (XMPP): core,
IETF RFC 6120 (March) (2011). [Online], Available https://tools.ietf.org/html/
rfc6120.
[39] P. Saint-Andre, “Extensible messaging and presence protocol (XMPP): instant
messaging and presence,” March 2011. [Online], Available https://tools.ietf.
org/html/rfc6121.
[40] Advanced Message Queuing Protocol 1.0 approved as an International Standard,
OASIS Standard, October 2012. [Online]. Available: https://www.amqp.org/.
[41] Data distribution services specification, Object Manage. Group (OMG), April
2015 [Online]. Available http://www.omg.org/spec/DDS/1.2/.
[42] I. Fette, A. Melnikov, The WebSocket protocol, IETF RFC 6455 (December)
(2011). [Online]. Available https://tools.ietf.org/html/rfc6455.
[43] J. Postel, Transmission control protocol, IETF RFC 793 (September) (1981).
[Online]. Available https://tools.ietf.org/html/rfc793.
[44] J. Postel, User datagram protocol, IETF RFC 768 (August) (1980). [Online].
Available https://www.ietf.org/rfc/rfc768.txt.
36 A. Colakovi ˇ ´c, M. Hadžiali´c / Computer Networks 144 (2018) 17–39
[45] S. Deering, R. Hinden, Internet protocol, version 6 (IPv6) specification, IETF
RFC 2460 (December) (1998). [Online]. Available https://tools.ietf.org/html/
rfc2460.
[46] A.J. Jara, L. Ladid, A. Skarmeta, The Internet of Everything through IPv6: an
analysis of challenges, solutions and opportunities, J. Wirel. Mob. Netw. Ubiquitous Comput. Dependable Appl. 4 (Spetember (3)) (2013) 97–118.
[47] T. Winter, et al., RPL: IPv6 routing protocol for low-power and lossy networks,
IETF RFC 6550 (March) (2012). [Online]. Available https://tools.ietf.org/html/
rfc6550.
[48] L. Catarinucci, D. de Donno, L. Mainetti, L. Palano, L. Patrono, M.L. Stefanizzi,
L. Tarricone, An IoT-aware architecture for smart healthcare systems, IEEE Internet Things J. 2 (March (6)) (2015) 515–526.
[49] L. Chunli, L. Donghui, Application and development of RFID technique, in:
Proc. 2nd International Conference on Consumer Electronics, Communications
and Networks (CECNet), 2012, pp. 900–903.
[50] Y. Choi, Y. Choi, D. Kim, J Park, Scheme to guarantee IP continuity for
NFC-based IoT networking, in: Proc. 19th International Conference on Advanced Communication Technology (ICACT), 2017, pp. 695–698.
[51] -N. Kushalnagar, G. Montenegro, C. Schumacher, IPv6 over low power wireless
personal area networks (6LoWPANs): overview, assumptions, problem statement, and goals, IETF RFC 4919 (August) (2007). [Online]. Available https:
//tools.ietf.org/html/rfc4919.
[52] Montenegro, Transmission of IPv6 packets over IEEE 802.15.4 Networks, IETF
RFC 4944 (September) (2007). [Online]. Available https://tools.ietf.org/html/
rfc4944.
[53] J. Hui, P. Thubert, Compression format for IPv6 datagrams over IEEE 802.15.4-
based networks, IETF RFC 6282 (September) (2011). [Online]. Available https:
//tools.ietf.org/html/rfc6282.
[54] B. Campbell, H. Tschofenig, An IETF URN sub-namespace for OAuth, IETF
RFC 6755 (September) (2007). [Online]. Available https://tools.ietf.org/html/
rfc6755.
[55] 802. 15, 4 IEEE Standard for Local and Metropolitan Area Networks – Part
15.4: Low-Rate Wireless Personal Area Networks (LR-WPANs), IEEE Standard,
2011 WG802.15.
[56] A. Al-Fuqaha, A. Khreishah, M. Guizani, A. Rayes, M. Mohammadi, Toward
better horizontal integration among IoT services, IEEE Commun. Mag. 53
(September (9)) (2015) 72–79.
[57] Bluetooth Core Specification, Bluetooth Special Interest Group (SIG),
2016 [Online]. Available https://www.bluetooth.com/specifications/
bluetooth-core-specification.
[58] S. Raza, P. Misra, Z. He, T. Voigt, Bluetooth smart: an enabling technology
for the Internet of Things, in: Proc. IEEE 11th International Conference on
Wireless and Mobile Computing, Networking and Communications (WiMob),
October 2015, pp. 155–162.
[59] J. DeCuir, Introducing bluetooth smart: part 1: a look at both classic and new
technologies, IEEE Consum. Electron. Mag. 3 (January (1)) (2014) 12–18.
[60] R. Frank, W. Bronzi, G. Castignani, T. Engel, Bluetooth low energy: an alternative technology for VANET applications, in: Proc. 11th Annual Conference on Wireless On-demand Network Systems and Services (WONS), 2014,
pp. 104–107.
[61] E. Mackensen, M. Lai, T.M. Wendt, “BluetoothLow Energy, (BLE) based wireless sensors, in: Proc. Sensors, 2012 IEEE, October 2012, pp. 1–4.
[62] M.O. Al Kalaa, W. Balid, N. Bitar, H.H. Refai, Evaluating bluetooth low energy
in realistic wireless environments, in: Proc. IEEE Wireless Communications
and Networking Conference, April 2016, pp. 1–6.
[63] J. Nieminen, T. Savolainen, M. Isomaki, B. Patil, Z. Shelby, C. Gomez, IPv6
over bluetooth(r) low energy, IETF 6Lo Working Group, August, 2015 [Online].
Available https://tools.ietf.org/html/draft- ietf-6lo-btle-17.
[64] T. Adame, A. Bel, B. Bellalta, J. Barcelo, M. Oliver, IEEE 802.11AH: the WiFi
approach for M2M communications, IEEE Wirel. Commun. 21 (January (6))
(2014) 144–152.
[65] E. Khorov, A. Krotov, A. Lyakhov, Modelling machine type communication in
IEEE 802.11ah networks, in: Proc. IEEE International Conference on Communication Workshop (ICCW), June 2015, pp. 1149–1154.
[66] U. Raza, P. Kulkarni, M. Sooriyabandara, Low power wide area networks: an
overview, IEEE Commun. Surv. Tutorials PP (January (99)) (2017) 1–1.
[67] M. Hasan, E. Hossain, D. Niyato, Random access for machine-to-machine communication in LTE-advanced networks: issues and approaches, IEEE Commun.
Mag. 51 (June (6)) (2013) 86–93.
[68] M.N. Tehrani, M. Uysal, H. Yanikomeroglu, Device-to-device communication
in 5G cellular networks: challenges, solutions, and future directions, IEEE
Commun. Mag. 52 (May (5)) (2014) 86–92.
[69] M. Jaber, M.A. Imran, R. Tafazolli, A. Tukmanov, 5G backhaul challenges
and emerging research directions: a survey, IEEE Access 4 (April) (2016)
1743–1766.
[70] U. Farooq, Muhammad Waseem, Sadia Mazhar, Anjum Khairi, Talha Kamal,
A review on Internet of Things (IoT), Int. J. Comput. Appl. 113 (March (1))
(2015) 1–7.
[71] N. Lin, W. Shi, The research on Internet of Things application architecture
based on web, in: Proc. Advanced Research and Technology in Industry Applications (WARTIA), IEEE Workshop, 2014, pp. 184–187.
[72] Noboru Koshizuka, Ken Sakamura, Ubiquitous ID: standards for ubiquitous
computing and the Internet of Things, IEEE Pervasive Comput. 9 (4) (2010)
98–101.
[73] H. van der Veer, A. Wiles, Achieving Technical, Interoperability – the
ETSI Approach, April 2008 ETSI White Paper No. 3[Online]. Available
http://www.etsi.org/images/files/ETSIWhitePapers/IOP%20whitepaper%
20Edition%203%20final.pdf .
[74] P.P. Ray, A survey on Internet of Things architectures, J. King Saud Univ. Comput. Inf. Sci. (October) (2016) 1319–1578.
[75] Machine-to-Machine communications (M2M), Functional architecture, ETSI TS
102 (2011) 690.
[76] Machine-to-Machine communications (M2M), M2M service requirements, ETSI
TS 102 (2013) 689.
[77] P2413 – Standard For an Architectural Framework For the Internet of Things
(IoT), BOG/CAG – Corporate Advisory Group, BOG – IEEE-SA Board of Governors, Active Project.
[78] Technical Report on IoT Reference Architecture, GISFI TR 06.001, 2012.
[79] A.H. Alhamedi, V. Snasel, H.M. Aldosari, A. Abraham, Internet of Things communication reference model, in: Proc. 6th International Conference on Computational Aspects of Social Networks, August 2014, pp. 61–66.
[80] O. Vermesan, P. Friess, Scalable integration framework for heterogeneous
smart objects, applications and services, in: Internet of Things – From Research and Innovation to Market Deployment, River Publishers Series in Communications, Denmark, 2014, pp. 225–239.
[81] O. Kaiwartya, A.H. Abdullah, Y. Cao, A. Altameem, M. Prasad, C-T. Lin, X. Liu,
Internet of Vehicles: motivation, layered architecture, network model, challenges, and future aspects, IEEE Access 4 (September) (2016) 5356–5373.
[82] H. Hada, J. Mitsugi, EPC based internet of things architecture, in: Proc. IEEE
International Conference on RFID-Technologies and Applications, September
2011, pp. 527–532.
[83] M. Yun, B. Yuxin, Research on the architecture and key technology of Internet
of Things (IoT) applied on smart grid, in: Proc. International Conference on
Advances in Energy Engineering, 2010, pp. 69–72.
[84] A.P. Castellani, N. Bui, P. Casari, M. Rossi, Z. Shelby, M. Zorzi, Architecture and
protocols for the Internet of Things: a case study, in: Proc. 8th IEEE International Conference on Pervasive Computing and Communications Workshops
(PERCOM Workshops), 2010, pp. 678–683.
[85] S. Hong, D. Kim, M. Ha, S. Bae, S.J. Park, W. Jung, J. Kim, SNAIL: an IP-based
wireless sensor network approach to the Internet of Things, IEEE Wirel. Commun. 17 (December (6)) (2010) 34–42.
[86] F. Andreini, F. Crisciani, C. Cicconetti, R. Mambrini, Context-aware location
in the Internet of Things, in: Proc. 2010 IEEE Globecom Workshops, 2010,
pp. 300–304.
[87] G. Pujolle, An autonomic-oriented architecture for the Internet of Things, in:
Proc. Modern Computing, 2006. IEEE JVA ’06, October 2006.
[88] S. Guoqiang, C. Yanming, Z. Chao, Z. Yanxu, Design and implementation of a
smart IoT gateway, in: Proc. IEEE International Conference on Green Computing and Communications and IEEE Internet of Things and IEEE Cyber, Physical
and Social Computing, 2013, pp. 720–723.
[89] I. Ishaq, J. Hoebeke, I. Moerman, P. Demeester, Internet of Things virtual networks: bringing network virtualization to resource-constrained devices, in:
Proc. IEEE International Conference on Green Computing and Communications, 2012, pp. 293–300.
[90] Z. Qin, G. Denker, C. Giannelli, P. Bellavista, N. Venkatasubramanian, A
software defined networking architecture for the Internet-of-Things, in:
Proc. IEEE Network Operations and Management Symposium (NOMS), 2014,
pp. 1–9.
[91] G.C. Fox, S. Kamburugamuve, R.D. Hartman, Architecture and measured characteristics of a cloud based Internet of Things, in: Proc. International Conference on Collaboration Technologies and Systems (CTS), 2012, pp. 6–12.
[92] A.P. Castellani, M. Dissegna, N. Bui, M. Zorzi, WebIoT: a web application
framework for the Internet of Things, in: Proc. IEEE Wireless Communications
and Networking Conference Workshops (WCNCW), 2012, pp. 202–207.
[93] Constrained RESTful Environments (Core), IETF Working Group, [Online]. Available: http://www.ietf.org/dyn/wg/charter/core-charter.html.
[94] B. Upadhyaya, Y. Zou, H. Xiao, J. Ng, A. Lau, Migration of SOAP-based services
to RESTful services, in: Proc. 13th IEEE International Symposium on Web Systems Evolution (WSE), 2011, pp. 105–114.
[95] D. Guinard, V. Trifa, E. Wilde, A resource oriented architecture for the web of
things, in: Proc. Internet of Things (IoT), December 2010, pp. 1–8.
[96] A.H.H. Ngu, M. Gutierrez, V. Metsis, S. Nepal, M.Z. Sheng, IoT middleware: a
survey on issues and enabling technologies, IEEE Internet Things J. PP (October (99)) (2016) 1–1.
[97] M.A. Razzaque, M. Milojevic-Jevric, A. Palade, S. Clarke, Middleware for Internet of Things: a survey, IEEE Internet Things J. 3 (February (1)) (2016) 70–95.
[98] I. Grønbæk, Architecture for the Internet of Things (IoT): API and interconnect, in: Proc. Second International Conference on Sensor Technologies and Applications (sensorcomm), 2008, pp. 802–807.
[99] B. Cheng, D. Zhu, S. Zhao, J. Chen, Situation-aware IoT service coordination
using the event-driven SOA paradigm, IEEE Trans. Netw. Serv. Manage. 13
(March (2)) (2016) 349–361.
[100] P. Spiess, S. Karnouskos, D. Guinard, D. Savio, O. Baecker, L.M. Sá de Souza,
V. Trifa, SOA-based integration of the Internet of Things in enterprise services,
in: Proc. IEEE International Conference on Web Services, 2009, pp. 968–975.
[101] Z.J. Muhsin, A. Al-Taee, M.A. Al-Taee, W. Al-Nuaimy, A. Al-Ataby, Mobile workflow management system based on the Internet of Things, in: Proc. 13th
International Multi-Conference on Systems, Signals & Devices (SSD), 2016,
pp. 436–441.
[102] F. Kawsar, G. Kortuem, B. Altakrouri, Supporting interaction with the Internet
of Things across objects, time and space, in: Proc. Internet of Things (IoT),
2010, pp. 1–8.
A. Colakovi ˇ ´c, M. Hadžiali´c / Computer Networks 144 (2018) 17–39 37
[103] A. Bhavana, Evaluating perception, characteristics and research directions for
Internet of Things (IoT): an investigational survey, Int. J. Comput. Appl. 121
(July (4)) (2015) 13–19.
[104] B. Soret, K.I. Pedersen, N.T.K. Jorgensen, V. Fernández-López, Interference coordination for dense wireless networks, IEEE Commun. Mag. 53 (January (1))
(2015) 102–109.
[105] 1905.1-2013, IEEE Standard for a Convergent Digital Home Network For Heterogeneous Technologies, IEEE Standard, 2013.
[106] S. Aguzzi, et al., Definition of a Research and Innovation Policy Leveraging
Cloud Computing and IoT Combination, 2014, pp. 64–67. A study prepared
for the European Commission–Final Report.
[107] C. Pereira, A. Aguiar, Towards efficient mobile M2M communications: survey
and open challenges, Sensors 14 (October (10)) (2014) 19582–19608.
[108] J. Santos, J. Rodrigues, B.M. Silva, J. Casal, K. Saleem, V. Denisov, An IoT-based
mobile gateway for intelligent personal assistants on mobile health environments, J. Netw. Comput. Appl. 71 (March) (2016) 194–204.
[109] G. Aloi, M. Di Felice, V. Loscri, P. Pace, G. Ruggeri, Spontaneous smartphone
networks as a user-centric solution for the future internet, IEEE Commun.
Mag. 52 (December (12)) (2014) 26–33.
[110] P. Vlacheas, et al., Enabling smart cities through a cognitive management
framework for the Internet of Things, IEEE Commun. Mag. 51 (June (6))
(2013) 102–111.
[111] M. Blackstock, R. Lea, IoT interoperability: a hub-based approach, in: Proc.
International Conference on the Internet of Things (IoT), 2014, pp. 79–84.
[112] L.E. Talavera, et al., The mobile hub concept: enabling applications for the
Internet of Mobile Things, in: Proc. IEEE International Conference on Pervasive Computing and Communication Workshops (PerCom Workshops), 2015,
pp. 123–128.
[113] I-R. Chen, J. Guo, F. Bao, Trust management for SOA-based IoT and its application to service composition, IEEE Trans. Serv. Comput. 9 (October (3)) (2016)
482–495.
[114] J. Kiljander, et al., Semantic interoperability architecture for pervasive computing and Internet of Things, IEEE Access 2 (August) (2014) 856–873.
[115] P. Barnaghi, W. Wang, C. Henson, K. Taylor, Semantics for the Internet of
Things: early progress and back to the future, Int. J. Semant. Web Inf. Syst.
8 (January (1)) (2012) 1–21.
[116] M. Ganzha, M. Paprzycki, W. Pawlowski, P. Szmeja, K. Wasielewska, Semantic
technologies for the IoT – an inter-IoT perspective, in: Proc. IEEE First International Conference on Internet-of-Things Design and Implementation (IoTDI),
2016, pp. 271–276.
[117] Y. Huang, G. Li, A semantic analysis for Internet of Things, in: Proc. International Conference on Intelligent Computation Technology and Automation, 1,
2010, pp. 336–339.
[118] Z. Song, A.A. Cardenas, R. Masuoka, Semantic middleware for the Internet of
Things, in: Proc. Internet of Things (IoT), 2010, pp. 1–8.
[119] Y. Wu, J. Li, J. Stankovic, K. Whitehouse, S. Son, K. Kapitanova, Run time assurance of application-level requirements in wireless sensor networks, in: Proc.
9th ACM/IEEE International Conference on Information Processing in Sensor
Networks, 2010, pp. 197–208.
[120] S-I. Choi, S-J. Koh, Use of proxy mobile IPv6 for mobility management in
CoAP-based Internet-of-Things networks, IEEE Commun. Lett. 20 (August (11))
(2016) 2284–2287.
[121] J. Huang, Q. Duan, Y. Zhao, Z. Zheng, W. Wang, Multicast routing for multimedia communications in the Internet of Things, IEEE Internet Things J. PP
(February (99)) (2017) 1–1.
[122] J. Ko, A. Terzis, S. Dawson-Haggerty, D.E. Culler, J.W. Hui, P. Levis, Connecting
low-power and lossy networks to the internet, IEEE Commun. Mag. 49 (April
(4)) (2011) 96–101.
[123] T. Clausen, U. Herberg, M. Philipp, A critical evaluation of the IPv6 routing
protocol for low power and lossy networks (RPL), in: Proc. IEEE 7th International Conference on Wireless and Mobile Computing, Networking and Communications (WiMob), 2011, pp. 365–372.
[124] S. Abdelwahab, B. Hamdaoui, M. Guizani, T. Znati, Cloud of things for sensing-as-a-service: architecture, algorithms, and use case, IEEE Internet Things
J. 3 (December (6)) (2016) 1099–1112.
[125] C. Perera, A. Zaslavsky, P. Christen, D. Georgakopoulos, Context aware computing for the Internet of Things: a survey, IEEE Commun. Surv. Tutorials 16
(1) (2014) 414–454 First Quarter.
[126] A. Mukherjee, H.S. Paul, S. Dey, A. Banerjee, ANGELS for distributed analytics
in IoT, in: In Proc. IEEE World Forum on Internet of Things (WF-IoT), 2014,
pp. 565–570.
[127] H. Madsen, B. Burtschy, G. Albeanu, Fl. Popentiu-Vladicescu, Reliability in the
utility computing era: towards reliable Fog computing, in: Proc. 20th International Conference on Systems, Signals and Image Processing (IWSSIP), 2013,
pp. 43–46.
[128] M. Chiang, T. Zhang, Fog and IoT: an overview of research opportunities, IEEE
Internet Things J. 3 (December (6)) (2016) 854–864.
[129] M. Roopaei, P. Rad, K-K.R. Choo, Cloud of Things in smart agriculture: intelligent irrigation monitoring by thermal imaging, IEEE Cloud Comput. 4 (March
(1)) (2017).
[130] M. Mosko, I. Solis, E. Uzun, C. Wood, CCNx 1.0 Protocol Architecture, A Xerox
Company, Computing Science Labaratory PARC, April 2017 [Online]. Available
http://www.ccnx.org/pubs/CCNxProtocolArchitecture.pdf.
[131] M.A. Rajan, P. Balamuralidhar, K.P. Chethan, M. Swarnahpriyaah, A self-
-reconfigurable sensor network management system for internet of things
paradigm, in: Proc. International Conference on Devices and Communications
(ICDeCom), 2011, pp. 1–5.
[132] M. Abu-Elkheir, M. Hayajneh, N.A. Ali, Data management for the Internet of Things: design primitives and solution, Sensors 13 (October) (2013)
15582–15612.
[133] C-W. Tsai, C-F. Lai, M-C. Chiang, L.T. Yang, Data mining for Internet of Things:
a survey, IEEE Commun. Surv. Tutorials 16 (1) (2014) 77–97 First Quarter.
[134] P. Rost, et al., Cloud technologies for flexible 5G radio access networks, IEEE
Commun. Mag., vol. 52, no. 5, pp. 68–76, Sep. 2014.
[135] H.H. Cho, C-F. Lai, T.K. Shih, H-C. Chao, Integration of SDR and SDN for 5G,
IEEE Access 2 (September) (2014) 1196–1204.
[136] M.A.S. Santos, B.A.A. Nunes, K. Obraczka, T. Turletti, B.T. de Oliveira,
C.B. Margi, Decentralizing SDN’s control plane, in: Proc. 39th Annual IEEE
Conference on Local Computer Networks, 2014, pp. 402–405.
[137] N. Bizanis, F.A. Kuipers, SDN and virtualization solutions for the Internet of
Things: a survey, IEEE Access 4 (September) (2016) 5591–5606.
[138] K. Sood, S. Yu, Y. Xiang, Software-defined wireless networking opportunities
and challenges for Internet-of-Things: a review, IEEE Internet Things J. 3 (August (4)) (2016).
[139] D. Bandyopadhyay, J. Sen, Internet of Things: applications and challenges in
technology and standardization, Wirel. Pers. Commun. 58 (May (1)) (2011)
49–69.
[140] M. Trnka, T. Cerny, Identity management of devices in Internet of Things environment, in: Proc. 6th International Conference on IT Convergence and Security (ICITCS), 2016, pp. 1–4.
[141] OMA LightweightM2M (LWM2M), OMA (Open Mobile Alliance) Specification, February 2017 [Online]. Available http://technical.openmobilealliance.
org/Technical/technical-information/release-program/current-releases/
oma-lightweightm2m-v1-0-2.
[142] V. Perelman, J. Schoenwaelder, M. Ersue, K. Watsen, Network configuration
protocol light (NETCONF Light), IETF Network Working Group, January 2012
[Online]. Available https://tools.ietf.org/html/draft-schoenw-netconf-light-01.
[143] Q. Wu, G. Ding, Z. Du, Y. Sun, M. Jo, A.V. Vasilakos, A cloud-based architecture for the internet of spectrum devices over future wireless networks, IEEE
Access 4 (June) (2016) 2854–286.
[144] M. Elkhodr, S. Shahrestani, H. Cheung, A smart home application based on
the Internet of Things management platform, in: Proc. IEEE International Conference on Data Science and Data Intensive Systems, 2015, pp. 491–496.
[145] T. Shah, A. Yavari, K. Mitra, S. Saguna, P.P. Jayaraman, F. Rabhi, R. Ranjan, Remote health care cyber-physical system: quality of service (QoS) challenges
and opportunities, IET Cyber Phys. Syst. 1 (January (1)) (2017) 40–48.
[146] A. Floris, L. Atzori, Quality of experience in the multimedia Internet of Things:
definition and practical use-cases, in: Proc. IEEE International Conference on
Communication Workshop (ICCW), 2015, pp. 1747–1752.
[147] Q. Wu, G. Ding, Y. Xu, S. Feng, Z. Du, J. Wang, K. Long, Cognitive Internet of
Things: a new paradigm beyond connection, IEEE Internet Things J. 1 (April
(2)) (2014) 129–143.
[148] A. Laya, L. Alonso, J. Alonso-Zarate, Is the random access channel of LTE and
LTE-a suitable for M2M communications? A survey of alternatives, IEEE Commun. Surv. Tutorials 16 (1) (2014) First Quarter.
[149] A. Lo, Y.W. Law, M. Jacobsson, A cellular-centric service architecture for machine-to-machine (M2M) communications, IEEE Wirel. Commun. 20 (October
(5)) (2013) 143–151.
[150] L. Li, S. Li, S. Zhao, QoS-aware scheduling of services-oriented Internet of
Things, IEEE Trans. Ind. Inf. 10 (May (2)) (2014) 1497–1505.
[151] F. Samie, V. Tsoutsouras, S. Xydis, L. Bauer, D. Soudris, J. Henkel, Distributed
QoS management for Internet of Things under resource constraints, in: Proc.
International Conference on Hardware/Software Codesign and System Synthesis (CODES+ISSS), 2016, pp. 1–10.
[152] C. Xiang, P. Yang, X. Wu, H. He, S. Xiao, QoS-based service selection with
lightweight description for large-scale service-oriented internet of things, Tsinghua Sci. Technol. 20 (August (4)) (2015) 336–347.
[153] M.B. Yassein, MohammedQ. Shatnawi, D. Al-zoubi, Application layer protocols
for the Internet of Things: a survey, in: Proc. International Conference Engineering & MIS (ICEMIS), 2016, pp. 1–4.
[154] Z. Sheng, S. Yang, Y. Yu, A.V. Vasilakos, J.A. Mccann, K.K. Leung, A survey on
the IETF protocol suite for the Internet of Things: standards, challenges, and
opportunities, IEEE Wirel. Commun. 20 (December (6)) (2013) 91–98.
[155] P. Masek, et al., Implementation of true IoT vision: survey on enabling protocols and hands-on experience, in: International Journal of Distributed Sensor
Networks, 12, 2016, pp. 1–18.
[156] W. Colitti, K. Steenhaut, N. De Caro, B. Buta, V. Dobrota, Evaluation of constrained application protocol for wireless sensor networks, in: Proc. 18th IEEE
Workshop LANMAN, 2011, pp. 1–6.
[157] N. De Caro, W. Colitti, K. Steenhaut, G. Mangino, G. Reali, Comparison of two
lightweight protocols for smartphone-based sensing, in: Proc. IEEE 20th SCVT,
2013, pp. 1–6.
[158] J.L. Fernandes, I.C. Lopes, J.J.P.C. Rodrigues, S. Ullah, Performance evaluation of
RESTful web services and AMQP protocol, in: Proc. Fifth International Conference on Ubiquitous and Future Networks (ICUFN), 2013, pp. 810–815.
[159] V. Pimentel, B.G. Nickerson, Communicating and Displaying Real-Time Data
with WebSocket, IEEE Internet Comput. 16 (May (4)) (2012) 45–53.
[160] A. Colakovi ˇ c´, T. Caršimamovi ˇ c´, The corresponding options of TCP variants for
fairness problem in AD HOC networks, Int. J. Soft Comput. Eng. 5 (May (2))
(2015) 1–8.
38 A. Colakovi ˇ ´c, M. Hadžiali´c / Computer Networks 144 (2018) 17–39
[161] R. Roman, P. Najera, J. Lopez, Securing the Internet of Things, Computer 44
(September (9)) (2011) 51–58.
[162] A.J. Jara, P. Martinez-Julia, A. Skarmeta, Light-weight multicast DNS and
DNS-SD (lmDNS-SD): IPv6-based resource and service discovery for the web
of things, in: Proc. Sixth International Conference on Innovative Mobile and
Internet Services in Ubiquitous Computing, 2012, pp. 731–738.
[163] S. Lanzisera, A.R. Weber, A. Liao, D. Pajak, A.K. Meier, Communicating power
supplies: bringing the internet to the ubiquitous energy gateways of electronic devices, IEEE Internet Things J. 1 (April (2)) (2014) 153–160.
[164] J. Qi, A.V. Vasilakos, J. Wan, J. Lu, D. Qiu, Security of the Internet of
Things: perspectives and challenges, Wirel. Netw. 20 (November (8)) (2014)
2481–2501.
[165] S. Babar, P. Mahalle, A. Stango, N. Prasad, R. Prasad, Proposed security model
and threat taxonomy for the Internet of Things (IoT), in: Proc. Communications in Computer and Information Science, 89, 2010, pp. 420–429.
[166] S. Kalra, S.K. Sood, Secure authentication scheme for IoT and cloud servers,
Pervasive Mob. Comput. 24 (December) (2015) 210–223.
[167] Internet of Things research study, Hewlett Packard Enterprise Report, 2015
[Online]. Available http://www8.hp.com/us/en/hp-news/press-release.html?
id=1909050#.WPoNH6KxWUk.
[168] L.D. Xu, W. He, S. Li, Internet of Things in industries: a survey, IEEE Trans.
Ind. Inf. 10 (November (4)) (2014) 2233–2243.
[169] Z. Yan, P. Zhang, A.V. Vasilakos, A survey on trust management for Internet
of Things, J. Netw. Comput. Appl. 42 (Jun. 2014) 120–134.
[170] K. Kang, Z. Pang, L.D. Xu, L. Ma, C. Wang, An interactive trust model for application market of the Internet of Things, IEEE Trans. Ind. Inf. 10 (May (2))
(2014) 1516–1526.
[171] D. Singh, G. Tripath, A. Jara, Secure layers based architecture for Internet of
Things, in: Proc. IEEE 2nd World Forum on Internet of Things (WF-IoT), 2015,
pp. 321–326.
[172] S.L. Keoh, S.S. Kumar, H. Tschofenig, Securing the Internet of Things: a standardization perspective, IEEE Internet Things J. 1 (June (3)) (2014) 265–275.
[173] M. Pei, N. Cook, M. Yoo, A. Atyeo, H. Tschofenig, The open trust
protocol (OTrP), IETF (2016). [Online]. Available https://tools.ietf.org/html/
draft-pei-opentrustprotocol-00.
[174] T. Dierks, C. Allen, The TLS protocol version 1.0, IETF RFC 2246 (1999). [Online]. Available https://www.ietf.org/rfc/rfc2246.txt.
[175] E. Rescorla, N. Modadugu, Datagram transport layer security, IETF RFC 4347
(2006). [Online] Availablehttps://tools.ietf.org/html/rfc4347.
[176] S. Vatari, A. Bakshi, T. Thakur, Green house by using IOT and cloud computing, in: Proc. IEEE International Conference on Recent Trends in Electronics,
Information & Communication Technology (RTEICT), 2016, pp. 246–250.
[177] F.K. Shaikh, E.Exposito S.Zeadally, Enabling technologies for green Internet of
Things, IEEE Syst. J. PP (April (99)) (2015) 1–12.
[178] S.K. Datta, C. Bonnet, N. Nikaein, An IoT gateway centric architecture to provide novel M2M services, in: Proc. 2014 IEEE World Forum on Internet of
Things (WF-IoT), 2014, pp. 514–519.
[179] D. Macedo, L.A. Guedes, I. Silva, A dependability evaluation for Internet of
Things incorporating redundancy aspects, in: Proc. Proceedings of the 11th
IEEE International Conference on Networking, Sensing and Control, 2014,
pp. 417–422.
[180] J. Yun, I-Y. Ahn, N-M. Sung, J. Kim, A device software platform for consumer
electronics based on the Internet of Things, IEEE Trans. Consum. Electron. 61
(November (4)) (2015) 564–571.
[181] H-L. Fu, P. Lin, H. Yue, G-M. Huang, C-P. Lee, Group mobility management for
large-scale machine-to-machine mobile networking, IEEE Trans. Veh. Technol.
63 (March (3)) (2014) 1296–1305.
[182] F. Bao, I-R. Chen, J. Guo, Scalable, adaptive and survivable trust management for community of interest based Internet of Things systems, in: Proc.
IEEE Eleventh International Symposium on Autonomous Decentralized Systems (ISADS), 2013, pp. 1–7.
[183] M. Masirap, et al., Evaluation of reliable UDP-based transport protocols for
Internet of Things (IoT), in: Proc. IEEE Symposium on Computer Applications
& Industrial Electronics (ISCAIE), 2016, pp. 200–205.
[184] R. Fantacci, T. Pecorella, R. Viti, C. Carlini, A network architecture solution
for efficient IOT WSN backhauling: challenges and opportunities, IEEE Wirel.
Commun. 21 (August (4)) (2014) 113–119.
[185] A. Taivalsaari, T. Mikkonen, A roadmap to the programmable world: software
challenges in the IoT Era, IEEE Software 34 (January (1)) (2017) 72–80.
[186] F. Al-Turjman, A. Radwan, S. Mumtaz, J. Rodriguez, Mobile traffic modelling for wireless multimedia sensor networks in IoT, Comput. Commun. 112
(November) (2017) 109–115.
[187] E. Borgia, The Internet of Things vision: Key features, applications and open
issues, Comput. Commun. 54 (December) (2014) 1–31.
[188] R.K. Lomotey, J. Pry, S. Sriramoju, Wearable IoT data stream traceability in a
distributed health information system, Pervasive Mob. Comput. 40 (September) (2017) 692–707.
[189] Y. Xu, V. Mahendran, W. Guo, S. Radhakrishnan, in: Fairness in fog networks: achieving fair throughput performance in MQTT-based IoTs, 2017,
pp. 191–196.
[190] A. Rizzardi, S. Sicari, D. Miorandi, A. Coen-Porisini, AUPS: an Open Source
AUthenticated Publish/Subscribe system for the Internet of Things, Inf. Syst.
62 (December) (2016) 29–41.
[191] M. Maier, M. Chowdhury, B.P. Rimal, D.P. Van, The tactile internet: vision, recent progress, and open challenges, IEEE Commun. Mag. 54 (May (5)) (2016)
138–145.
[192] S. Sicaria, A. Rizzardia, D. Miorandib, C. Cappielloc, A secure and quality-aware prototypical architecture for the Internet of Things, Inf. Syst. 58
(June) (2016) 43–55.
[193] D. Mazza, D. Tarchi, G.E. Corazza, A unified urban mobile cloud computing
offloading mechanism for smart cities, IEEE Commun. Mag. 55 (March (3))
(2017) 30–37.
[194] H. Lamaazi, N. Benamar, A. J.Jara, RPL-based networks in static and mobile
environment: a performance assessment analysis, J. King Saud Univ. Comput.
Inf. Sci. (April) (2017) 1–14.
[195] R. Hamilton, J. Iyengar, I. Swett, A. Wilk, QUIC: a UDP-based secure and reliable transport for HTTP/2 draft-tsvwg-quic-protocol-02,
Internet-Draft, January 2016, [Online]. Available: https://tools.ietf.org/html/
draft-tsvwg-quic-protocol-02.
[196] Z. Shelby, M. Huttunen, M. Saarnivala, J. Riihijärvi, O. Raivio, P. Mähönen,
nanoIP, Internet Draft, February 2005, [Online]. Available: https://tools.ietf.
org/html/draft-shelby-nanoip-00.
[197] Eddystone format, April 2018, [Online]. Available: https://developer.estimote.
com/eddystone/.
[198] IEEE Std 802.11ah-2016, WG802.11 – wireless LAN working group, 2016, [Online]. Available: https://standards.ieee.org/findstds/standard/802.11ah-2016.
html.
[199] Weightless Specification, November 2017, [Online]. Available: http://www.
weightless.org/about/weightless-specification.
[200] Standardization of NB-IOT, June 2016, [Online]. Available: http://www.3gpp.
org/news-events/3gpp-news/1785-nb_iot_complete.
[201] Progress on 3GPP IoT, February 2016, [Online]. Available: http://www.3gpp.
org/news-events/3gpp-news/1766- iot_progress.
[202] Release 14, February 2016, [Online]. Available: http://www.3gpp.org/
release-14.
[203] LoRaWANTM Specification, November 2015, https://lora-alliance.org/
about-lorawan.
[204] Sigfox Technology Overview, 2017, https://www.sigfox.com/en/
sigfox-iot-technology-overview.
[205] Random Phase Multiple Access technology, Ingenu Inc, 2016, https://www.
ingenu.com/technology/rpma/.
[206] Wi-Fi HaLow, Wi-Fi Alliance®, January 2016, https://www.wi-fi.org/
discover-wi-fi/wi-fi-halow.
[207] A. Munir, P. Kansakar, S.U. Khan, IFCIoT: integrated fog cloud IoT: a novel architectural paradigm for the future Internet of Things, IEEE Consum. Electron.
Mag. 6 (June (3)) (2017) 74–82.
[208] A. Brogi, A: Forti, QoS-aware deployment of IoT applications through the fog,
IEEE Internet Things J. PP (May (99)) (2017) 1–8.
[209] W. Li, I. Santos, F.C. Delicato, P.F. Pires, L. Pirmez, W. Wei, H. Song, A. Zoma

IoT Technology Disruptions: A Gartner Trend
Insight Report
Published: 15 June 2017 ID: G00331334
Analyst(s): Sanjit Ganguli, Ted Friedman
IoT’s rapidly evolving technologies have ushered in innovative disruptions at
a breakneck pace. This report highlights Gartner research that covers these
IoT technical disruptions in five key areas: security, artificial intelligence, data
and analytics, communications, and endpoint technology.
Opportunities and Challenges
■ The unique volume, variety and velocity characteristics of IoT data present scalability and
integration challenges to harness and gain business insights, and may require an architectural
realignment.
■ Communication and endpoint IoT technologies are rapidly evolving to support new and
disruptive IoT use cases, with numerous competing and fragmented standards and protocols.
■ The deployment of IoT technology introduces a multifaceted “attack surface” that must be
secured.
What You Need to Know
■ Advances in arti€cial intelligence algorithms, advanced analytics like deep learning and new
data management approaches enabled by abundant computational power can successfully
expedite the delivery of business outcomes promised by IoT.
■ IoT communications networks should be implemented to balance network ef€ciency and
compatibility, based on an understanding of evolving standards, protocols and implications to
existing enterprise networks.
■ IoT’s multifaceted attack surface can be secured with new and innovative approaches in IoT
security, including advanced asset discovery, identity management, authentication, visibility and
security analytics.
Insight From the Analyst
Technology Evolving to Turn IoT Hype Into Reality
Sanjit Ganguli, Research Director
Ted Friedman, Vice President, Distinguished Analyst
The Internet of Things (IoT) has seen years of growing interest among enterprises and service
providers with the promise of enabling new digital business initiatives and unlocking operational
ef€ciencies. Practically, much of this interest has been rooted in the discovery phase of IoT, where
CIOs and CTOs wondered, “What is IoT?” and “Why do I need to care?” Today, as demonstrated in
our published research and observed in our client interactions, this has transitioned to the
implementation phase, where executives are asking, “How do we get started with IoT?” (see the
Evidence section). This shift is largely fueled by the disruptions in IoT technology that have
occurred, enabling a feasible and achievable path to creating business value.
Theme
Advances in arti€cial intelligence, data and analytics, security, communications and endpoint
technology have been critical IoT enablers. Their applicability and state of the art are evolving
rapidly, and organizations planning and deploying IoT solutions need to know what is new and
emerging in these technology areas, and answer the questions:
■ How will these IoT technology disruptions impact my ability to drive new digital business and
become more operationally ef€cient?
■ How will these technology developments disrupt my current IoT initiatives?
■ Will these technology disruptions in IoT eventually disrupt traditional IT and other areas of the
business?
In this IoT Trend Insight report, we highlight the Gartner research that analyzes these very
disruptions. This research will highlight Cool Vendors that are innovative and game-changing,
Page 2 of 17 Gartner, Inc. | G00331334
market trends in evolving technology areas, best practices for enabling these technologies, and
vertical-industry-speci€c use cases.
This collection will be organized into the €ve disruptive technology areas, as shown in Figure 1:
■ Sensing: IoT Endpoints
■ Communicating: IoT Communications
■ Securing: IoT Security
■ Understanding: IoT Data and Analytics
■ Acting: IoT Arti€cial Intelligence
Figure 1. Five Areas of IoT Technology Disruption
Source: Gartner (June 2017)
Executive Overview
De€nition
The Internet of Things is de€ned as:
Gartner, Inc. | G00331334 Page 3 of 17
A network of dedicated physical objects (things) that contain
embedded technology to sense or interact with their internal
state or the external environment. This excludes generalpurpose devices such as smartphones, tablets and PCs.
The various technology disruptions discussed in this research fall into different layers of the IoT
Reference Model (see Figure 2). In the model, the layers de€ne what capabilities an IoT component,
function or process must possess, while the tiers de€ne where a component, function or process
operates in the IoT architecture. The interfaces de€ne how data and control ‚ow into, out of and
through the system.
Endpoint technology inhabits the Device Layer; communications technology comprises the
Communication Layer; data and analytics reside in the Information Layer; and arti€cial intelligence
exists on the Function Layer. Security technology, as it is pervasive in its nature, touches each of
these layers (see “Architect Your Internet of Things System by Using the Gartner IoT Reference
Model”). Understanding the interaction and interdependence of each technology area with the IoT
system as a whole will enable a better understanding of their impact.
Page 4 of 17 Gartner, Inc. | G00331334
Figure 2. Gartner IoT Reference Model at a Glance
Source: Gartner (June 2017)
Finally, it is important to not think of these technology disruptions as trends that will only impact IoT
initiatives. Instead, consider their impacts within the larger context of IT and digital business. New
IoT communications technologies will likely have a major in‚uence on how enterprise networks are
architected in the future, with enhancements in wireless networking and edge computing. Similarly,
data analytics and arti€cial intelligence (AI) technologies, which will enable the harvesting of
knowledge from the massive amounts of IoT data, can be applied to traditional enterprise IT or
business data to enable new digital business initiatives. Furthermore, the increasing miniaturization,
dynamism, mobility and modularity of traditional IT systems will, over time, resemble many of the
characteristics of IoT systems.
Gartner, Inc. | G00331334 Page 5 of 17
Given all of the promise of IoT technology, there remain signi€cant perceived barriers to IoT success
in the minds of CIOs and CTOs. While their perception of IoT has changed after their initial
investigations, much of their concern is based on the current de€ciencies of IoT systems. In fact,
Gartner’s latest Internet of Things Backbone Survey showed that security was cited as the top
barrier to IoT success (35% of respondents), with privacy concerns (25% of respondents) and
potential risks and liabilities (25% of respondents) all in the top €ve. Additionally, the complexity of
implementation and integration ranked third with 29% of respondents, and 21% felt that the
technology itself is immature (see “Survey Analysis: 2016 Internet of Things Backbone Survey” and
Figure 3). The ability to stay abreast of IoT’s technology disruptions and in‚uence IoT initiatives to
make best use of them will be paramount to ensuring success of future IoT projects and
overcoming these barriers.
Figure 3. Barriers to IoT Success
Source: Gartner (June 2017)
Research Highlights
Sensing IoT Endpoints
Endpoints of the Internet of Things will grow at a 32.9% compound annual growth rate (CAGR) from
2015 through 2020, reaching an installed base of 20.4 billion units. With this massive growth in the
number of endpoints come major technology disruptions in sensor, device, gateway and digital twin
technologies. This is coupled with the growth of edge computing technologies, where computing
and analytics move closer to the endpoint to enable more real-time use cases. Understanding these
technology disruptions can help minimize the complexity involved in selecting, deploying, managing
and operating these IoT endpoints. The research highlighted here discusses many of these
disruptions.
Page 6 of 17 Gartner, Inc. | G00331334
Related Research
“Cool Vendors in ‘Thingi€cation’, 2017” focuses on desktop development tools that enable CIOs to
create IoT prototypes and small-run projects within existing resources. Cool Vendors in this
research each provide unique technologies that can enable new business models with minimal
investment.
“Cool Vendors in the Internet of Things, 2017” provides recommendations for CIOs looking to
enable disruptive IoT solutions and business outcomes. These include supporting the business
unit’s IoT cost optimization or operational building projects by establishing a team to build an IoT
requirements checklist and impact assessment. Key elements of the checklist and assessment
include IoT architectural elements such as gateways, platform, security and integration into
enterprise solutions.
“Cool Vendors for Smart City Applications and Solutions, 2017” provides CIOs with advice on how
integrated smart data with cool software solutions containing multiple functionalities is vital for
smart building, multimodal mobility and collaboration.
“Cool Vendors in Enterprise Wearable and Immersive Technologies, 2017” explores how I&O leaders
struggle to build a complete platform to support wearable devices in the enterprise. It highlights
innovative tools that go beyond hardware and addresses issues key to a successful enterprise
wearable device strategy.
“Cool Vendors in Personal Devices, 2017” shows how new device categories like virtual personal
assistant (VPA)-enabled speakers are evolving, going beyond voice UIs and integrating multimodal
interactions (i.e., via cameras and displays) to create more contextualized and personalized
experiences.
“Market Guide for Edge Computing Solutions for Industrial IoT” demonstrates that the edge
computing solution vendor landscape is rapidly evolving. Most vendors in the IoT market have
recognized that edge computing is an integral part of an IoT solution. Not all data needs to be sent
to the cloud or core data center, as that is cost-prohibitive, bandwidth-intensive, has performance
implications or is not practical, as in the case of remote locations. Therefore, it is imperative to have
data aggregation and processing deployed at the source of data generation as an aid to rapid
decision making using real-time analytics.
“Market Trends: The Connected Home, 2017” highlights how 2017 promises to be an exciting year
for the connected home. We fully expect big-brand virtual personal assistant (VPA)-enabled
speakers, such as Amazon Echo and Google Home, and home ecosystems, such as Works with
Nest, Apple HomeKit and the growing number of Alexa-enabled devices, to breathe new life into the
connected home market, thanks to both the novelty and the network effect. But it is not just these
giants that are likely to proliferate in the market.
“Market Trends: The Need for Real-Time Insights Offers Market Opportunities in Edge Computing”
demonstrates how edge computing provides alternative design elements that can be used,
depending on the speci€c application requirements. In edge computing, cloudlike functionalities,
such as compute, storage and networking (including applications of virtualization and softwareGartner, Inc. | G00331334 Page 7 of 17
de€ned data center technologies), are pushed to the edge of the network — where IoT data is
generated by sensors and other IoT endpoints. Edge computing serves as the decentralized
extension of cloud at the edge of the campus networks, at the edge of the data center networks
and at the edge of the cloud.
“Emerging Technology Analysis: Energy Harvesting Enables Autonomous IoT Endpoints” shows that
IoT deployments face signi€cant and increasing battery management costs as networks scale up
and move to battery-operated endpoints for increased autonomy and lower installation costs.
Gartner forecasts between 2 billion and 2.5 billion battery-operated endpoints by 2020. However,
battery management issues are often underestimated, resulting in delayed IoT initiatives and
overbudget IoT costs.
“Architect Your Internet of Things System by Using the Gartner IoT Reference Model” describes the
Gartner IoT Reference Model and three-step process. The reference model provides a framework
that enables technical professionals to de€ne their system architecture. The three-step process
provides a methodology to guide technical professionals toward this goal. Organizations can use
the model and process irrespective of what technology they use, which vendors they select or what
business outcome they are trying to achieve.
“Cool Vendors in IoT Edge Computing, 2017” explores the growing need for real-time insights
closer to the point of IoT data generation. IT leaders responsible for IT/OT alignment face an
increasing urgency to deploy decentralized, scalable and secure computing technologies at the
edge of data center, cloud and campus networks.
“Use the Internet of Things in Smart Buildings to Achieve Work-Life Ambience” focuses on smart
buildings, which have gained the attention of real-estate and facilities management leaders under
pressure to reduce operational costs, meet sustainability goals and improve the employee
experience. CIOs can use this research to plan how to work with their colleagues to achieve all
three.
Communicating: IoT Communications
The IoT communications space is in the midst of a standards and protocol battle, with far-reaching
implications in the technology choices that IoT end users make. Readers should consider the
varying technologies across a wide range of coverage, including:
■ PAN — 802.11ad, Bluetooth 4, Near Field Communication (NFC)
■ LAN — 802.11n, ZigBee, Thread, Bluetooth 5
■ WAN — LoRa, Sigfox, narrowband IoT (NB-IoT), LTE-M
Selecting the correct network architecture to address the IoT application requirement requires an
understanding of the network’s ability to address performance, scalability, security, interoperability
and device connectivity. Research highlighted here discusses the bene€ts and weaknesses of these
communications technologies and their broader implications on networking architecture.
Page 8 of 17 Gartner, Inc. | G00331334
Related Research
“Market Trends: LoRa Offers Low-Risk, High-Reward LPWA Opportunity” describes how Sigfox,
Random Phase Multiple Access (RPMA), LoRa and NB-IoT are competing for market share, but
have different business model risks. This research highlights the options for technology strategic
planners at semiconductor companies who risk missing market opportunities by wasting investment
in inappropriate wireless technologies for IoT solutions.
“The Role of the Alarms and Noti€cations Platform in the Real-Time Health System” explores how
alarms and noti€cations (A&N) platforms acquire, €lter, manage and deliver clinical alarms and event
data from patient monitors and medical devices. As a key enabler of the real-time health system,
healthcare provider CIOs will need to familiarize themselves with the A&N platform value
proposition. As more systems and devices are engaged in the delivery of care, at least one A&N
vendor is positioning its platform as an IoT platform.
“2017 Strategic Roadmap for IoT Network Technology” highlights Gartner research that has
identi€ed four basic IoT network architecture domains, each with its own communication
technologies, as well as network architecture dependencies and opportunities. Selecting the wrong
network architecture for the desired outcome means reducing the network’s ability to address
performance, scalability, security or device connectivity to address the IoT application requirements.
Unfortunately, it typically also means throwing everything away and starting from the beginning for
Round 2 to provide network communication for the IoT project.
“The Top Three Impacts of IoT on Networks” highlights the regular failure of IT to identify and
remediate the IoT implementation’s impact on network elements, creating headaches for
networking leaders.
“Automaker CIOs Should Move Quickly to Implement Over-the-Air Update Capability” focuses on
over-the-air (OTA) capability as an absolute requirement for the automotive industry to begin using
new technology and to try out different business models. Despite the fast-spreading connectivity of
vehicles, automakers have been slow to add this capability, and have limited it mostly to noncritical
entertainment functions. The pace of introductions of the capability will increase.
“Emerging Technology Analysis: Time-Sensitive Networking” looks at time-sensitive networking
(TSN) as an IoT wireless and wired solution that can overcome the weaknesses of Wi-Fi and
Ethernet. Technology strategic planners focused on the IoT and enterprise connectivity must
develop an aggressive pricing strategy by limiting functionality premiums to entice organizations to
upgrade to TSN capabilities as part of normal switching refresh cycles.
“Innovation Insight: How CIOs Can Leverage the IoT to Break Down Building Management Silos” is
recommended for business leaders looking to boost their green credentials. According to Energy
Star, the average building wastes 30% of its energy through inef€ciencies. Much of this energy can
be conserved by using the IoT and IT infrastructure to enable communication between the different
building management systems.
“How to Realign ITOM Product Offerings for IoT Use Cases” explores how, while it may seem that
marketing and selling IT operations management (ITOM) products for IoT use cases may be an easy
Gartner, Inc. | G00331334 Page 9 of 17
and natural extension, the reality is that the vast majority of vendors don’t have the vertical-speci€c
capabilities or expertise to viably participate in the IoT market. Many of these vendors also lack an
understanding of whether the data collected can assist in solving the vertical market business
problem.
Securing: IoT Security
IoT security is cited as the top barrier to IoT success, as the explosion of varying types of IoT
endpoints creates an attack surface that has never before seen. IoT-security-related technology
disruptions have been broad in their scope and can be broken down into the following security
methods:
■ Asset discovery, pro€ling and tracking
■ Authentication
■ Network-based protection
■ Secure software development
■ Visibility through monitoring, detection and response
The integration of these IoT security technologies with traditional IT and operational technology (OT)
security infrastructures and practices remains a challenge, and is exacerbated by the speci€c
vertical and use-case nature of IoT systems. Research covered here highlights both technologyspeci€c and vertical-industry-speci€c trends in IoT security technology.
Related Research
“Cool Vendors in IoT Security, 2017” describes vendors that bring notable approaches to software
composition analysis, enterprise mobility management and asset discovery. Security and risk
management leaders should see these offerings as leading indicators of emerging priorities in IoT
utilization.
“Semiconductor Design, Key Management and In-Situ Updates Will Enable Secure IoT Solutions”
considers how the widespread deployment of IoT endpoints ampli€es the possibility for major
security risks. Technology product management leaders at semiconductor vendors that fail to
integrate security features and processes into their IoT-targeted products will lose design-ins and
suffer brand devaluation.
“Top 10 Strategic Technology Trends for 2017: Adaptive Security Architecture” drives home the fact
that security considerations must be factored in from the earliest stages of solution design. Existing
blocking and prevention capabilities are insuf€cient — comprehensive protection requires an
adaptive process to predict, prevent, detect and respond to security breaches.
“Market Trends: Grow Your IoT Security Business by Investing in Real-Time Discovery, Visibility and
Control” focuses on the chief information security of€cer’s (CISO’s) concern that security and risk
management leaders in the consumer and industrial IoT verticals don’t know what assets they have,
Page 10 of 17 Gartner, Inc. | G00331334
whether or not assets are connected to the internet, and whether protection is required. This is due
to a lack of network and device visibility; discovery is a prerequisite to IoT security.
“Healthcare Provider CIOs Need to Address IoT’s Security Risks Now” shows that the widely varied
data from almost every aspect of daily operation in a healthcare facility is a security risk. This data
includes speci€cs such as how the facility is performing physically (heating/cooling/lighting) and
patient condition — both within the hospital and at home — with many more use cases to be
discovered, all of which must be protected.
“Don’t Let Your IoT Projects Fail: Use the Right IoT Security Pattern to Protect Them” shows that
security and risk management leaders must recognize that IoT projects have security risks and that
they must do something about them today, including identifying the security controls and
technology needed for the IoT pattern used for the project.
Understanding: IoT Data and Analytics
Data from things and the insights derived from that data fuel the business value and transformative
nature of IoT. Many existing data and analytics capabilities can be applied to IoT initiatives and IoT
data, but organizations need to modernize in three key areas:
■ Styles of analytics (expanding into predictive)
■ Styles of integration (diversifying and adding real-time and virtualized)
■ New data persistence models (cloud and NoSQL)
Research highlighted here exposes critical technology developments in data and analytics that are
directly applicable to IoT.
Related Research
“Competitive Landscape of IoT Platform Vendors” highlights that there is no dominant provider of
IoT platforms. In fact, the market is extremely crowded with a very broad range of companies, with
a mix of startups, big industrial players, system integrators and traditional IT companies all
competing to provide solutions in the space. In addition, a signi€cant number of vendors are adding
IoT capabilities to their portfolios, making the landscape unclear.
“Align Data Integration and Data Quality to Strengthen IoT Solutions” looks at the rapid evolution
toward digital business, where IoT projects create signi€cant data management challenges for
integration and quality. A process for sharing and certifying IoT data requires technologies such as
data quality tools that must work alongside data integration activities.
“Cool Vendors in Internet of Things Analytics, 2017” pro€les innovative vendors in IoT analytics.
These vendors focus on some of the hottest areas of IoT — visibility into the manufacturing
process, enabling new analytics users, and device diagnostics, repair and maintenance — to help
data and analytics leaders increase the value of IoT projects.
Gartner, Inc. | G00331334 Page 11 of 17
“Use Master Data Management Principles for Identity Management in the Internet of Things”
considers that complex digital business initiatives include IoT devices which can involve hundreds
of thousands or even millions of “entities” — digital identi€ers of people, devices, applications or
subsets of those entities. The means to identify, label and manage those entities can already be
found in multiple management systems and technologies.
“A Guide to Deploying IoT Analytics, From Edge to Enterprise” shows how, for successful digital
transformation, technical professionals must establish new ways to leverage and monetize IoT data
by embedding analytics throughout the IoT architecture. This report offers guidance on deploying
analytics across IoT solutions, from edge to enterprise.
“Innovation Insight for Digital Twins — Driving Better IoT-Fueled Decisions” considers how
exploiting the digital-twin concept is becoming an important innovation that enables stakeholders to
monitor and make informed decisions about the state of the actual physical things, their context
and the required action needed to optimize their future state.
“Market Guide for Energy Management Systems, IoT” shows that as IoT-based systems emerge,
energy management systems (EMSs) have evolved into platforms that monitor and manage all
energy use in a building, including the HVAC system, lights, major equipment, renewable energy and
plug load. Consequently, EMSs not only enable energy management, but also create an
environment of energy economics, sustainability and operational ef€ciency.
“The Impact of Event-Driven IT on API Management” explains how digital business trends —
including the IoT, real-time decision making and microservice architecture — are driving application
leaders to focus on event-driven IT.
“Embrace Your Bias to Enable Analytics Clarity” shows that bias is inherent in the development of
analytic models, data selection and the associated algorithms. The bias continuum provides a
discussion template to data and analytics leaders for transparently exposing statistical bias to
ensure real business impacts.
“Harness Streaming Data for Real-Time Analytics” warns analytics leaders that stream processing
compute platforms are the core of serving data for real-time analytics, especially for the IoT
systems, yet mastering these platforms requires a mental shift toward treating data as dynamic, not
static.
Acting: IoT Arti€cial Intelligence
AI will be used to render new insights, transform decision making and drive improved business
outcomes. The complexity, speed and distribution of IoT solutions and the data they generate may
obviate traditional decision-making techniques, both manual and automated. Advances in AI, in the
form of new algorithms, increasing computational power and breakthroughs in deep learning, will
enable transformative IoT solutions that would otherwise be impossible. This research highlights key
trends in AI and critical connections to the IoT.
Page 12 of 17 Gartner, Inc. | G00331334
Related Research
“Develop Your Arti€cial Intelligence Strategy Expecting These Three Trends to Shape Its Future”
looks at the three major trends to affect AI during the coming years, including better
communication, deeper integration and richer ecosystems.
“AI on the Edge: Fusing Arti€cial Intelligence and IoT Will Catalyze New Digital Value Creation ”
describes how AI and the IoT are symbiotic technologies that will be the foundation of a new
platform for digital business value creation. CIOs engaged in IoT initiatives should leverage these
capabilities for strategic advantage.
“Market Guide for Conversational Arti€cial Intelligence in China” explores how the technology is
progressing in China with a wide range of solutions, as the market remains immature and
challenging.
“Market Trends: How AI and Affective Computing Deliver More Personalized Interactions With
Devices” shows how current platforms for detecting and responding to emotions are mainly
proprietary and specialized in a few isolated use cases. We expect Google, Apple, Facebook and
Amazon to disrupt the emotion-sensing digital device market by offering tools that will enable
affective computing for broader use cases.
“Market Guide for Augmented Reality” highlights how augmented reality solutions are poised for
rapid growth, empowered by the IoT, digital business and next-generation smartphones. Enterprise
architecture and technology innovation leaders must carefully evaluate providers as the market
consolidates around platforms.
“Preparing and Architecting for Machine Learning” explains the business value machine learning
(ML) provides, the basics of the architecture, process and skills needed for ML, and what steps
should be taken to get started in ML.
“Top 10 Strategic Technology Trends for 2017: Arti€cial Intelligence and Advanced Machine
Learning” focuses on the key drivers for IoT and machine learning that will impact all businesses.
“Prepare for Big Changes in Software and SaaS Pricing, Driven by AI and IoT” considers the
upheaval in negotiations and software evaluation from changing service types.
Gartner, Inc. | G00331334 Page 13 of 17
Related Priorities
Table 1. Related Priorities
Priority Focus
Succeeding With
Semiconductor-Based
Technology
This initiative enables technology providers to improve their competitiveness by
using products and services out of the semiconductor and electronics industry,
and investing in emerging technologies.
Delivering Effective Identity and
Access Management
Capabilities
The delivery of effective IAM capabilities involves tools and best practices that
manage identity, privileges, access and trust to facilitate security, risk management
and business imperatives.
Building and Expanding a Digital
Business
Digital business is the creation of new business designs by blurring the digital and
physical worlds. Digital business involves the interaction of people, businesses and
intelligent “things.”
Supply Chain Strategy,
Leadership and Governance
Designing strategy, optimizing networks, developing the organization and
managing performance must work interdependently to execute an ef€cient
demand-driven supply chain.
Source: Gartner
Gartner Analysts Supporting This Trend
Mark Hung
Alfonso Velosa
Benoit Lheureux
Earl Perkins
Saniye Burcu Alaybeyi
Eric Goodness
Tim Zimmerman
Svetlana Sicular
Whit Andrews
Related Resources
Webinars
“Cool Vendors for 2017: The Digital Nail Gets Hammered, So Be the Hammer”
Page 14 of 17 Gartner, Inc. | G00331334
“The IoT Roadmap for the Digital Business”
“Rapidly Architect Your IoT System With the IoT Reference Model”
“Digital Twins: The Future of Better IoT Fueled Business Decisions”
“The Gartner Top 10 Strategic Technology Trends for 2017”
“Preparing and Architecting for Machine Learning”
Articles
“Let Machine Learning Boost Your Business Intelligence”
“Navigating the Security Landscape in the IoT Era”
“Gartner Reveals 2017 Cool Vendors That Can Help Keep Pace With Digital Innovation”
“How to Address Threats in Today’s Security Landscape”
“Digital Has Changed the Security Landscape”
“5 Steps to Address IoT Integration Challenges”
Gartner Recommended Reading
Some documents may not be available as part of your current Gartner subscription.
“Internet of Things Primer for 2017”
“IoT Communications Architecture Demysti€ed”
“Hype Cycle for the Internet of Things, 2016”
“IoT’s Challenges and Opportunities in 2017: A Gartner Trend Insight Report”
“Hype Cycle for IoT Standards and Protocols, 2016”
“Ready-for-Development Semiconductor Solutions Massively Reduce Drone Market Entry Barriers”
Evidence
The analysis and advice provided in this document are built from constant scanning of the market,
as well as from the aggregation of analysts’ experience and ongoing interactions with end users and
technology and service providers. We used a range of sources to feed our perspective on the topics
discussed in this document:
Gartner, Inc. | G00331334 Page 15 of 17
■ Gartner customer inquiry and conversations, which has shown an average of 40% year-overyear increase in inquiry volume from IT leaders
■ Discussions between Gartner analysts with expertise in key technologies or relevant vertical
markets
■ Previous Gartner analysis of digital business, IoT and related technologies
Gartner analysts also leverage secondary sources of information, including government agencies,
standards organizations and so forth.
Page 16 of 17 Gartner, Inc. | G00331334
GARTNER HEADQUARTERS
Corporate Headquarters
56 Top Gallant Road
Stamford, CT 06902-7700
USA
+1 203 964 0096
Regional Headquarters
AUSTRALIA
BRAZIL
JAPAN
UNITED KINGDOM
For a complete list of worldwide locations,
visit http://www.gartner.com/technology/about.jsp
© 2017 Gartner, Inc. and/or its af€liates. All rights reserved. Gartner is a registered trademark of Gartner, Inc. and its af€liates. This
publication may not be reproduced or distributed in any form without Gartner’s prior written permission. It consists of the opinions of
Gartner’s research organization, which should not be construed as statements of fact. While the information contained in this publication
has been obtained from sources believed to be reliable, Gartner disclaims all warranties as to the accuracy, completeness or adequacy of
such information. Although Gartner research may address legal and €nancial issues, Gartner does not provide legal or investment advice
and its research should not be construed or used as such. Your access and use of this publication are governed by Gartner Usage Policy.
Gartner prides itself on its reputation for independence and objectivity. Its research is produced independently by its research
organization without input or in‚uence from any third party. For further information, see “Guiding Principles on Independence and
Objectivity.”
Gartner, Inc. | G00331334 Page 17 of 17

The post Discussion Board Topic 1 appeared first on Versed Writers.