1

Internet of Things-aided Smart Grid: Technologies,

Architectures, Applications, Prototypes, and Future
Research Directions
Yasir Saleem, Student Member, IEEE, Noel Crespi, Senior Member, IEEE, Mubashir Husain Rehmani, Senior
Member, IEEE, and Rebecca Copeland

Abstract—Traditional power grids are being transformed into monitored to ensure that the power plants generate electrical
Smart Grids (SGs) to solve the problems of uni-directional energy in accordance with the needs of the consumers within
information flow, energy wastage, growing energy demand, relia- the constraints of power systems. Nearly, all the generation,
bility and security. SGs offer bi-directional energy flow between
service providers and consumers, involving power generation, transmission and distribution of electrical energy is owned
transmission, distribution and utilization systems. SGs employ by the utility companies who provide electrical energy to
various devices for the monitoring, analysis and control of consumers and bill them accordingly to recover their costs
the grid, deployed at power plants, distribution centers and and earn profit.
in consumers’ premises in a very large number. Hence, an The traditional power grid worked very well from its
SG requires connectivity, automation and the tracking of such
devices. This is achieved with the help of Internet of Things inception in 1870 until 1970 [1]. Even though the consumers’
(IoT). IoT helps SG systems to support various network func- demand for energy grew exponentially, it was still rather
tions throughout the generation, transmission, distribution and predictable. However, there has been a dramatic change in
consumption of energy by incorporating IoT devices (such as the nature of electrical energy consumption since 1970, as
sensors, actuators and smart meters), as well as by providing the load of electronic devices has become the fastest growing
the connectivity, automation and tracking for such devices. In
this paper, we provide the first comprehensive survey on IoT- element of the total electricity demand and new sources of high
aided SG systems, which includes the existing architectures, electricity consumption have been developed, such as electric
applications and prototypes of IoT-aided SG systems. This survey vehicles (EVs). The power grids endure a significant wastage
also highlights the open issues, challenges and future research of energy due to a number of factors, such as consumers’
directions for IoT-aided SG systems. inefficient appliances and lack of smart technology, ineffi-
Index Terms—Internet of things (IoT), Smart grid (SG), Home cient routing and dispensation of electrical energy, unreliable
Area Network (HAN), Neighborhood Area Network (NAN), Wide communication and monitoring, and most importantly, lack
Area Network (WAN) of a mechanism to store the generated electrical energy [2]–
[4]. Furthermore, power grids face some other challenges as
I. I NTRODUCTION well, including growing energy demand, reliability, security,
emerging renewable energy sources and aging infrastructure
A. Motivation problems to name a few.
A traditional power grid consists of a large number of In order to solve these challenges, the Smart Grid (SG)
loosely interconnected synchronous Alternate Current (AC) paradigm has appeared as a promising solution with a va-
grids. It performs three main functions: generation, transmis- riety of information and communication technologies. Such
sion and distribution of electrical energy [1], in which electric technologies can improve the effectiveness, efficiency, relia-
power flows only in one direction, i.e., from a service provider bility, security, sustainability, stability and scalability of the
to the consumers. Firstly in power generation, a number of traditional power grid [5]. SG solves the problem of electrical
large power plants generate electrical energy, mostly from energy wastage by generating electrical energy which closely
burning carbon and uranium based fuels. Secondly in power matches the demand [2]–[4]. SG helps to make important
transmission, the electricity is transmitted from power plants decisions according to the demand of energy, such as real-
to remote load centers through high voltage transmission time pricing, self healing, power consumption scheduling and
lines. Thirdly in power distribution, the electrical distribution optimized electrical energy usage. Such decisions can signifi-
systems distribute electrical energy to the end consumers cantly improve the power quality as well as the efficiency of
at reduced voltage. Each grid is centrally controlled and the grid by maintaining a balance between power generation
and its usage [6].
Y. Saleem, N. Crespi and R. Copeland are with the Department SG differs from traditional power grids in many aspects.
of Wireless Networks and Multimedia Services, Telecom SudParis,
Institut Mines-Telecom, 91011 Evry, France (e-mail: yasir- For instance, SG offers a bi-directional communication flow
saleem shaikh@telecom-sudparis.eu; noel.crespi@it-sudparis.eu; between service providers and consumers, while a traditional
rebecca.copeland@coreviewpoint.com). power grid only offers only uni-directional communication
M.H. Rehmani is with the Department of Electrical Engineering, COM-
SATS Institute of Information Technology, Wah Cantt, Pakistan (e-mail: from the service provider to the consumer. SG provides
mshrehmani@gmail.com). supervisory control and data acquisition (SCADA), advanced
 2

metering infrastructure (AMI), smart meters, fault tolerance, • A discussion on the current prototypes for IoT-aided SG
unauthorize usage detection, and load balancing [6]–[8], as systems;
well as self-healing, i.e., detection and recovery from faults • A discussion on big data management in IoT-aided SG
[9]. systems;
SG deploys various types of devices for monitoring, an- • An overview of IoT and non-IoT communication tech-
alyzing and controlling the grid. Such monitoring devices nologies for SG systems; and
are deployed at power plants, transmission lines, transmission • A presentation of the open issues, challenges and future
towers, distribution centers and consumers premises, and their research directions of IoT-aided SG systems.
numbers amount to the hundreds of millions or even billions Table I presents the list of acronyms used in this paper.
[1]. One of the main concerns for SG is the connectivity,
automation and tracking of such large number of devices, TABLE I
L IST OF ACRONYMS AND CORRESPONDING DEFINITIONS .
which requires distributed monitoring, analysis and control
through high speed, ubiquitous and two-way digital commu- Acronyms Definitions
nications. It requires distributed automation of SG for such ADR Adaptive Data Rate
AMI Advanced Metering Infrastructure
devices or “things”. This is already being realized in the real ATM Asynchronous Transfer Mode
world through the Internet of Things (IoT) technology. CGI Common Gateway Interface
DER Distributed Energy Resource
Initially, the Internet served as a connectivity of people-to- DSL Digital Subscriber Lines
people and people-to-things. However by 2008, the number EPON Ethernet Passive Optical Network
of things connected to the Internet surpassed the number of FAN Field Area Network
GPS Global Positioning System
people in the world, thus the impact of IoT technology keeps GSM Global System for Mobile
rising. IoT is a network of physical objects or things connected GUI Graphical User Interface
HAN Home Area Network
to the Internet. Such objects are equipped with embedded IaaS Infrastructure as a Service
technology to interact with their internal and external envi- IoT Internet of Things
IPv6 Internet Protocol version 6
ronments. These objects sense, analyze, control and decide ISM Industrial, Scientific and Medical
individually or in collaboration with other objects through high LTE Long Term Evolution
speed and two-way digital communications in a distributed, MAC Medium Access Control
M2M Machine-to-Machine
autonomous and ubiquitous manner. This is exactly what is MBWA Mobile Broadband Wireless Access
required for the SG. Hence, IoT technology can help SGs by NAN Neighborhood Area Network
NAT Network Address Translation
supporting various network functions throughout the power NIST National Institute for Standards and Technology
generation, storage, transmission, distribution and consump- NMEA National Marine Electronics Association
LWM2M Lightweight Machine-to-Machine
tion by incorporating IoT devices (such as sensors, actuators OPGW Optical Power Ground Wire
and smart meters), as well as by providing connectivity, OPLC Optical Power Line Communications
automation and tracking for such devices [10]. PaaS Platform as a Service
PLC Power Line Communications
SG is considered as one of the largest applications of the PKI Public Key Infrastructure
IoT [11]. Today, the number of things that use electricity is PSTN Public Switched Telephone Network
RF Radio Frequency
higher than the number of things connected to the Internet, and RFID Radio Frequency Identification
essentially everything which uses electricity could be made SaaS Software as a Service
SCADA Supervisory Control and Data Acquisition
more useful by connecting it to the Internet. Hence in the SG Smart Grid
future, the IoT would be larger than the SG today, and the SIM Subscriber Identity Module
modern and intelligent grid (i.e., SG) will not be possible WPAN Wireless Personal Area Network
UPnP Universal Plug and Play
without the IoT technology. By making the IoT technology WiFi Wireless Fidelity
as a global standard for the communications and the basis for WiMAX Worldwide inter-operability for Microwave Access
WMAN Wireless Metropolitan Area Network
SG, new doors will be opened for maximizing the prospects VoLTE Voice over Long Term Evolution
for future innovations. VM Virtual Machine
UAV Unmanned Aerial Vehicles
UWB Ultra-Wideband
B. Contributions of this Survey Article WAN Wide Area Network
To the best of our knowledge, there is no survey that covers WLAN Wireless Local Area Networks
WWW World Wide Web
IoT-aided SG systems. In this survey, we comprehensively
cover architectures, applications and prototypes, as well as
the open issues, challenges and future research directions for
IoT-aided SG systems. The contributions of this survey are C. Comparison with Existing Survey Articles
summarized as follows: While there have been a large number of separate surveys
• The first survey on the intersection of IoT and SG on the IoT and the SG, to the best of our knowledge, there
systems, i.e., IoT-aided SG systems; is no existing survey that covers the intersection of IoT and
• A detailed discussion on existing architectures for IoT- SG systems. This survey differs from previous individual
aided SG systems; surveys on IoT and SG systems because it combines IoT and
• A detailed discussion on the existing and potential appli- SG systems together and it specifically covers IoT-aided SG
cations of IoT-aided SG systems; systems.
 3

There is a number of surveys on the IoT that have focused communication among smart devices through various commu-
on the overview, principles, vision, applications, challenges nication technologies, such as ZigBee, WiFi, Bluetooth and
and architectures of the IoT [12]–[15]. Many surveys have also cellular communications. and connects them to the Internet.
focused on security [16]–[18] and cloud computing for the IoT The semantic oriented concept realizes a variety of applica-
[19]. Moreover, social IoT has become a very hot topic of IoT tions with the help of smart devices.
[20], [21]. Other groups of surveys on the IoT have focused Over the past few years, the IoT technology has gained
on software defined networking (SDN) [22], data mining [23], significant attention in various applications, and has allowed
LTE uplink scheduling [24] and context aware computing [25]. for the interconnection of the Internet to various network-
The standardization of the IoT is also surveyed in [26], [27]. embedded devices used in daily life [46]. It has automated
Another survey [28] has focused on experimentation with the the operation of various systems, such as health care, trans-
IoT. portation, military, home appliances, security, surveillance,
Surveys focused on the SG have presented the general agriculture and power grids. In some areas, IoT devices are
overviews of the SG concept [4], including challenges and equipped with transceivers, micro controllers and protocols,
standardization activities [29], [30], as well as applications enabling their communication with other devices as well as
and architectures [5], [31]–[33]. Moreover, information man- with external entities to allow the realization of completely
agement schemes for SG have been surveyed in [34]. Security automated systems [47].
issues, which are important factors to be considered in the
SG, are surveyed in [35], [36], and privacy-aware metering is B. Smart Grid
reviewed in [37]. Other areas which have been surveyed in the
The SG has been promoted as a promising solution for
domain of SG include energy efficiency [38], power demand
minimizing the wastage of electrical energy and as a means
forecasting [39], load balancing [40], optimization and pricing
to solve the problems of traditional power grids, making pos-
methods [41], and renewable energy [42]. Additionally, some
sible advances in efficiency, effectiveness, reliability, security,
surveys focus on solving the problem of spectrum scarcity in
stability, and increasing demand of electrical energy [48]. The
SG through Cognitive Radio (CR) technology [4], [7], [10],
main SG attributes are that it offers self-healing, improved
[43]. A survey on cloud-assisted IoT-based SCADA systems
electricity quality, distributed generation and demand response,
security has very recently been published in [44].
mutual operation and user participation, and effective asset
management.
D. Article Organization As presented in Figure 2, the SG completely revolutionizes
The organization of the paper is presented in Figure 1 and the energy generation, transmission, distribution and consump-
is as follows: In Section II, we provide an overview of IoT tion in four sub-systems. It is comprised of three types of
technology and SG systems, as well as the integration of IoT networks, a Home Area Network (HAN), a Neighborhood
technology into SG systems. In Section III, we present the Area Network (NAN) and a Wide Area Network (WAN).
current applications of IoT-aided SG systems, followed by a HAN is the first layer; it manages the consumers’ on-demand
detailed analysis of the existing architectures of IoT-aided SG power requirements and consists of smart devices, home
systems in Section IV. In Section V, we present prototypes appliances (including washing machines, televisions, air con-
and experimentations for IoT-aided SG systems. We discuss ditioners, refrigerators and ovens), electrical vehicles, as well
big data management in IoT-aided SG systems in Section as renewable energy sources (such as solar panels). HAN is
VI. Section VII presents an overview of IoT and non-IoT deployed within residential units, in industrial plants and in
communication technologies for SG systems. In Section VIII, commercial buildings and connects electrical appliances with
we highlight the open issues, challenges and future research smart meters. NAN, also known as Field Area Network (FAN),
directions for IoT-aided SG systems. Finally, we conclude in is the second layer of an SG and consists of smart meters
Section IX. belonging to multiple HANs. NAN supports communication
between distribution substations and field electrical devices for
power distribution systems. It collects the service and metering
II. I NTERNET OF T HINGS AND S MART G RID : A N information from multiple HANs and transmits it to the data
OVERVIEW AND I NTEGRATION collectors which connect NANs to a WAN. WAN is the third
A. Internet of Things layer of an SG and it serves as a backbone for communication
between network gateways or aggregation points. It facilitates
The IoT is defined as a network that can connect any
the communication among power transmission systems, bulk
object with the Internet based on a protocol for exchanging
generation systems, renewable energy sources and control
information and communication among various smart devices
centers [2].
in order to achieve monitoring, tracking, management and
location identification objectives [45]. The IoT focuses on the
realization of three main concepts, namely things-oriented, C. Integration of the IoT into a SG
Internet-oriented and semantic-oriented. The things oriented The SG has already achieved wide adoption in information
concept involves smart devices, such as RFID tags, sensors, sensing, transmission and processing, and now IoT technology
actuators, cameras, laser scanners, the Global Positioning plays a significant role in grid construction. IoT technology
System (GPS) and NFC. The Internet oriented concept enables provides interactive real-time network connection to the users
 4

Fig. 1. Article organization.

and devices through various communication technologies, SG through application interfaces.

power equipment through various IoT smart devices, and the IoT sensing devices are generally comprised of wireless
cooperation required to realize real-time, two-way and high- sensors, RFIDs, M2M (machine-to-machine) devices, cam-
speed data sharing across various applications, enhancing the eras, infrared sensors, laser scanners, GPSs and various data
overall efficiency of a SG [49]. The application of the IoT in collection devices. The information sensing in an SG can
SGs can be classified into three types. Firstly, IoT is applied be highly supported and improved by IoT technology. The
for deploying various IoT smart devices for the monitoring IoT technology also plays an essential role in the infras-
of equipment states. Secondly, IoT is applied for information tructure deployment of data sensing and transmission for
collection from equipment with the help of its connected IoT the SG, assisting in network construction, operation, safety
smart devices through various communication technologies management, maintenance, security monitoring, information
(see Section VII). Thirdly, IoT is applied for controlling the collection, measurement, user interaction etc. Moreover, the
 5

Fig. 2. Smart grid (SG) architecture presenting power systems, power flow and information flow. The SG is comprised of four main subsystems (power
generation, transmission, distribution and utilization) and three types of networks (a wide area network (WAN), a neighborhood area network (NAN) and a
home area network (HAN). The power flows through the subsystems while the information flows through the networks.

IoT also enables the integration of information flow, power aided SG systems.
flow and distribution flow in a SG [50], [51]. Additionally, In the rest of this section, we describe the suitability of
existing SG architectures mainly focus on the needs of power IoT technology and preferred communication technologies for
distributors to manage the complete power grid [52]. The various functions of the three SG layers, i.e., for HAN, NAN
consumers are accessed with a network of smart meters by and WAN.
means of General Packet Radio Service (GPRS) or other mo- 1) Home Area Networks (HANs): HANs may have either
bile networks. The new reality where consumers may already a star topology or a mesh topology. The preferred communi-
have other smart home infrastructures (such as WiFi) has cation technologies for HANs are powerline communications
not yet been incorporated in the network communications of (wired technology), ZigBee, Bluetooth and WiFi (wireless
existing SG architectures [53], [54]. While some architectures technologies). A HAN is comprised of a variety of IoT smart
do consider existing smart home infrastructures, they are not devices, such as a home gateway, smart meters, sensor and
designed for scalability in large deployments [55], [56]. The actuator nodes, smart home appliances and electric vehicles.
protocols specific to IoT and SG systems therefore cannot A home gateway connects to smart meters and periodically
be directly applied to IoT-aided SG systems, as they only collects power consumption data of the home appliances.
consider the individual characteristics of either the IoT or the HANs perform two-fold functions, commissioning and con-
SG systems, which is not sufficient for an integrated IoT-aided trol. The commissioning function identifies new devices and
SG system. manages the devices. The control function enables commu-
A SG is comprised of four main subsystems, power genera- nication among smart devices by establishing the links and
tion, power transmission, power distribution and power utiliza- performs reliable operation for the various SG layers. A HAN
tion. IoT can be applied to all these subsystems and appears uses two-way communications for demand response manage-
as a promising solution for enhancing them, making the IoT ment services [60]–[62]. In the forward communication direc-
a key element for SG. In the area of power generation, the tion, the smart meters’ load and real-time power consumption
IoT can be used for the monitoring and controlling of energy information of the home equipment, connected to IoT smart
consumption, units, equipment, gas emissions and pollutants devices, are collected by home gateways and transmitted from
discharge, power use/production prediction, energy storage the consumer side (the HAN) to the NAN to be forwarded to
and power connection, as well as for managing distributed a utility center. In the backward communication direction, the
power plans, pumped storage, wind power, biomass power and home gateway acts as a central node and receives dynamic
photo-voltaic power plants [57]–[59]. In the area of power electricity pricing information from the NAN, which is then
transmission, the IoT can be used for the monitoring and provided to smart meters or IoT smart devices for triggering
control of transmission lines and substations, as well as for the required action for home appliances.
transmission tower protection [57], [58]. In the area of power 2) Neighbor Area Networks (NANs): The communication
distribution, IoT can be used for distributed automation, as technologies for NANs need to cover a radius of a thousand
well as in the management of operations and equipment. In meters. The communication channels between smart meters
the area of power utilization, the IoT can be used for smart and data aggregation points therefore must be interference
homes, automatic meter reading, electric vehicle charging and free [63]. A gateway in the NAN collects consumers’ energy
discharging, for collecting information about home appliances’ consumption data from smart meters in HANs and transmits
energy consumption, power load controlling, energy efficiency the collected data to the utility companies through either
monitoring and management, power demand management and private or public WANs. Basically, the topology of a NAN
multi network consumption [57], [58]. Figure 3 presents the is comprised of two types of gateways, NAN gateways and
existing (see Section III) and potential applications of IoT- HAN gateways. A NAN gateway connects various HAN
 6

Fig. 3. Existing and potential applications of IoT-aided SG systems classified into WAN, NAN and HAN. These are further classified into subsystems, i.e.,
power generation, transmission, distribution and utilization. The blue (shaded) boxes represent existing applications while the white (empty) boxes represent
potential applications. The existing applications are discussed in Section III.

gateways and serves as an access point to provide a single is another standard for IoT which is gaining popularity due to
hop connection to HAN gateways in a hybrid access manner. its greater simplicity. Additionally, the Internet Engineering
The HAN gateways transmit their energy consumption data Task Force (IETF) has developed a set of protocols and
to NAN gateways through either wired (e.g., PLC, DSL) or standards for connecting the IoT devices to the Internet [27].
wireless (e.g., cellular, mobile broadband wireless access or For instance, the Routing Protocol for Low Power and Lossy
digital microwave technology) communication technologies. Networks (RPL) and the Constrained Application Protocol
3) Wide Area Networks (WANs): The WAN is the back- (CoAP) are examples of two protocols developed by IETF
bone for communication between network gateways, NANs, to connect lower power devices to the Internet. Machine-to-
distributed grid devices, utility control centers and substations. Machine (M2M) communication is an enabling technology for
It is comprised of two interconnected networks, core networks IoT. M2M-based communication networks and their protocol
and backhaul networks. The core network provides communi- stack is discussed in [68], wherein authors briefly discuss
cation to utility control centers with low latency and high data protocol layers and standardization efforts to connect M2M
rate through fiber optics or cellular communications. The back- devices to the Internet. A security protocol is key requirement
haul networks provide broadband connections and monitoring for the successful operation of IoT. In this regard, IETE’s
devices to NANs through wired (e.g., optical networks, DSL), DICE working group (WG) has put some effort to define and
wireless (e.g., cellular network, mobile broadband wireless adapt the Datagram Transport Layer Security (DTLS) protocol
access) or hybrid fiber-wireless networks. [69].
2) Standardization Activities in SG: In SG, there are several
worldwide standardization efforts. IEEE, ANSI, NIST, and
D. Standardization Activities for IoT-aided SG systems
IETF are the key organizations which are involved in the
For the practical realization of technologies, standardization standardization process of SG systems. In Europe, various
activities play an important role. Though efforts have taken Expert Groups have been established which look at the stan-
place on both IoT and SG systems from the perspective of dardization efforts. Besides this, the European Committee for
standardization at the individual level, there is still a need Standardization (CEN), the European Committee for Elec-
to put joint efforts towards standardization of IoT-aided SG trotechnical Standardization (CENELEC), and the European
systems. Below, we summarize standardization activities in Telecommunications Standards Institute have been working
IoT and SG systems. jointly to solve issues raised in the deployment of SG [29],
1) Standardization Activities in IoT: In IoT, several stan- [70], [71]. IEEE-SA has been closely working on SG systems
dardization activities are going on. A very detailed discussion and in this regard, the IEEE P2030 standard has been pro-
on standardization activities from the perspective of IoT and posed, which deals with the interoperability issues of different
different standardization bodies can be found in [26], [64]. power systems. Another effort by IEEE is the approval of
The oneM2M [65], [66] is standard for M2M and IoT, which IEEE’s 1547 standard which deals with the interconnection
is widely adopted. OMA LWM2M (Lightweight M2M) [67] of distributed energy resources. To deal with the challenges
 7

of NAN communications, the IEEE 802.15.4g standard was smart appliances. IoT technology also provides its services to
introduced. This standard deals with the physical layer issues NANs, linking a group of smart homes in a neighborhood
associated with NAN communications [72]. The IEEE P1901 through a NAN to form a smart community [78]. Smart
working group is developing standards for multimedia com- homes in the smart community can thus share the results
munication in cognitive radio-based SG systems [73]. of outdoor surveillance cameras to detect any accident or
In summary, there are still many issues which need to suspicious activity and inform the appropriate emergency
be addressed by the standardization community to deal with centers and police stations autonomously. The concept of
interoperability issues of IoT-aided SG systems. the smart community could be extended to form a smart
city. Similarly, a comprehensive surveillance system could be
III. E XISTING A PPLICATIONS OF I OT- AIDED SG S YSTEMS developed in a smart city to monitor various activities within
There are many existing applications of IoT-aided SG an entire city or even a country. Some IoT requirements and
systems, and many more have been proposed, as represented considerations for smart buildings are discussed in [90] for
in Figure 3. Table II presents the work done on the appli- building management systems and energy optimization.
cations of IoT-aided SG systems. In this section, we discuss 2) Information Management System for Electric Vehicles:
existing applications of IoT-aided SG systems in the literature Electric Vehicles (EVs) provide an eco-friendly transportation
[50], [74]–[85]. option by reducing carbon dioxide emissions [50], [91]. This
is an interesting opportunity for IoT aided SG systems. The
charging system for EV is composed of a power supply
A. HAN Applications system, charging equipment and a monitoring system. The
1) Smart Home: IoT technology plays a major role in power supply system is responsible for the output and man-
the SG for the realization of smart homes and appliances agement of electricity. The charging equipment charges and
[76], [86], [87], such as smart TVs, home security systems, discharges the EVs, and includes both AC and DC chargers.
smart refrigerators, washing machines, fire detection, lighting AC chargers provide slow charging and are generally imple-
control and temperature monitoring. The smart home includes mented in home. Rapid EV charging requires DC chargers,
sensor and actuator nodes for environmental monitoring that generally implemented at public charging stations. Both types
transmit the surveillance data to the home’s control unit [77]. of chargers also include a billing function. The monitoring
The control unit enables users to monitor and control the system is responsible for real-time monitoring of the charging
appliances remotely from anywhere and at any time. The system and its security. IoT technology plays a leading role in
smart home is thus a key element of the SG to realize real- this monitoring system, providing an information management
time interaction between users and the grid, improve the system that integrates different components of the charging
quality of services and enhance the capacity of integrated system [50]. For instance, the IoT technology enables the
grid services, as well as to fulfill users’ energy demands power supply and real-time monitoring systems to send their
in a most efficient possible way. Optimizing daily power information to the information management system, which
consumption makes broad use of smart home services. For in turns passes information to the control station, so that
instance, users can turn on heaters or air conditioners before necessary actions can be taken. EVs are equipped with GPS,
coming home in order to enjoy their desired environment which let the IoT to help drivers to manage their batteries
without waiting. Additionally, users can switch on their power- more efficiently by locating the nearest, most suitable charging
intensive electrical appliances, such as their washing machine, station with the shortest waiting time, as well as providing
in the middle of the night when the electricity price is lower. traffic and parking information [50].
The control unit also uses surveillance data to detect suspicious 3) Automatic Meter Reading: Traditionally, power con-
activities and inform users to take appropriate actions. All sumption information was collected manually on site at spe-
these functions can only able to be realized, thanks to the IoT cific time intervals. This practice inevitably led to inadequacies
technology. Viswanath et al. [88] have designed a system and in terms of accuracy and timeliness. IoT enables Automatic
developed a testbed for smart home with a house having one Metering Infrastructure (AMI) [92] or remote meter reading
living and three bedrooms that can accommodate six to nine systems based on WSN, PLC and Optical PLC (OPLC) by
persons. This system controls the energy based on dynamic using public or private communication networks. An AMI
pricing and performs as an energy management system. Hence, system is one of the most important functions of the SG,
it avoids home appliances usage during the peak hours. The collecting the real-time electricity consumption data with high
authors developed an Android application for allowing remote reliability, processing this information and hence providing
access to consumers as well. Additionally, Shah et al. [89] real-time monitoring, statistics and power consumption anal-
proposed a model for security enhancement in smart home by ysis. The importance of this system lies in the timeliness,
implementing Reed Solomon Codes for error detection and efficiency and accuracy in the power consumption data. By
correction. using this system, IoT technology could help users to save
The IoT is applied to various aspects of the SG in smart money by adjusting their electricity usage behavior based on
homes, for instance, in a smart home’s sensor LAN protocol the analysis of their power consumption [75].
to control smart appliances, multi-meter reading, information 4) Integration of Distributed Energy Resources (DERs):
gathering of power consumption (including electricity, water Renewable energy generators, such as solar cells, photovoltaic
and gas), load monitoring and control and user interaction with cells and wind turbines, are progressively integrated into
 8

TABLE II
S UMMARY OF APPLICATIONS OF I OT- AIDED SG SYSTEMS .

Applications References SG Area SG Sub-Area

Transmission Tower Protection [74] WAN Power Transmission
Monitoring of Power Transmission Lines [75] WAN Power Transmission
Smart Patrol [50] NAN Power Distribution
Smart Home [76]–[78] HAN Power Consumption
Information Management Systems for EVs [50] HAN Power Consumption
Automatic Meter Reading [75] HAN Power Consumption
Distributed Energy Resources [77], [79], [80] HAN Power Consumption
Power Demand Management [81]–[84] HAN Power Consumption

today’s power grid. They have recently attracted considerable component of SG [93]. It is the component that is directly
attention in SG studies due to climate changes and environ- connected to users in smart grid. Smart distribution gid
mental pressures. Renewable energy has a positive impact consists of communication system, power distribution remote
on the global environment by generating electricity without unit, master unit and station unit [93]. With the help of
carbon emissions [79]. Reducing or holding back the annual IoT technology, the smart distribution grid can immediately
increase in greenhouse gas emissions contributes to limiting identify the faults in case of any disorder and can overcome
the Earth’s increasing temperature. In the past few years, the fault instantly. IoT technology helps smart distribution grid
governments and organizations have installed a substantial by providing various types of sensors for collecting data about
number of solar cells and wind turbines to satisfy part of temperature, humidity, noise etc. which ensures monitoring
their power requirements [77]. Power generation patterns of and secure operation of distribution grid [94]. The first
renewable energy sources (solar and wind), which are dis- demonstration of smart distribution by using IoT technology
tributed over the grid, are intermittent in nature and dependent in distribution network of smart grid was performed in Henan
on location and climate, so they pose significant challenges Hebi IoT demonstration project in Hebi, China [94]. It uses
for the predictability and reliability of the power supply. Such temperature, noise and tower tilt sensors, as well as ZigBee,
problems are addressed using the seamless interoperability and GPRS, 3G and power fiber. It realizes online monitoring,
connectivity provided by the IoT technology. Furthermore, online inspection and life cycle management. This project
the IoT technology uses sensors to collect real-time weather was performed using 10 kV underground and overhead laying
information which helps in forecasting energy availability in mixed line which covers 45 utility distribution transformers,
the near future. A Kalman Filter-based state estimation and 68 units of surge arresters, circuit breakers on 20 pillars, a
discrete-time linear quadratic regulation method for controlling ring counter, and four cable branch box. Readers are referred
the state deviations is proposed in [80], utilizing IoT aided SG to [94] for detailed description and results of this project.
systems. It uses IoT technology and WSNs to sense, estimate 2) Smart Patrol: The patrolling of power generation, trans-
and control the states of DERs. Furthermore, a web of things- mission and distribution used to be mainly a manual task,
based SG architecture is proposed in [85] for the remote performed regularly at specific time intervals. However, due to
monitoring and controlling of renewable energy sources. climate conditions and both human and environmental factors,
5) Power Demand Management: Power demand manage- the quality and quantity of patrolling is not always as desired.
ment, also known as demand-side energy management, is Furthermore, it is usually not easy for power workers to patrol
defined as the change in the energy consumption profiles unattended substation equipment. The IoT technology offers
of consumers according to the time-varying electricity prices a promising solution to this problem by introducing Smart
from utility companies [81], [82]. It is used to minimize the patrol [50]. Smart patrol is comprised of WSN and RFID
consumer’s electricity bill, and to reduce the operational cost tags which are connected to the power substation with the
of the power grid and energy losses, as well as to shift the help of IoT technology, and are used to locate the power
demand load from peak times [83]. IoT devices collect the equipment in order to improve the quality of patrolling, as well
energy consumption requirements of various home appliances as to enhance the stability, efficiency and reliability of a power
and transmit them to the home control units. Subsequently, system and its supply. Smart patrol can be used for a number
the SG control unit schedules the energy consumption of home of applications, such as patrol staff positioning, equipment
appliances based on the users’ defined preferences so that each status reports, environment monitoring, state maintenance and
consumer’s electricity bill is minimized. Demand-side energy standard operations guidance [50].
management can be performed at different levels of the SG.
For instance, it can be performed at a home-level to maintain
C. WAN Applications
consumers’ privacy. It can also be performed at higher levels to
not only benefit the consumers but also the utility companies 1) Transmission Tower Protection: An integral part of
by generating a more effective scheduling plan [84]. power transmission, the transmission tower protection is a
WAN application of IoT-aided SG systems, developed to
enhance the safety of transmission towers from physical dam-
B. NAN Applications age by plundering of components, natural disasters, unsafe
1) Smart Distribution: Smart distribution is based on ad- construction and growing trees under the foundations [74].
vanced automated IoT technology and is one of the important Burglary and intentional damage by people are the major
 9

causes of transmission tower damage. Natural disasters, such time images to the sink if the threat is identified. For huge
as typhoons, strong thunderstorms and global warming affects construction machinery and trees that get too close to the
also can cause transmission towers to collapse. Additionally, transmission lines, the video cameras directed towards the
key infrastructure projects, such as highways and high speed transmission lines identify the threat and transmit real-time
railways, are often constructed near the transmission towers images to the sink. In all of the threat cases, the sink generates
and must sometimes cross high voltage transmission lines. Of- an alarm and forwards the images to the monitoring center
ten the construction companies are not fully aware of the risks through the Internet. The monitoring center handles the real-
involved in operating near high voltage transmission towers. time data of a series of transmission towers. The alarm signal
They use a number of large construction machines which not and images from the sink inform the monitoring center staff
only pose serious dangers to their crew, but can also damage about any threat to a transmission tower and the staff then
transmission lines and towers. Such construction contractors takes appropriate actions to handle such threats.
sometimes do not inform the relevant power transmission 2) Online Monitoring of Power Transmission Lines: The
departments which makes it impossible for power transmission online monitoring of power transmission lines is one of the
employees to inspect and monitor all the power transmission most important applications of the IoT in the SG, specifically
facilities, which could lead to risks to transmission towers. for disaster prevention and mitigation. In recent years, natural
Currently, the main method of transmission towers protec- disasters have highlighted the challenges of security, reliability
tion is manual patrol by the staff. However, regular manual and stability inherent to high voltage power transmission
patrol of high voltage transmission lines and towers by staff lines. Traditionally, high voltage transmission line monitoring
is very difficult due to manpower realities, divisions of respon- has been performed manually. Sensors measuring conduc-
sibility and the level of knowledge of the staff. Furthermore, tor galloping, wind vibration, conductor temperature, micro-
some transmission towers are difficult to approach due to their meterology and icing can now be used to achieve real-time
physical positioning. Hence, the patrolling quality cannot be online monitoring of power transmission lines [75]. This new
guaranteed. The patrolling period varies from 1-10 weeks, online power transmission line monitoring system is comprised
which means insufficient monitoring and higher security risks. of two parts. In the first part, the sensors are installed on
While it is true that some equipment, such as cameras and the power transmission lines between transmission towers to
infrared alarms, are installed on transmission towers to monitor monitor the states of power transmission lines. The second
burglary and other potential damage, the accuracy and stability part has sensors installed on the transmission towers in order
of this equipment is not yet satisfactory [74]. to monitor their states and their environmental parameters. IoT
With the help of WSNs, IoT technology can provide remote enables the communication between the power transmission
monitoring in addressing these security threats. The IoT-aided line sensors and the transmission tower sensors.
transmission tower protection system contains various sensors
which generate early warnings of threats to high voltage D. Summary and Insights
transmission towers, enabling quick responses. The sensors
include vibration sensors, anti-theft bolts, a leaning sensor and We have comprehensively surveyed the existing applications
a video camera. These sensors and the sink node form a WSN of IoT-aided SG systems. While there could be many appli-
[74]. The sensors detect any threat, and send the relevant cations of IoT-aided SG systems, as presented in Figure 3,
signals to the sink node. The sink node receives these signals to date there is very little work on these applications. Some
from the sensors, processes them into data and transmits the applications of IoT-aided SG systems are already deployed, but
data to the monitoring center through the Internet or any other many more are yet to materialize, when the full capabilities
public/private communication network. of instantaneous knowledge and massive data processing are
The sensors’ deployment is presented in Figure 4 and is as exploited. Applications today tackle several areas of interest,
follows. The anti-theft bolts are deployed on the lower part such as:
of the tower. One vibration sensor is deployed underground in • Surveillance of premises or of power equipment installa-

the base of the transmission tower and the other is deployed tions (towers and power transmission lines);
on the tower, about 3-5 meters above the ground. The leaning • Adjusting home consumption by dynamic scheduling,

sensor is deployed close to the vibration sensor on the tower. which takes advantage of fluctuating pricing;
The camera is installed on the tower about 6-8 meters high, • Meter reading and consumption monitoring, residential

directed towards the transmission line. Finally, the sink node and commercial;
is deployed in the middle of the transmission tower. • Electric Vehicle charging and parking;

The vibration sensors monitor the vibration signals of the • Power demand and supply management, including inte-

ground and tower. When they detect signals that indicate grated renewable energy sources; and
excavation construction close to the transmission tower, they • Maintenance of power supply systems, by detecting line

transmit those signals to the sink node. The vibration sensor faults and failures.
on the tower, the leaning sensor and the anti-theft bolts can This survey reveals that there is little literature on the
detect the signals of a burglary or vandalism in progress, and application of IoT-aided SG systems in the area of power
subsequently transmit those signals immediately to the sink generation, transmission and utilization. The potential of
node. In both cases, the sink node combines and processes renewable energy may depend, for example, on IoT-based
the signals, and triggers the video cameras to send real- prediction of weather conditions that can regulate energy flow
 10

Fig. 4. IoT-aided transmission tower protection system for the safety of transmission towers from the threats of buglary, natural disasters, barbaric construction
and growing trees. This system is comprised of a sink node and various sensors which generate early warnings to the monitoring centers about the threats to
high voltage transmission towers. The sensors include two vibration sensors (one underground in the base of tower and other on the tower), anti-theft bolts,
leaning sensor and video cameras [74].

between different regions, and on monitoring the efficiency of the perception layer to the telecommunication protocols [95].
the involved equipment (solar panels, wind turbines). Subsequently, it transmits the mapped data to the application
layer through the relevant telecommunication network. The
IV. A RCHITECTURES FOR I OT- AIDED SG S YSTEMS core network, i.e., the Internet is responsible for the routing,
A number of architectures are available for IoT-aided SG information transmission, and control. The access network
systems. We survey such architectures in this section, and their will be based on other telecommunication networks. The
classification is represented in Figure 10. IoT management and information centers also belong to the
network layer. The network layer can rely on public as well
A. Three-layered Architecture as on industry-specific communication networks.
3) Application Layer: The application layer is the inte-
A three layered architecture for IoT-aided SG systems, gration of IoT technology and industry expertise for the
based on the characteristics of the IoT, has been widely used realization of a broad set of IoT-aided SG applications [96].
in the literature and it has been proposed in [45], [46], [50], Its function is to process the information received from the
[75], [95]–[97]. As presented in Figure 5, this architecture network layer, and based on this information, it monitors and
is comprised consists of three layers, a perception layer, a troubleshoots the IoT devices and SG environment in real-
network layer and an application layer. time. It provides various applications of IoT-aided SG systems
1) Perception Layer: This layer enables the main objectives
which are presented in Figure 3. It is comprised of application
of sensing and collecting information in IoT-aided SG systems
infrastructure/middleware and various types of servers related
using a variety of devices. It is comprised of various kinds
to content, web services and directory services. The applica-
of IoT sensing devices, such as RFID tags, cameras, WSN,
tion infrastructure/middleware provides computing, processing
GPS and M2M devices, in order to collect data in a SG. It is
and resources for IoT technology. The key elements provided
categorized into two sub-layers, a perception control sub-layer
by the application layer are information sharing and security.
and a communication extension sub-layer. The perception
The application layer is set to grow, especially for SGs that
control sub-layer realizes the perception of the physical world
can provide much richer data sets. In turns, these applications
by processing IoT devices, information acquisition, monitoring
dictate what data is required, at what time intervals, from the
and control, while the communication extension sub-layer has
sensors.
a communication module which connects IoT devices with the
network layer.
2) Network Layer: The network layer is comprised of B. Four-layered Architecture
the converged network formed by various telecommunication A four-layered architecture for IoT-aided SG systems based
networks and the Internet [96]. The network layer has been on the characteristics of SG information and communication
widely accepted due to its mature technologies. Its function systems was proposed in [58]. As presented in Figure 6, this
is to map the information collected by the IoT devices in architecture is comprised of a terminal layer, a field network
 11

layer, a remote communication layer and a master station

system layer. According to the three-layered IoT hierarchical
model, the terminal and field network layers in this architecture
correspond to the IoT perception layer, the remote communi-
cation layer corresponds to the IoT network layer, and the
master station system layer corresponds to the IoT application
layer.
The terminal layer is comprised of IoT devices deployed
in various SG functions, such as power generation, trans-
mission, distribution and utilization. The IoT devices include
remote terminal units, information collection devices, smart
meters, smart devices and intelligent electronic devices. This
layer collects information from IoT devices and transmits the
collected data to a field network layer. The field network
layer can be wired or wireless. Depending on the type of IoT
devices, the appropriate communication network is used. As an
example, ZigBee is used by sensor nodes in order to transmit
the collected data to a remote communication network layer.
The remote communication network layer can also be
Fig. 6. Four-layered architecture of IoT-aided SG systems based on SG
wired or wireless. It is comprised of various communication characteristics. It is comprised of a terminal layer, a field network layer, a
networks which provide connectivity to the Internet, such as remote communication layer and a master station system layer [58].
2G, 3G, and LTE as wireless networks, and optical networks
as a wired network. This layer serves as a middleware between
IoT devices and the master station system layer. The master station system layer is the control and information system of a
SG. It controls and manages all the SG functions. It can also
be considered as an interface to the IoT-aided SG applications.

C. Energy Efficient Architecture

Improving a building’s energy efficiency is an important
aspect of the SG, an aspect that is required for global sus-
tainability. Thus, smart energy is an important research area
of the IoT. In the United States, buildings are responsible for
around 71% of the total electrical energy consumption [98], a
major motivation for designing green buildings to be energy
efficient. However, thus far, such green buildings have not
been as energy efficient as expected [99]. This may be due
to the ill-fitting of rigid and static schedules to the lifestyle of
consumers, or to the unpredictability of business requirements.
Now, with the help of IoT technology, the ability to change
energy settings can be entrusted to the users, using their smart-
phones or PCs. This allows users to suit their own schedules
at will, and respond to events instantaneously by adjusting the
policies as and when necessary. In 2015, an energy efficient
location-based automated energy control IoT framework was
proposed, using smartphones and cloud computing technolo-
gies [99]. That proposed framework changes the currently
static energy management and centralized control modes to
dynamic and distributed energy control at the consumer-side
SG, which is composed of usually various buildings.
This framework is represented in Figure 7 and is comprised
of four main components, (i) multi-source energy saving
policies, (ii) monitoring and control via mobile devices, (iii)
location-based automatic control, and (iv) a cloud computing
platform for data storage and computation. An organization’s
Fig. 5. Three-layered architecture of IoT-aided SG systems, comprised of
a perception layer, a network layer and an application layer. The perception
premise consists of various parts, such as a campus, build-
layer consists of two sub-layers, a communication extension sub-layer and a ings, departments, labs and rooms/offices, each with different
perception control sub-layer. energy requirements and policies that are important to take
 12

into account while managing energy consumption. Even in a message will be sent to the policy server to trigger the policy
a single home, each family member has his/her own energy control process, e.g., start the heating. The now deserted home
consumption preference which needs to be considered. Hence, will transit into an energy saving mode while the office build-
this framework has a tree-like structure control plane as ing will start the processes based on the user’s preferences,
presented in Figure 7, comprised of several layers of energy such as cooling/heating to the user’s desired temperature.
saving policies at different levels (e.g., building, department, Finally, this framework uses the cloud computing platform
lab and room). for data storage and modeling computation and analysis. The
Smartphones are equipped with multiple networking in- cloud provides the basic data storage and retrieval services
terfaces, such as Wireless Fidelity (WiFi), 3G, Long Term for each building’s energy consumption data. Most of the
Evolution (LTE), bluetooth and WiMAX, as well as GPS, computation-intensive modeling and analysis jobs are per-
which helps to obtain precise location positioning. Thanks to formed in the cloud. Moreover, the cloud provides security,
the global Internet access, smartphones are the ideal choice for reliability and configurability for the network communication
monitoring and controlling energy control systems remotely between the cloud and the user. The cloud has an application
from any place at any time. After an initial authentication and layer which contains a user friendly web interface for manag-
authorization, consumers can dynamically modify their energy ing the buildings’ energy systems, so that a remote user can
saving policies through their smartphones by interacting with easily configure and manage the system.
the policy servers of their home and their office buildings.
The consumer can be connected to the Internet through smart- D. Web-enabled SG Architecture
phones, tablets, or laptops with WiFi or with 3G networks. An architecture for IoT-aided SG systems is proposed in
Furthermore, the location information of smartphones is [85] based on the web of things (Figure 8). The web of things
used in designing automatic control policies which can switch is comprised of a number of web services provided on top of
energy consuming devices ON or OFF in homes and office the IoT devices in which the web browser acts as an interface
buildings by detecting the location and direction of user to these web services. There are two types of energy sources,
movements. These dynamic adjustment policies also enable non-renewable and renewable energy sources. Non-renewable
the coordination between buildings’ policies. For example, energy sources include thermal power plants (combusting coal
as presented in Figure 7, when a user has moved out of or oil) that release carbon emissions to the environment, as
a predefined distance range from the home premises and is well as nuclear power plants. The renewable energy sources
moving into a predefined distance range of an office building, are environmental friendly and are comprised of hydropower,
wind farms, solar, biogas plants and biofuel sources, as well
as geothermal and tidal/wave sources.
The energy sources in this architecture are connected to
individual digital energy meters. These digital energy meters
are responsible for collecting household energy consumption
data. The meter readings from energy meters of non-renewable
and renewable sources are collected by separate IoT gate-
ways which communicate regularly with these meters. The
collected data from IoT gateways are updated to the server
periodically, and the server provides web services on top of
these IoT devices. These web services include the locations of
houses connected through the SG, and the meter information.
Moreover, for each home, the scheduling of power sources and
the controlling of the energy sources by switching the source
controllers remotely through IoT devices are provided as web
services. Through the Internet, by connecting to any device,
a user can access these services. The energy sources for each
household are switched through source changers which are
controlled through IoT devices; these IoT devices switch the
energy sources upon receiving instruction from the user via
the server.

E. Last-meter SG Architecture
The last-meter SG is a part of the SG that is closer to the
Fig. 7. Smart location-based automated energy control IoT framework for home, i.e., the part with which the consumers interact. An
energy efficiency, changing the current static and centralized energy control architecture for the last meter SG embedded in the IoT is
to dynamic and distributed energy control. It is comprised of four main
components, i.e., multi-source energy saving policies, monitoring and control
proposed in [100], [101] and is presented in Figure 9. This
via mobile devices, location-based automatic control, and a cloud computing architecture is comprised of three main components, sensor
platform for data storage and computation [99]. and actuator networks, an IoT server and user interfaces.
 13

Fig. 9. Last-meter SG architecture embedded in the IoT [100], [101]. It is

comprised of three main components, sensor and actuator networks, an IoT
server and user interfaces. The sensor and actuator networks consist of sensor
and actuator nodes, and IP gateways. The IoT server consists of a message
dispatcher, an SG database, a data management unit, a configuration database,
a configurator unit, a user database and a secure access manager. The user
interfaces consist of a visualization interface, a configuration interface and a
web service API.

Fig. 8. Web-enabled SG architecture comprised of web services on the top

of IoT devices [85]. It has two types of energy sources, renewable and non- corresponding network address, provided by the local network
renewable, both of which are connected to digital energy meters. The digital coordinator. Subsequently, when a node sends a message to the
energy meters collect household energy consumption data which is further
collected by IoT gateways to communicate to the server containing the web
IoT server, the gateway node translates its network address into
services. a unique ID, and vice versa.
b) IP gateway: The IP gateway connects sensor and
actuator nodes to the IoT server via an IP link if there
1) Sensor and Actuator Networks for Various SG applica- is no IP capability. For uplink communication, the gateway
tions: The sensor and actuator networks enable the realization collects data from sensor/actuator nodes, performs encapsu-
of various SG applications and contain two main components: lation/reformatting (if required) and forwards the data to the
sensor and actuator nodes, and IP gateways. message dispatcher in the IoT server over a secure TCP/IP
a) Sensor and actuator nodes: The sensor and actuator link. For downlink communication, it forwards the commands
nodes can be part of wired or wireless networks. This ar- received from the IoT server to the intended receiver nodes.
chitecture can accommodate heterogeneous sensor networks, In general, the gateway performs a conversion of data into
as the data management unit in the IoT server translates a universal format. However in this architecture, such con-
information into the format required by the SG database. The version is performed by the IoT server, and therefore the
bi-directional node communication enables an IoT server to gateway sends the messages over a TCP/IP link in the native
configure, program and interrogate the sensor and actuator format. This procedure has two main advantages. Firstly,
nodes. This architecture supports the possibility of adding or the gateway will have low computational complexity and
removing network components in real time, even if specific hardware requirements. It only has to ensure an IP connection,
node characteristics depend upon the network implementation. encapsulate the messages into TCP/IP packets and ensure the
Hence, any node can join the network without changing security level required by specific applications. Secondly, new
the network implementation and the newly added node is functionalities and applications can be developed and added
automatically identified and accessible from the network ad- without modifying the gateway.
ministration interface for registration and configuration. 2) IoT Server: The IoT server has four main components:
Each sensor and actuator node has to be uniquely identified the first one is a message dispatcher, the second one is a
in order to ensure global accessibility. However, the node data management unit and database storage, the third one is a
addresses in sensor and actuator networks are generally only configurator unit and database, while the fourth one is a secure
unique within a single network and may change over time. access manager and user database.
Therefore, the IoT server assigns a unique ID to each node a) Message dispatcher: To perform bi-directional com-
of the network and maintains the mapping of each ID and its munication between each gateway and the rest of the system,
 14

the message dispatcher will be used. The message dispatcher service API.
will listen to new connections from IP nodes that want to join a) Visualization interface: It displays the current and
the system for uplink communication. It decrypts incoming past information of household energy consumption. Addition-
packets for every connection and forwards them to the data ally, it allows authorized users to send commands to actuator
management unit for storage and interpretation. For downlink nodes attached to home appliances.
communication, it encrypts the messages from the configurator b) Configuration interface: It enables the users to re-
unit into a TCP message and then forwards them to the motely manage and configure their networks. Additionally,
destination gateway. users can set the data visibility of their household appliances
b) Data management unit and database storage: The as well as third-party applications access. This interface also
data management unit is a collection of software modules. In allows administrators to remotely register new gateways and
this unit, each software module is able to manage the messages configure new network connections. To establish a new con-
of a specific sensor/actuator network. These modules receive nection, each gateway requires the IP address and network
messages in their native format and then extract their payload. name of the message dispatcher, the port number on which
Two storing mechanisms are performed based on the payload. the message dispatcher can accept the connections and the
Firstly, the messages are stored in a unique format in the network AES security key.
SG database if the payload contains either measurement data c) Web service API: It opens the platform (IoT server) to
from a sensor node or a notification from an actuator node. service providers and new client applications. APIs offer an
Secondly, the message is stored in the original format into the easy and unified way to retrieve information collected from
configurator database if the payload contains specific network heterogeneous sources. Utilities, service providers and third-
messages (such as configuration, management information, a parties use web service APIs to obtain single, multiple or
node address or a communication channel). aggregated measurement data, as well as to notify consumers
The SG database decouples the data collection from the data about dynamic changes in tariffs, of weather related data and
processing and visualization so that the users do not have to alarms. For security and privacy purposes, only registered
interrogate the nodes directly. The decoupling allows the nodes users and authorized third-party applications can access energy
to stay in sleep mode most of the time and to only wake up consumption data from an SG database through web service
periodically to receive commands and configuration messages, APIs.
as well as to send measurement and status data. This approach
is very useful when the sensor networks are heterogeneous, as
well as when nodes are battery powered. The sensors’ data F. Summary and Insights
in a SG database is represented by a unique format that is We have surveyed the existing architectures for IoT-aided
associated with the physical nodes through the unique node SG systems. As presented in Figure 10, the major focus
ID, independent of the local sensor network protocol. In this of the existing studies is either generic layered architectures
manner, data can be processed and visualized independently or the HAN architectures specifically for controlling and
of the characteristics of the physical source. managing home appliances remotely. However, the layered
c) Configurator unit and database: The configurator unit architectures (three-layered and four-layered) are very generic
is a collection of software modules, where each module is architectures and they do not cover many specific aspects
dedicated to a specific type of sensor/actuator network. This of SG, such as all the networks (HAN, NAN and WAN)
unit configures the nodes and networks according to the inputs and systems (power generation, transmission, distribution and
from applications and users, as well as according the stored consumption). Hence, it is very important to consider these
system status in the configurator database. specific aspects in an architecture for IoT-aided SG systems.
d) Secure access manager and user database: This Additionally, four-layered architecture is an extension of three-
component coordinates all the communication between the layered architecture by dividing the network layer into field
end users and the IoT server by ensuring privacy and data network layer and remote communication network layer. It is
protection. It only allows authorized users and third-party very important to analyze why there is a need of four-layered
applications to access the stored information and the network architecture and why we should not have five-layered or six-
configuration based on the user database, and the users’ layered architectures.
access permission for each resource. The network owners have The other three architectures (energy efficient, last meter
administrative privileges over their networks by default. SG and web-enabled SG) are mainly designed for HAN.
3) User interfaces: The user interfaces allow users, appli- The energy efficient architecture saves energy by using smart
cation developers and service providers to interact with an location-based automated energy control framework, the last-
IoT server. The user interfaces offer different functionalities meter SG architecture aims at automating the HAN of SG
to standard and administrative users. Standard users can only using IoT technology and the web-enabled SG architecture
access the energy consumption data pertaining to their own also connects HAN to the web by offering a number of web
households. Administrative users, meanwhile, have higher services on the top of IoT devices. However, these current
access privileges and they can also view the configuration and studies on architectures for IoT-aided SG systems are very
status of the IoT devices, and configure them dynamically. limited; there are still no architectures for NAN and WAN of
The user interface is divided into three main components: a IoT-aided SG systems, and other areas of HAN are also yet
visualization interface, a configuration interface and a web lacking architectures. Hence, we strongly recommend future
 15

Fig. 10. Architectures for IoT-aided SG systems classified into layered architectures and HAN architectures.

research on architectures for NAN, WAN and the other HAN and smart devices with location sensors that have a GlobalSat
areas, which should focus on the individual characteristics and GPS module. This prototype requires two software packages.
requirements of HANs, NANs and WANs. Subsequently, the One package handles the data recording of GPS locations
individual architectures for HAN, NAN and WAN will need and sends that information to the server in National Marine
to be combined as a whole for the realization of complete Electronics Association (NMEA) 0183 compliant format. The
IoT-aided SG systems. Since, IoT and non-IoT communication other package is a configuration and management software for
technologies play a vital role in the realization of IoT-aided SG WiFi routers that provides a port mapping service for accessing
systems in order to provide connectivity, it is very important the server from outside the Network Address Translation
to consider communication technologies and topologies in the (NAT). The server executes Python codes programmed with
overall architectures for IoT-aided SG systems. Common Gateway Interface (CGI) scripts to control the WeMo
devices through the Universal Plug and Play (UPnP) protocol.
V. P ROTOTYPES FOR I OT- AIDED SG S YSTEMS The smart device is able to control the electrical appliances
in both locations through the inter-operation of hardware and
Prototypes play an important role in the development of
software.
a system in order to test various functions and verify the
operation before the actual commercial implementation. Some
prototypes have been developed for IoT-aided SG systems as
well, however they are very few and there is a need to develop
more prototypes for IoT-aided SG systems. In this section, we
discuss the existing prototypes of IoT-aided SG systems.

A. A Simple Prototype for Energy Efficiency

A simple prototype for smart energy involving one user
was designed in a Leadership in Energy and Environmen-
tal Design (LEED)-gold-certificated green office building. It
linked electrical appliances in two locations, an apartment
and an office in an office building [99]. This prototype is
a simplified scenario of the framework presented in Figure
7. The objective of this prototype was to enable a user to
dynamically control and manage appliances in two locations. Fig. 11. Smart location-based automated energy control IoT framework
for energy efficiency with the objective of enabling users to dynamically
The prototype enables the server to trigger energy policy control and manage the appliances in residential and office buildings. This is
control process by turning electrical appliances ON or OFF a very simple prototype in which one user controls devices in two locations.
in the two locations after detecting changes in the user’s The hardware includes Kill-A-Watt electric meters, WeMo controllers, WiFi
routers, one server in each location and smart devices with GPS modules. Its
location. This helps users to effortlessly implement and control software includes a GPS location data recording software, and a configuration
their own energy policies in real time, making their energy and management software for WiFi routers.
consumption directly proportional to their actual utilization.
Although this simple prototype only involves one user with In this prototype, a smart mobile device (smartphone,
control devices in two locations, this system could be easily tablet or laptop), with a location sensor, sends its location
scaled up with multiple users controlling their devices at at predefined intervals to two servers in two locations (i.e.,
different locations simultaneously. the apartment and the office room). The servers inside the
Figure 11 presents the structure of this prototype. The NAT or firewall are accessed from outside by means of port
hardware includes Kill-A-Watt electrical meters [102], WeMo mapping technology. The servers calculate the distance of
controllers [103], WiFi routers, one server in each location, the mobile device from themselves, and if the distance is
 16

within the predefined threshold, they trigger the energy policy Internet in order to transmit or to receive data. Thirdly,
control process that instructs the controller to turn the electrical terminating the connection upon completion of a transmission
appliances ON or OFF in the two locations according to pre- or reception.
defined energy policies. This simple example of automation The software includes a Graphical User Interface (GUI) for
demonstrates the ambition to move from explicit (but hit-or- accessing the web services and user account management. A
miss) user actions to intuitive and sensor-based activation that consumer can access web services via the web browser (any
are more reliable. Further sensing (e.g., weather) and other sort) of any device connected to the Internet. A consumer ini-
data (calendar, habitual behavior) can be combined to refine tially registers himself, and after the required verification and
the automatic activation control. authentication, the administrative staff adds the consumer’s
home to the system. Subsequently, the consumer can monitor
his home’s average power consumption, making it possible to
B. Integration of Renewable and Non-renewable energy
keep track of energy consumption and plan the scheduling of
Sources at Home
power sources.
A prototype for integrating renewable and non-renewable
energy sources at home using IoT is presented in [85]. This
prototype is designed for the HAN architecture presented in C. In-Home Appliance Monitoring Implementation
Figure 8. The electric meters are connected to non-renewable An in-home IoT-aided SG prototype is implemented in
and renewable energy sources to record the current and voltage [100], [101] for the architecture presented in Figure 9. This
readings. This configuration has a direct connection to IoT prototype includes a ZigBee network connected to the IoT
embedded devices, and hence the IoT embedded devices server through a ZigBee IP gateway. In order to collect the
change the energy source by controlling the source changers real time energy consumption of home appliances, smart plugs
connected to the SG supplies of the houses. The web services are deployed between each home appliance and its wall socket
allow consumers to track energy consumption on various to collect the loading information of the each home appliance.
scales, such as daily, weekly, monthly and yearly, as well as The smart plug is also an actuator node that can turn the load
to compare the consumption data for different times, and thus ON and OFF, configured through the user interface.
enable consumers to configure and schedule the switching of The communication with the ZigBee network is provided
energy sources (e.g., non-renewable or renewable) in advance. through a Freescale MC13224 SoC. The board includes an
The consumer also has the ability to reconfigure the current ARM7 processor with 128KB of flash, 96KB of RAM and
energy source instantly in case of an emergency. 80KB of ROM memory. For energy measurement, an analog
The pilot hardware and software components are as follows: device ADE7953 was used. Load control was implemented
The hardware includes an ARM cortex M3 processor to using a single-pole bistable 12V relay which can support up
design the IoT device. An LPC1768 processor from NXP to 16A. The power supply unit in the board can supply 12V
[104] is used as the ARM processor, interfaced with an for the relay and 3V for the ADE7953 IC and MC13224 SoC.
Ethernet port, an LCD and an RS232 port. CMSIS [105] is BeeStack, a Freescale ZigBee stack, is used to implement the
used as a real-time operating system for task optimization, firmware, running on the smart plug.
and LwIP protocol stack is implemented to support TCP/IP Since the smart meter is a ZigBee device, a ZigBee/IP
functionality on the board. The ARM processor interfaces gateway is used for the communication with IoT server. The
its Universal Asynchronous Serial Transmission (UART) port ZigBee/IP gateway is comprised of a micro controller, an
with MAX232 ICI in order to communicate with its RS232 Ethernet interface and a ZigBee RF transceiver. The Freescale
port. However, the collected data from electric meters is Kinetis K60 is used as a micro controller which is based
in the form of RS485 port out. Hence, the output from on an ARM Cortex M4 processor with hardware encryption.
RS232 is converted to RS485 for compatibility and the RS484 The message dispatcher is implemented as a multi process
MODBUS protocols can then transmit serial data up to 1.2km. application on a Linux machine. It continuously listens to new
Several electric meters can thus be accommodated by a single connections from the gateways.
processor board within a 1km radius. Current and voltage The data collection unit is implemented using CoMo soft-
transformers are required if the the current is higher than 5A ware [106]. It is designed for fast prototyping of network data
and the voltage is higher than 450V. The transformers also mining applications and it is used in large testbed deploy-
help to isolate the electric meters from the high current and ments, e.g., PlanetLab [107]. Hence, it is scalable to very
voltage of the input supply. The electrical source changers are large systems with high data rates. The CoMo architecture
basically DC voltage-controlled, interfaced with the embedded provides an abstraction between the IoT server and the network
controller board through relay controllers (such as H-bridge interface. It follows a modular approach and it is comprised
drivers). of several modules. Each CoMo module interprets the packets
To incorporate the Internet with this prototype, an Ethernet of a specific type of sensor network that belongs to the
port RJ45 is interfaced to the LPC1768 processor. The Internet households and extracts data for inclusion in either the SG
connection is established with the help of the LwIP protocol database or a configuration database. The configurator and user
suite in three steps. Firstly, initializing the Internet connection, databases are implemented with MySQL. After the required
i.e., mapping a MAC address to the IP address with access authentication, the Tornado implements direct access to the
to the world wide web (www). Secondly, connecting to the home appliances data using HTTP.
 17

The user interface is implemented as a web interface. The

web interface, the secure access module and the configurator
unit are based on an open source non-blocking web server,
known as Tornado [108]. The graphical interface is based on
Twitter Bootstrap [109]. The web interface mainly consists of
five functionalities, home appliances’ energy consumption data
visualization, registration and configuration of new networks,
privileges management, registration of new sensors for home
appliances, and sending commands. Firstly, the visualization of
energy consumption of the home appliances provides a time
range selector and a graph where energy consumption data
is plotted as a function of time. The time option allows the Fig. 12. Prototypes for IoT-aided SG systems.
desired measurement period to be selected. Secondly, there are
dedicated web interfaces that administrators use to register
and configure new networks. The network configuration is VI. B IG DATA AND C LOUD FOR I OT- AIDED SG S YSTEMS
stored in the configurator database. Thirdly, the administrators
A. Need of Big Data Management in IoT-aided SG Systems
can manage privileges by assigning or removing different
privileges to different users and third-parties. These privileges The integration of IoT technology with SG comes with a
include the modification of specific network options (such cost of managing huge volumes of data, with frequent process-
as the communication channel and the security level), as ing and storage. Such data includes consumers load demand,
well as managing sensors’ visibility of home appliances and energy consumption, network components status, power lines
security. Fourthly, any unregistered home appliances that are faults, advanced metering records, outage management records
sending data through a registered network or gateway are and forecast conditions. This means that the utility companies
automatically detected by the system. The system then informs must have hardware and software capabilities to store, manage
administrators about such home appliances, and they decide and process the collected data from IoT devices efficiently and
whether to register these new home appliances or not. Finally, effectively [111].
authorized users send commands to the actuator nodes in home Big data is defined as data with huge volume, variety and
premises so that the required actions can be performed. velocity (three V’s). The high frequency of data collection
by IoT devices in SG makes the data size very large. The
variety is represented by the different sensors that produce
different data. The data velocity represents the required speed
for the data collection and processing. Hence, IoT-aided SG
D. Summary and Insights systems can apply the techniques of big data management and
processing (such as hardware, software and algorithms).
As presented in Figure 12, there are not many published The frequency of data processing and storage for IoT-
prototypes in this area, and they tend to be very simple. aided SG systems varies from application to application.
For instance, the prototype for energy efficiency (see Sec- For instance, some applications perform their tasks during a
tion V-A) is a very simple prototype involving one user specific time of a day, such as weather forecasting, which
controlling home appliances in two locations (i.e., at their can be performed daily at the night time. Other applications
home and in an office building). Similarly, the integration of perform their tasks all the times, such as real-time online
renewable and non-renewable energy sources prototype (see monitoring of transmission power lines, so these requirements
Section V-B) and in-home appliance monitoring prototype need consideration in managing and processing their data [77],
(see Section V-C) associate the home appliances with the [112]. Big data analytics can help to manage real-time and
web through web services. Consumers can access these web huge data [113].
services through any browser and use them to control their In SG, the Supervisory Control and Data Acquisition
home appliances, to track energy consumption and to switch (SCADA) system is the main element of decision making. It
the energy sources (e.g., renewable and non-renewable). More collects data from IoT devices which are distributed over the
challenging prototypes would require cooperation of HAN, grid and provides real-time online monitoring and controlling.
NAN and WAN, affecting the various subsystems (i.e., power Additionally, it helps to manage the power flow throughout the
generation, transmission, distribution and utilization) of SG. network in order to achieve consumption efficiency and power
In addition, combining different IoT knowledge systems with supply reliability. Generally, it is located on local computers
different sensor input systems will yield further intelligent at various sites of the utility companies. With the growing size
functionality than available today. of SGs, utility companies face a challenge in keeping SCADA
From this viewpoint of the prototype literature, it becomes systems updated and upgraded. In order to solve this problem,
apparent that there are no easily available open-source testbeds cloud computing is a good solution to host SCADA systems.
and simulation tools to enable developing experimentations Cloud computing enables on-demand access to a shared pool
and performance evaluation of IoT aided SG systems. For this of computing resources, such as storage, computation, net-
purpose, [110] is a good source of guidance. work, applications, servers and services [114]–[116].
 18

Fig. 13. Classification of big data management in IoT-aided SG systems into platforms and techniques. The platforms include cloud computing and fog
computing while the techniques include MapReduce and stream processing.

B. Platforms to the Cloud, and locally stored data is processed by devices

Cloud computing provides a service-driven model which located at the edge of the network [121]. Hence, fog computing
can be categorized into three types, namely Infrastructure as is a good alternative to Cloud computing for IoT-aided SG
a Service (IaaS), Platform as a Service (PaaS) and Software systems. Jalali et al. [122] explore and evaluate the usage
as a Service (SaaS) [117]. IaaS offers the provisioning of of Fog computing and microgrids together for reducing the
computer infrastructure as a service, generally via Virtual energy consumption of IoT-aided SG systems.
Machines (VMs), such as Amazon EC2, Flexiscale, GoGrid
and Joyent. PaaS provides the support of operating system C. Techniques
and software development to the developers to deploy their There are two main techniques for big data management,
applications within the Cloud, such as Microsoft Windows namely MapReduce and stream processing [77]. MapReduce
Azure, Force.com and Google app engine. SaaS is a multi- [123] is suitable for static and non real-time applications of
tenant platform and can be referred as on-demand services IoT-aided SG systems, such as weather forecasting, while
over the Internet. It provides a ready to use application for the stream processing is suitable for real-time applications, such
end users, such as IBM, Microsoft, Oracle and SAP. Hence, as online monitoring, self-healing and fraud detection, as well
the utility companies can transfer their SCADA systems en- as non-real-time applications. The MapReduce technique is a
tirely to the Cloud and utilize the storage, processing and tool for analyzing large historical data and is frequently used.
other services offered by Cloud computing. In this manner, It splits big data sets into smaller data sets and processes these
the utility companies can get many advantages, such as no smaller data sets concurrently on multiple machines using the
maintenance cost and overhead, better collaboration, pay per same code. Stream processing [124] is considered ideal for
usage and reduced energy cost [118]. Meloni et al. [119] processing sensor data and big data streams. Its architecture is
proposed a cloud-based architecture for SG based on REST designed to handle big data in real-time with high scalability
API by considering interoperability, reusability and security in and fault tolerance. Hence, stream processing has a great
their architecture. The authors also presented a case study to potential for big data management in IoT-aided SG systems
highlight the benefits of this architecture. [77].
However, there is a security risk of using Cloud computing
for SCADA systems due to the shared storage among several
users, which makes it vulnerable to various attacks. Such D. Summary and Insights
security concerns are discussed in detail in [120]. Fog com- IoT-aided SG systems involve processing data that have the
puting, a low latency and highly distributed model, extends the same characteristics as data that requires Big Data techniques.
paradigm of Cloud computing to network ’edge’ to overcome Figure 13 presents the classification of big data management in
this issue. In fog computing, there is no need to transfer data IoT-aided SG systems. Big data management is highly suitable
 19

for IoT-aided SG systems due to its huge volume of data and appliances, and secondly, between smart meters and utility
its ’velocity’ - the need for real-time processed information. control centers. The first data flow can be achieved through
The operation platforms for big data management in IoT- powerline communications or wireless communications, such
aided SG systems include cloud computing and fog computing. as by using 6LowPAN, ZigBee and Z-wave [128]. The second
Cloud computing provides on-demand access to a shared pool data flow can be achieved by using cellular communications
of computing resources. It is a service driven model which or via the Internet. However, there are various key limiting
is categorized by IaaS, PaaS and SaaS. Sending and storing factors that should be considered in the deployment process
data on the Cloud raises security risk issues, due to the shared of smart metering, such as operational costs, deployment
storage among several users, which makes it vulnerable to time, availability of technology, and the operating environment
attacks. Hence, fog computing is used to solve this security (such as rural, urban, indoor, outdoor). Hence, the choice of
risk. Fog computing does not require data transfer to the cloud technology that suits one environment may not be applicable
and the data is processed by devices that are located at the for another. We discuss communication technologies for SGs
network edge. The techniques for big data processing include in the rest of this section.
MapReduce and stream processing. MapReduce is suitable
for static and non-real time applications and it analyzes large A. IoT Technologies
historical data. It splits big data sets into smaller data sets 1) Wireless Technologies:
and processes these smaller data sets concurrently on multiple a) Z-Wave: Z-Wave [129], [130] is a low-power radio
machines. Stream processing is suitable for both real and non- frequency (RF) IoT communication technology for short range
real time applications and is ideal for sensors and big data communication which is mainly suitable for home automation
streams. It is highly scalable and fault tolerant, and it has [131], [132]. It operates on 1 GHz band and offers a data
a great potential for big data management in IoT-aided SG rate of 100 Kbps with coverage range of 30m. Z-Wave is
systems. So far, IoT-aided SG systems do not present new developed by Sigma Designs [133] and it applies 128-bit AES
issues for Big Data processing. However, different SG Big encryption for ensuring security [134].
Data applications with different data sampling frequencies and Advantages: The main advantage of Z-Wave is that it
data types should be further investigated. does not interfere with WiFi and other wireless technologies
operating in 2.4 GHz band, such as ZigBee and Bluetooth
VII. I OT AND N ON -I OT C OMMUNICATION [134], [135]. It is optimized for reliable and low-latency
T ECHNOLOGIES FOR SG communication of small data packets. It is also very scalable
There are many IoT and non-IoT communication technolo- and can control up to 232 devices [135]. It can also support
gies available for SGs and the users have to select among them, whole mesh networks without needing any coordinator.
suitable to their needs. There is a lack of proper guidelines for Disadvantages: The main disadvantage of Z-wave is that
the selection of communication technologies for SG. Therefore since it is mainly designed for small data packets, it is not
in this section, we have listed and discussed various IoT much suitable for NAN and WAN because NAN and WAN
communication technologies together with non-IoT communi- have large aggregated data packets. Additionally, it is a short
cation technologies including their characteristics, advantages range and low data transmission rate communication technol-
and disadvantages. Table III provides a summary of each ogy. Moreover, there is only one company, Sigma Designs, for
communication technology which is discussed in detail in each developing the chips of Z-Wave, unlike multiple sources for
sub-section [125]–[127]. From this section, users will have other wireless technologies (e.g., ZigBee) [135].
a broader view of available communication technologies for b) 6LowPAN: 6LoWPAN (IPv6 over Low-power Wire-
SG and subsequently, they can make better decision for the less Personal Area Networks) [136] is a key IP-based com-
selection of most suitable communication technology based munication technology. It enables IPv6 packets to be carried
on their needs. In the rest of this section, we first briefly within small link layer frames efficiently which are defined
explain the role and need of communication technologies for by IEEE 802.15.4. The key attribute of 6LoWPAN is the
SG and subsequently provide discussion on each IoT and non- IPv6 (Internet Protocol version 6) stack, which has the main
IoT communication technology. role in recent year for enabling IoT. IPv6 provides around
The IoT and non-IoT communication technologies used in 5x1028 addresses for each person in the world, which allows
the SG for data transmission between smart meters and electric any device or embedded object to have its own unique IP
utilities are categorized into two types, wireless and wired address and to connect to the Internet. It is mainly designed
technologies. Wireless technologies have some advantages for home automation (including smart metering). Initially,
over wired technologies in some cases, such as their ease 6LoWPAN was designed to support IEEE 802.15.4 low-power
of connection to otherwise unreachable or difficult areas, as wireless networks in 2.4 GHz band, however it is now adapted
well as their low cost infrastructure. However in wireless to support a number of other wireless newtorking media,
technologies, signal attenuation may occur due to the nature of including sub-1 GHz low-power RF [131]. Unlike Bluetooth
the transmission path. On the other hand, wired technologies and ZigBee, 6LoWPAN provides encapsulation and header
do not have such interference issues and, unlike wireless compression mechanisms [135]. Similar to Z-Wave, it also
technologies, their functions do not depend upon batteries. applies strong 128-bit AES encryption for link layer security
An SG requires two types of information flows. Firstly, defined in IEEE 802.15.4. It offers a data rate of 250 Kbps
between smart meters and IoT devices, sensors and home and coverage range of 10m to 100m.
 20

TABLE III
I OT AND NON -I OT COMMUNICATION TECHNOLOGIES FOR THE SMART GRID

Tech. Protocol Advantages Disadvantages Applicability SG Application Areas Data rate Coverage
range
Z-Wave No interference with other wire- Not suitable for NAN and WAN; HAN Home automation 100 Kbps 30 m
less technologies; reliable; low only one propriety company for
latency; scalable making chips; short range
6LoWPAN Robust; low-power; support Require extensive training and HAN Smart metering; home au- 250 Kbps 10-100 m
IoT Wireless Technologies

large mesh network topology; knowledge; short range; low data tomation
can be applied across various rate
communication platforms
LoRaWAN Low-power; long range; no in- – NAN; WAN Management of operation and 0.3-50 Kbps 2-5km
terference with different data equipment; online monitoring (urban en-
rates; enhance gateways capac- of power transmission lines vironment);
ity by creating virtual channels; and tower 15km
low-cost secure bi-directional (suburban
communication environ-
ment)
ZigBee 16 channels each with 5 MHz of Low data rates; limited energy of HAN Energy monitoring; smart 250 Kbps 10-100 m
bandwidth in 2.4 GHz spectrum; battery; low processing capabil- lightning; home automation;
low power usage; low complex- ities; short range automatic meter reading
ity; low deployment cost
Bluetooth Low power consumption Low data rate; short range; weak HAN Home automation 721 Kbps 1-100 m
security; prone to interference
from IEEE 802.11 WLANs
Cellular commu- Wide area coverage; improved Network congestion due to high HAN; NAN; Monitoring and management 60-240 10-50 km
nications QoS density; critical for emergency WAN of DERs; SCADA Kbps
applications; non-guaranteed
service in unfavorable conditions
(e.g., wind storms)
Non-IoT Wireless Technologies

Wireless mesh Low cost; self healing; self or- Prone to interference and fad- HAN; NAN Monitoring and controlling of Depends Depends
ganization; high scalability; high ing; network coverage problem DERs; automation and protec- upon upon
data rate in rural areas due to low density tion of substation automation protocol deployment
of smart meters; prone to loop
problem due to inclusion of mul-
tiple relay nodes
WiMAX Long range; high data rate Trade-off between performance HAN Real time pricing; automatic 75 Mbps 10-50 km
and distance; costly RF hard- meter reading; outage detec- (LOS); 1-5
ware; low frequencies are al- tion and restoration km (NLOS)
ready licensed and require leas-
ing; high frequencies do not pen-
etrate obstacles
Mobile Low latency; high mobility; high Communication infrastructure NAN; WAN Broadband communication 20 Mbps Vehicular
broadband bandwidth not readily available; high cost; for electric vehicles; SCADA standard
wireless access moderate data rate system; wireless backhaul for (up to 240
SG monitoring km/h)
Digital Long distance coverage; high Prone to multipath interference HAN; NAN Transfer trip between DER 155 Mbps 60 km
microwave data rate; high bandwidth and precipitation and distributed substation
technology
Powerline Cost effective; low installation Noisy and harsh medium; sensi- HAN; NAN Low voltage distribution; au- 2-3 Mbps 1-3 km
communication cost; wide availability; utility’s tive to disturbances; signal qual- tomatic meter reading
own ownership and control; ded- ity gets affected by the type and
Non-IoT Wired Technologies

icated network number of devices, wiring dis-

tance between nodes and net-
work topology; cost of owner-
ship; complexity of management
Digital High speed; low latency; low Quality is distance dependent; HAN; NAN; Smart metering 1-100 Mbps 5-28 km
subscriber lines installation cost; high data rate, high installation cost in low den- WAN
(DSL) high capacity; long range; wide sity (rural) areas; unreliable
availability
Optical commu- Long distance communication; High deployment cost of fiber WAN; NAN Physical network infrastruc- Up to 100 10-60 km
nications ultra-high bandwidth; robustness installation; high cost of terminal ture Tbps
against radio and electromag- equipment; difficult to upgrade;
netic interference; high reliabil- not suitable for metering appli-
ity cations

Advantages: The characteristics of 6LoWPAN make it ideal short range and low data transmission rate communication
for SG HAN including home automation and smart meter- technology, therefore it is not much suitable for NAN and
ing [131]. It is very robust and low-power communication WAN.
technology which can also support a large mesh network
c) LoRaWAN: LoRAWAN [137] is a Low Power Wide
topology [131]. Additonally, its standard has the liberty of
Area Network (LPWAN) communication technology which is
physical layer and frequency band, and it can be applied across
designed for wireless battery-operated devices in Internet of
various communication platforms (such as IEEE 802.15.4,
Things [138]. It is mainly designed for wide area network
WiFi, Ethernet and sub-1 GHz band).
coverage in IoT and it can be applied in NAN and WAN of
Disadvantages: The main disadvantage of 6LoWPAN is SG. It provides mobility, secure bi-directional communication
the extensive training required to work with IPv6 protocol. and localization services, as well as interoperability. It offers a
The end users cannot deploy this protocol easily because it data rate of 0.3 Kbps to 50 Kbps and coverage range of 2-5km
requires extensive knowledge of IPv6 stack and its functional- in urban environment and 15km suburban environment. Most
ity. Moreover, one of its disadvantages are similar to Z-Wave, of the existing wireless communication technologies have been
i.e., it is designed for small link layer frames, and it has designed to maximize their peak data rate, however LoRa is
 21

mainly designed for low-power consumption and long commu- The main advantage of Bluetooth is its low power usage,
nication range by sacrificing its peak data rate [139]. Further- which is useful for HANs, as well as for online monitoring
more, LoRaWAN network server manages the RF output and of substation automation [142]. Its main disadvantages are its
data rate for each IoT device through adaptive data rate (ADR) low data rate, short range and weak security. Furthermore,
approach in order to enhance network capacity and battery life Bluetooth configured devices may interfere with WLANs
of each IoT device [137]. It is comprised of wide spectrum (Wireless Local Area Networks) due to their high influence
bands, there is no interference with different data rates during on the surrounding communication links.
the communication. Furthermore, it creates virtual channels b) Cellular Communications: Cellular networks can be a
which enhances the capacity of gateways. Additioanlly, it good choice for communication between smart meters and util-
provides low-power WAN coverage with features required to ity centers in an SG. The use of existing cellular communica-
support low-cost secure bi-directional communication [135]. tion infrastructures save the costs and time of the utility centers
d) ZigBee: ZigBee is a cost effective, reliable, low- to build a private and dedicated communication infrastructure
power, low complexity and low data rate wireless commu- for a SG. A number of cellular communication technologies,
nication technology, developed by the ZigBee Alliance. It is including 2G, 3G, LTE and WiMAX are available to utility
suitable for home automation, energy monitoring, smart light- centers for smart metering deployments in SG. Currently,
ning and automatic meter reading in SGs. ZigBee has been cellular networks have been used in SG by a number of
considered as the most suitable communication technology companies. For instance, Echelon’s networked energy services
for HANs in SGs by the National Institute for Standards system uses T-Mobile’s Global System for Mobile (GSM)
and Technology (NIST) [140]. Many AMI vendors, such as communications for its deployment by integrating T-Mobile’s
Elster, Itron and Landis Gyr, integrate ZigBee with smart SIM (Subscriber Identity Module) into Echelon’s smart meters
meters for communication with IoT devices that control the to enable communication between smart meters and the back-
home appliances in HANs [71]. ZigBee is also used to inform haul servers [125]. Since T-Mobile’s GSM network handles all
consumers about their real-time energy consumption. the communication requirements of smart metering in the SG,
Advantages: ZigBee operates on the 868 MHz, 915 MHz investments in a new dedicated communication network are
bands and also uses 16 channels in the 2.4 GHz band (each not required by the utility centers. Similarly, Telecom Italia,
having 5 MHz of bandwidth). It provides compatibility with Telenor, Vodafone and China Mobile have also offered their
the IEEE 802.15.4 standard and offers a data rate of 250 GSM networks for communication in SG.
Kbps with coverage of 10-100m. It supports tree, mesh and Advantages: Since cellular networks have existing infras-
star topologies and uses 128-bit AES encryption for security. tructure, utilities do not have to incur the additional costs of
Due to its lightweight nature, ZigBee is a good choice for building the dedicated communication infrastructure for SG.
metering management, it is considered as a good choice for the Due to their ubiquity and cost effectiveness, cellular network
SG implementation in HANs due to its simplicity, robustness, is the leading communication technology for the SG. Further-
mobility, low operational cost and low bandwidth requirement. more, since GSM capacity is now progressively freed up due
Disadvantages: ZigBee has some limitations for practical to the migration to VoLTE (Voice over LTE), mobile operators
implementations, such as low memory and processing capabil- can offer such solutions for a relatively low cost. In SG, the
ities, high latency and interference with other devices due to its data collection by IoT devices is performed at short intervals,
wireless medium because multiple devices operate on the same which generates a huge amount of data in short bursts, and so
spectrum [140]. Such interference in the vicinity of ZigBee can cellular networks provide sufficient bandwidth to handle such
damage the communication in the whole SG network [141]. data. Moreover, cellular networks also ensure the security of
Therefore, interference avoidance and detection schemes must data transmission security in SG systems. Since the cellular
be designed, as well as energy efficient routing protocols in network coverage has reached almost 100% in the developed
order to provide reliable performance and to maximize the world, cellular networks can provide the best coverage for
network lifetime. SGs. Additionally, lower maintenance and deployment costs,
and rapid installation make cellular networks a better choice
for SG applications, such as AMI, HAN, outage management
B. Non-IoT Technologies and demand response management.
1) Wireless Technologies: Disadvantages: The cellular networks’ services are shared
a) Bluetooth: Bluetooth is a part of the Wireless Personal among multiple customers which may cause network conges-
Area Network (WPAN) standard, known as IEEE 802.15.1, tion and performance deterioration in emergency situations,
which is a low power, short range radio frequency com- presenting a severe challenge for mission critical applications
munication standard. It provides a data rate of 721 Kbps of SG which require continuous communications. In unfavor-
and operates on the 2.4 - 2.4835 GHz unlicensed Industrial, able climate conditions (e.g., wind storms), the cellular net-
Scientific and Medical (ISM) band. Bluetooth can facilitate works may be unable to provide guaranteed service, compared
point-to-point and point-to-multipoint communication config- to wired networks. The utilities therefore, have to build their
urations. It offers coverage between 1m - 100m, depending own private communication networks in order to fulfill the
on the communication configuration. Furthermore, Bluetooth- communication needs of mission critical SG applications.
configured devices are comprised of the whole seven-layered c) Wireless Mesh: A wireless mesh network comprises
OSI communication stack. of a group of wireless nodes where new nodes can join the
 22

network and each node can act as an router. If any node leaves It provides a real time data rate of 1 Mbps up to a high speed
the network, the self healing characteristic of the network data rate of 20 Mbps.
allows it to establish an alternative route through the active MBWA’s main advantages are its high mobility support,
nodes. A wireless mesh network was applied in a SG by the high bandwidth and low latency. It can be used in NANs and
Pacific Gas, SkyPilot Networks and Electric Company in a WANs for SG applications, such as broadband communication
smart metering system. In such a system, each smart device for electric vehicles, SCADA systems and wireless backhaul
is equipped with a radio module and each device routes the for SG monitoring. Its main disadvantage is that since it is
metering data through neighboring meters. In this manner, a new technology, there is no communication infrastructure
each meter acts as a relay until the collected data reaches readily available for this technology. Hence, the use of this
the electric network access point. The collected data are then technology would be costly, compared to other available
tranmitted to the utility. technologies.
Advantages: The wireless mesh networking is an effective f) Digital Microwave Technology: Digital microwave
solution with low cost, self healing, self organization, self technology operates on the 2 - 40 GHz licensed bands and
configuration and high scalability properties. It enhances the provides a data rate of up to 155 Mbps. It offers long distance
network performance, balances the network load, and provides and wide coverage of up to 60 km, and accepts data from
scalability, and network coverage range [143]. Urban and ATM (Asynchronous Transfer Mode) or Ethernet ports and
suburban areas can be well-covered with the help of a multi- forwards it as microwave radio.
hop routing technique. Since smart meters act as relay nodes, a The main advantages of digital microwave technology are
higher number of relay nodes maximizes the network coverage its long distance coverage and high bandwidth. It can support
and capacity. Wireless mesh networking is highly suitable for point-to-point communications for SG applications, such as
home energy management and AMI applications of SGs. transfer trips between Distributed Energy Resources (DERs)
Disadvantages: The major challenges of wireless mesh and a distribution substation [126]. Its main disadvantage is
networking are their network capacity, fading and interference. that it is prone to two types of signal fading, namely multipath
Wireless mesh networks are difficult to implement in rural interference and precipitation. Moreover, it incurs additional
areas because the smart meters’ density is not sufficient to latency due to its encryption of messages for security, which
cover the whole communication network. Moreover, as data makes the messages larger.
travels though multiple relay nodes, there could be a routing 2) Wired Technologies:
looping problem which would cause additional overhead and a) Powerline Communication (PLC): PLC uses existing
affect the available bandwidth [141]. powerlines to transmit high speed data of about 2-3 Mbps
d) WiMAX: WiMAX (worldwide interoperability for mi- from one device to another. PLC has a direct connection
crowave access) is based on the IEEE 802.16 standard for to the meter, and therefore, it was the first communication
Wireless Metropolitan Area Networks (WMAN) with the main technology for electricity meters [141]. In a PLC network,
goal of achieving worldwide interoperability for microwave the data from smart meters is transmitted through a PLC to a
access. WiMAX has its own spectrum bands for interoper- data concentrator, as they are directly connected to each other.
ability. It has been assigned the 3.5 GHz and 5.8 GHz bands Subsequently, the data from data concentrators is transmitted
for fixed communication, while the 2.3 GHz, 2.5 GHz and to a utility control center using other wireless communication
3.5 GHz bands are reserved for mobile communications. The technologies (such as cellular networks), as they are not
spectrum bands 2.3 GHz, 2.5 GHz and 3.5 GHz are licensed directly connected to each other through powerlines [125].
spectrum, while the 5.8 GHz spectrum band is unlicensed. It Advantages: PLC reduces the installation cost by using
provides up to 70 Mbps data rate and 48km coverage. The the existing powerline infrastructure instead of developing a
licensed spectrum bands are more suitable for long distance private communication infrastructure. The strengths of PLC
communications as they allow for long range transmission and are its cost-effectiveness (by reusing existing infrastructure),
high power. widely available infrastructure, standardization efforts and
WiMAX’s main advantages are the long range and the high ubiquitous nature. HAN is one of the largest applications of
data rate. It is mainly applied in HANs and AMI, for the real PLC technology. Since the PLC infrastructure broadly covers
time pricing, automatic meter reading, and outage detection almost all the urban areas in the range of a utility company’s
and restoration. Its two main disadvantages include costly Ra- service territory, it can be suitable for SG applications in urban
dio Frequency (RF) hardware for WiMAX towers and a trade- areas.
off between network performance and distance. Furthermore, Disadvantages: The powerline transmission medium of the
high WiMAX frequencies cannot penetrate through obstacles PLC technology is noisy and harsh, increasing the challenges
and lower frequencies have been already assigned, and so of channel modeling [71]. Moreover, the quality of signals are
require leasing from third-parties. adversely affected by using powerlines in SGs due to various
e) Mobile Broadband Wireless Access: Mobile Broad- parameters, such as the number and types of IoT devices
band Wireless Access (MBWA) technology, also known as connected through powerlines, the wiring distance between
MobileFi, is based on the IEEE 802.20 standard. It provides transmitters and receivers, and the network topology [125].
high mobility, high bandwidth and low latency. It operates on b) Digital Subscriber Lines (DSL): DSL is a high speed
the 3.5 GHz licensed band by exploiting the features of the digital data wired transmission technology which uses tele-
IEEE 802.11 WLANs as well as the IEEE 802.16 WMANs. phony networks land wires. A number of SG projects have
 23

selected DSL as their communication technology. As an ex- to broadband networks [146].

ample, an SG project was carried out at Deutsche Telekom
for the Stadtwerke Emden municipal utility in Germany in C. Summary and Insights
which Deutsche Telekom was responsible for providing data
communications to electric and gas meters. A communication In this section, we have discussed several IoT and non-IoT
box was installed at each consumer’s premises, which then communication technologies for SG. The main motivation of
transmitted the consumption data to Stadtwerke Emden utility discussing these communication technologies for SG is to pro-
over DSL. Deutsche Telekom offered a number of services vide some guidelines for the selection of communication tech-
in this project, such as installation and operation, analyzing nologies of SG based on the requirements. For this purpose,
consumption data and data transmission. we have discussed which communication technologies are
Advantages: DSL reduces the installation cost by exploiting preferred for which scenarios, along with their characteristics,
the existing broadband infrastructure. The main strengths of advantages and disadvantages. We came to know that most
DSL technology are its wide availability (in the developed of the communication technologies are designed by focusing
world), high bandwidth data transmission and low cost. These HAN. However, there is a new IoT communication technology,
strengths make it suitable for electric utilities to implement named LoRaWAN, which is a very good candidate for NAN
SGs with smart metering and data transmission applications. and WAN and it is a long range and low-power communication
Disadvantages: DSL technology requires installation and technology. Furthermore, it is important to note that there
maintenance of communication cables which makes it difficult is no overall the best technology, but certain ones are more
to implement in rural areas due to the high cost of installation suitable to particular SG applications than others. In general,
and maintenance of fixed infrastructure for low density areas. wired technologies such as DSL, PLC and optical fiber are
Additionally, the quality of the DSL connection depends upon expensive for wide area deployments, especially in rural areas.
the distance between the subscriber and the serving telephone However, they can maximize both communication capacity
exchange, which makes it difficult to characterize the per- and security. Wireless technologies, on the other hand, can
formance of DSL technology. Also, due to its unreliability reduce installation costs but they have bandwidth and security
and down time, it is not appropriate for mission critical limitations.
applications.
c) Optical Communications: Optical communication VIII. O PEN I SSUES , C HALLENGES , AND F UTURE
technologies have been widely used by electric utilities in R ESEARCH D IRECTIONS
the last decade by building the communication backbone In this section, we highlight open issues, challenges and
for connecting substations with control centers. Fiber optic future research directions which are also categorized in Figure
communications play a significant role in SGs. Recently, the 14.
usage of optical fibers in SGs has been proposed to provide SG
services directly to consumers [144], [145]. Furthermore, the
Ethernet Passive Optical Network (EPON) technology enables A. Physical Layer
the use of a standard Ethernet communication protocol over 1) Adverse Environmental Conditions and Constrained De-
an optical network, and it is therefore gaining considerable vices: IoT-aided SG systems operate under different envi-
attention from grid operators because of its interoperability ronments including some very severe conditions, such as
with existing IP-based networks [146]. the monitoring of power transmission lines. Therefore, it is
Advantages: There are three main advantages of optical important to consider requirements for reliability, availability,
communication technology. Firstly, it can transmit data packets compatibility for hybrid communication technologies and sig-
over long distances (about several kilometers) by providing naling coverage at adverse environmental conditions [96]. IoT
up to tens of Gbps. Secondly, it is robust against radio solutions for self-healing and self-organization should also be
and electromagnetic interference, which makes it a desirable considered. As an example, when a set of IoT devices fail,
candidate for high-voltage environments. Thirdly, a special an alternative route should be selected by the self-healing
type of optical cables, known as Optical Power Ground Wire capability, so that the reliability of IoT-aided SG systems
(OPGW), combines the features of optical and grounding should not be compromised.
communications which allows long distance transmission at The limitations of IoT-aided SG systems also include the
high data rates. Hence, OPGW can be used for building constrained devices that are used. Such devices may be
transmission and distribution lines [146]. ruggedized for adverse conditions, but they lack memory and
Disadvantages: Although optical communications are very processing power that limits their ability to perform local
favorable for SG, they do have some limitations. The main functions. The lifecycle for such devices tends to be rather
disadvantage of optical communications is the high deploy- long (up to 10 years), so backward compatibility remains a
ment cost of the fiber installation and the terminal units. serious issue.
Furthermore, they are very difficult to upgrade. Optical net- 2) Energy Acquisition: IoT end devices and sensors operate
works, which provide high quality wideband, are not suit- on batteries in many applications of IoT-aided SG systems. For
able for metering applications, because metering devices are example, the online monitoring of power transmission lines
memory-constrained, requiring only narrow band communica- includes various sensors, video cameras and backbone nodes
tion. Therefore, these devices are not likely to be connected installed on the transmission towers and transmission lines,
 24

Fig. 14. Layered-wise classification of open issues, challenges and future directions for IoT-aided SG systems.

which normally operate on batteries. Therefore the energy appliance scheduling towards automated power management
acquisition for power consumption of these devices is a serious and mission-critical power supply, the reliability and speed
problem in realizing the application of IoT technology in SG of the communication become paramount. As Section VII
[50], [96]. For this purpose, efficient energy storage sources discusses several IoT communication networks, IoT-aided SG
for IoT devices, and energy generation devices coupled with systems utilize a hybrid combination of communication net-
energy harvesting using energy conversion need to be used works at different stages in the same operation, a situation
designed [57]. Already, the new generation of batteries can that is not common in other fields. This communication path
boast over ten years of life [147] at reasonable usage. However, goes through several ‘legs’, from devices to local network,
there are still limitations on the level of power usage at the then to gateways, then to core servers and potentially to the
device level. Cloud too, using different protocols each time. This means
3) Environmentally Safe Devices: IoT devices in some that more elaborate applications of IoT-aided SG systems need
SG applications are deployed outdoors and at severe electro- to be assured of solid network support, despite fragmented
magnetic conditions, such as power generation, transmission communication phases, and multiple network providers.
and distribution substations. Therefore, it is very crucial to 2) Data Fusion: Data fusion is the process of combining
protect the instruments and to consider and embed new tech- data from multiple sources. The IoT devices in IoT-aided SG
nologies, such as dustproof, waterproof, anti-electromagnetic, systems are resource constrained and have limited battery life,
anti-vibration, low temperature and high temperature in the processing, bandwidth and storage capabilities. Therefore, it
manufacturing of IoT devices and their chips that will prolong is not efficient for IoT devices to transmit all the data to a
their lifetime under such severe environments [50]. gateway in the process of data collection, as it would consume
significantly high energy and bandwidth. It is desirable to use
B. Network Layer data fusion technologies to filter and aggregate only useful data
from multiple IoT devices, which will enhance the efficiency
1) Communication Networks: Information and communi-
of data collection, as well as save energy and bandwidth [96].
cation networks are very important for the realization of
The techniques of identifying the significant data, such as
transmission and collaboration of IoT devices in IoT-aided
smart aggregation, is a new field that will undoubtedly impact
SG systems. The information and communication networks
IoT-aided SG systems.
can be divided into two main categories based on the range
of transmission [96], namely wide area and short range com-
munication networks. In wide area communication network, C. Transport Layer
the long distance information transmission is achieved through 1) Congestion: Congestion causes delay and packets loss,
IP-based Internet, PLC, OPLC, 2G/3G mobile networks, LTE which are the important performance parameters for SG [148].
and satellite networks. In short range communication net- An input from a NAN gateway can be delayed or missed by the
work, Bluetooth, ZigBee (IEEE 802.15.4) and Ultra-Wideband control center due to congestion, which may affect important
(UWB) are used. As an example, the wireless sensor nodes in decision making by the control center, thus causing perfor-
IoT-aided SG systems have the characteristics of low power, mance degradation or non-fulfillment of the user requirements.
low rate and short distance with the limitations of storage It may also be possible that a message could be dropped by a
and processing capabilities. Hence, ZigBee is an appropriate HAN gateway due to congestion and its memory overload if
communication network in this scenario. As IoT-aided SG multiple messages arrive to the HAN gateway from a higher
applications progress from merely informative facilities and number of IoT devices simultaneously. In such scenarios, the
 25

IoT devices have to retransmit the packets on the request of systems. However, standardization of IoT data collection under
the HAN gateway upon the expiration of the acknowledge- OneM2M [65], [66] is in full swing, but the energy industry
ment interval at HAN gateway, which contributes to higher regards it as an overkill and unsuitable for constrained devices.
delay. Since some SG applications are delay sensitive, it is OMA’s standard for LWM2M (Lightweight M2M) [67] is
necessary to minimize the communication delay in IoT-aided gaining greater popularity due to its greater simplicity, which
SG systems. For many IoT-aided SG systems, it is essential to is required for such applications.
accommodate a higher number of simultaneous messages from De facto standards emerge ’organically’ in the web world
multiple IoT devices without causing a major impact on delay by wide adoption of what is usually open source or freely
and packets loss. This means not only high bandwidth, but available software components. This may eventually occur
also minimizing the number of messages from multiple IoT for IoT-aided SG systems. However, the security requirements
devices to each HAN gateway, i.e., careful network design and need more urgent solutions.
optimal numbers of nodes and gateways [149].
F. Security Issues
D. Application Layer Although IoT technologies have been widely applied in
1) Interoperability and Integration of Devices: Interoper- SG, these could lead to various security vulnerabilities. Since
ability is defined as the ability of two or more heterogeneous the monitoring and controlling in IoT-aided SG systems is
networks/devices to exchange information between them, and performed over the open Internet, it has no more than the
to use the exchanged information in a common function [57]. Internet-based security, which has inferior security than the
The IoT-aided SG system is comprised of a large number managed mobile and fixed networks. The Internet is far more
of different types of IoT devices and gateways which vary vulnerable to cyber attacks. By manipulating the data either
in characteristics, operation, resources (such as computation generated by smart objects or sent from the utility, an attacker
power, memory, energy, bandwidth, time sensitivity), as well can affect the real-time balance between energy production
as their implemented communication stacks and protocols (for and consumption, and cause considerable financial loses to the
non IP-based devices). The lack of device interoperability and utility and power assets [150]. For instance, very recently, the
integration imposes a serious constraint on the development of management of contingencies in smart grid through Internet
IoT-aided SG systems. One possible solution proposed in [1] is of Things have been proposed [151]. Therefore, security
to convert the networks based on proprietary protocols into IP- considerations for IoT-aided SG systems is a high priority.
based networks for the realization of IoT-aided SG systems. Such considerations consist of several aspects are discussed
This will enable the SG to get benefits from the seamless below and some calibrated security measures for centralized
integration of various types of networks/devices for achieving IoT-aided SG systems are discussed in [152].
1) Resource Constrained: In IoT-aided SG system, sev-
interoperability. Moreover, future IoT devices should integrate
eral IoT devices and smart objects are resource constrained,
different communication protocols and standards which oper-
specially those that are deployed in great numbers. Since
ate at different frequencies and allow different architectures
they have limited computational and storage capacities, they
to communicate with other networks [57]. Additionally, the
cannot run complex security algorithms. This constraint makes
interoperability issues need to be differentiated at different
the application of classical security solutions (such as public
levels, such as in the communication layer, the physical layer
key infrastructure (PKI) and public key cryptography) more
or the application layer. A holistic approach to services,
challenging [150]. Hence, there is a need to take special care
devices and semantics should be addressed for solving the
while developing security solutions for IoT-aided SG systems
interoperability in IoT-aided SG systems [57].
to ensure that the resource constrained IoT devices are able to
2) Huge Data Handling: The integration of IoT technology
accommodate the proposed security solution.
with SG comes with the cost of more frequent processing
2) Privacy: The smart meters and the appliances in the
and storing the huge volume of data which would impose
houses could provide information much more than just energy
a higher load on the IoT communication networks. Such
consumption, such as the habits of consumers (wake up,
data includes energy consumption, consumers load demand,
sleeping, lunch and dinner timings), whether they are away
advanced metering records, power lines faults etc. Using high
from premises, or on vacation. Such information can be used
bandwidth and data rates, such as the ones offered by LTE, in-
for marketing (e.g., timing of adverts) but in the wrong hands,
creases the ability to transport such data but creates bottleneck
it can be used for burglaries, for example. Utilities must guar-
elsewhere. Consequently, the utility companies should need to
antee that such private user data should not be obtained without
design systems with enhanced capabilities to store, manage
the users’ approval and that such data would only be used
and process the collected data efficiently and effectively [111].
for the intended purpose [153]. A secure and differentiated
access to data for IoT-aided SG systems is proposed in [100]
E. Miscellaneous Issues to ensure privacy. In this design, the consumers have complete
1) Need of Standardization: Standardization is important fine-grained access to their data while the distributors and
for interoperability, compatibility, reliability and security. Al- energy utilities receive only aggregated statistical data and
though there have been separate investigations on the stan- have coarse-grained access to consumers data. Addtionally,
dardization activities of IoT [26], [27] and SG [70], there some other privacy considerations for IoT-aided SG systems
is no standardization activity specifically for IoT-aided SG are discussed in [154].
 26

3) Trust Management: A certain level of trust establish- IX. C ONCLUSION

ment is required for two devices to communicate. It is easy Smart Grid (SG) is the future grid which solves the
to establish a trust relationship between devices which are problems of uni-directional information flow, energy wastage,
owned/managed by the same entity, but not where they are growing energy demand, reliability and security in the tradi-
managed by different entities, such as the consumers for tional power grid. The Internet of Things (IoT) technology
appliances and operators for smart meters. In a large scale provides connectivity anywhere and anytime. It helps SG
IoT-aided SG system, it is very challenging to establish trust by providing smart devices or IoT devices (such as sensors,
between IoT devices which are owned/managed by different actuators, and smart meters) for the monitoring, analysis and
entities [150]. controlling the grid, as well as connectivity, automation and
4) Authentication and Authorization: The energy provider tracking of such devices. This realizes the IoT-aided SG system
has to authenticate each smart meter in order to bill the corre- which supports and improves various network functions at the
sponding consumer. The identity of the IoT device in an SG is power generation, transmission, distribution, and utilization.
also authenticated in order to avoid any misuse of the system. In this paper, we have presented a comprehensive survey
Only the authenticated user or device should be authorized on IoT-aided SG systems. We have surveyed architectures
to accomplish the tasks or is granted the required privileges and applications of IoT-aided SG systems. We discussed
to access the resources. For example, the configuration of a existing and potential applications of IoT-aided SG systems.
smart meter can only be done by the field agent, who should We also presented various IoT and non-IoT communication
be authorized and granted privileges [150]. technologies for SG systems by presenting their advantages,
5) Data Integrity: Data integrity is also very important to disadvantages, and applicability. Additionally, some existing
ensure that the received data from IoT devices (such as smart prototypes of IoT-aided SG systems were surveyed which can
meters) cannot be modified by an unauthorized party. In IoT- serve as a baseline for future research in this area. Since
aided SG systems, the IoT devices generally communicate IoT-aided SG systems can generate huge amount of data,
using the open Internet. Therefore, data exchanges can be therefore, we also surveyed and provided solutions for big data
easily comprised by the attacker. Energy consumption of processing in IoT-aided SG systems. We concluded the survey
households can be altered by an attacker and can also modify by presenting open issues, challenges, and future research
the data exchanges in SG systems to lower the prices during directions for IoT-aided SG systems.
the peak hours, for example. This would significantly increase
the energy consumption of households (such as charging R EFERENCES
cars) instead of minimizing it which will eventually result
[1] S. E. Collier, “The Emerging Enernet: Convergence of the Smart Grid
in an overloaded power network. Therefore, data integrity with the Internet of Things,” in Rural Electric Power Conference
in IoT-aided SG systems ensures that consumers are charged (REPC), 2015, pp. 65–68.
according to their exact energy consumption [150]. [2] R. Deng, Z. Yang, M.-Y. Chow, and J. Chen, “A Survey on Demand
Response in Smart Grids: Mathematical Models and Approaches,”
6) Cyber Attacks: SG involves various physical objects, IEEE Transactions on Industrial Informatics, vol. 11, no. 3, pp. 570–
such as smart meters, cables and transformers that are man- 582, 2015.
aged by IoT. Hence, SG is vulnerable to cyber attacks which [3] S. Temel, V. C. Gungor, and T. Kocak, “Routing Protocol Design
Guidelines for Smart Grid Environments,” Computer Networks, vol. 60,
could subvert the management and cause indirect damage to pp. 160–170, 2014.
these assets. Attacks can prevent smooth operations, and in- [4] R. Ma, H.-H. Chen, Y.-R. Huang, and W. Meng, “Smart Grid Com-
terfere with maintenance procedures. Attacks can also prevent munication: Its Challenges and Opportunities,” IEEE Transactions on
Smart Grid, vol. 4, no. 1, pp. 36–46, 2013.
the billing process to proceed, and the energy supply to be [5] W. Wang, Y. Xu, and M. Khanna, “A Survey on the Communication
balanced. The more IoT-aided SG applications are developed, Architectures in Smart Grid,” Computer Networks, vol. 55, no. 15, pp.
the greater the need to prevent cyber attacks. 3604–3629, 2011.
[6] E. Yaacoub and A. Abu-Dayya, “Automatic Meter Reading in the Smart
7) Scalability: The IoT-aided SG systems consist of a large Grid using Contention Based Random Access over the Free Cellular
number of IoT devices and smart objects which are installed Spectrum,” Computer Networks, vol. 59, pp. 171–183, 2014.
over large areas which may comprise of few cities in a country. [7] M. Yigit, V. C. Gungor, and S. Baktir, “Cloud Computing for Smart
Grid Applications,” Computer Networks, vol. 70, pp. 312–329, 2014.
Hence, scalability is a serious challenge. [8] H. Sun, A. Nallanathan, B. Tan, J. S. Thompson, J. Jiang, and H. V.
8) Confidentiality: SG data should be protected from pry- Poor, “Relaying Technologies for Smart Grid Communications,” IEEE
ing eyes of unauthorized parties. Confidentiality means that the Wireless Communications, vol. 19, no. 6, pp. 52–59, 2012.
[9] S. Bush, “Network Theory and Smart Grid Distribution Automation,”
stored and transmitted data is only accessible to the concerned IEEE Journal on Selected Areas in Communications, vol. 32, no. 7,
persons. As an example, the energy consumption data of pp. 1451–1459, 2014.
consumers should not be accessible to anyone except the SG’s [10] W. Meng, R. Ma, and H.-H. Chen, “Smart Grid Neighborhood Area
Networks: A Survey,” IEEE Network, vol. 28, no. 1, pp. 24–32, 2014.
operator and the energy providers’ appropriate departments. [11] Smart Grid News, “Smart Grid 101: The Internet of Things and the
9) Identity Spoofing: In identity spoofing, an attacker takes Smart Grid (Part 1),” 12 November 2013, accessed: January
an identity of a legitimate user/thing and uses it for com- 2016. [Online]. Available: http://www.smartgridnews.com/story/
smart-grid-101-internet-things-and-smart-grid-part-1/2013-11-12
munication on its behalf in an unauthorized manner. In IoT- [12] A. Al-Fuqaha, M. Guizani, M. Mohammadi, M. Aledhari, and
aided SG sytems, the identity of a smart meter at someone’s M. Ayyash, “Internet of Things: A Survey on Enabling Technologies,
home can be spoofed by the attacker and in this manner, the Protocols and Applications,” IEEE Communications Surveys & Tutori-
als, 2015.
legitimate user would be charged for energy consumption of [13] E. Borgia, “The Internet of Things Vision: Key Features, Applications
the attacker. and Open Issues,” Computer Communications, vol. 54, pp. 1–31, 2014.
 27

[14] I. Mashal, O. Alsaryrah, T.-Y. Chung, C.-Z. Yang, W.-H. Kuo, and [36] N. Komninos, E. Philippou, and A. Pitsillides, “Survey in Smart Grid
D. P. Agrawal, “Choices for Interaction with Things on Internet and and Smart Home Security: Issues, Challenges and Countermeasures,”
Underlying Issues,” Ad Hoc Networks, vol. 28, pp. 68–90, 2015. IEEE Communications Surveys & Tutorials, vol. 16, no. 4, pp. 1933–
[15] M. Nitti, V. Pilloni, G. Colistra, and L. Atzori, “The Virtual Object 1954, 2014.
as a Major Element of the Internet of Things: a Survey,” IEEE [37] S. Finster and I. Baumgart, “Privacy-Aware Smart Metering: A Survey,”
Communications Surveys & Tutorials, 2015. IEEE Communications Surveys & Tutorials, vol. 17, no. 2, pp. 1088–
[16] J. Granjal, E. Monteiro, and J. Silva, “Security for the Internet of 1101, 2015.
Things: A Survey of Existing Protocols and Open Research Issues,” [38] Q. Sun, H. Li, Z. Ma, C. Wang, J. Campillo, Q. Zhang, F. Wallin, and
IEEE Communications Surveys & Tutorials, vol. 17, no. 3, pp. 1294– J. Guo, “A comprehensive review of smart energy meters in intelligent
1312, 2015. energy networks,” IEEE Internet of Things Journal, in Print, 2016.
[17] J. Granjal, E. Monteiro, and J. S. Silva, “Security in the Integration of [39] L. Hernandez, C. Baladron, J. M. Aguiar, B. Carro, A. J. Sanchez-
Low-Power Wireless Sensor Networks with the Internet: A Survey,” Esguevillas, J. Lloret, and J. Massana, “A Survey on Electric Power
Ad Hoc Networks, vol. 24, pp. 264–287, 2015. Demand Forecasting: Future Trends in Smart Grids, Microgrids and
[18] S. Sicari, A. Rizzardi, L. Grieco, and A. Coen-Porisini, “Security, Smart Buildings,” IEEE Communications Surveys & Tutorials, vol. 16,
Privacy and Trust in Internet of Things: The Road Ahead,” Computer no. 3, pp. 1460–1495, 2014.
Networks, vol. 76, pp. 146–164, 2015. [40] A. Rahman, X. Liu, and F. Kong, “A Survey on Geographic Load
[19] A. Botta, W. de Donato, V. Persico, and A. Pescape, “Integration of Balancing based Data Center Power Management in the Smart Grid
Cloud Computing and Internet of Things: A survey,” Future Generation Environment,” IEEE Communications Surveys & Tutorials, vol. 16,
Computer Systems, 2015. no. 1, pp. 214–233, 2014.
[20] L. Atzori, A. Iera, G. Morabito, and M. Nitti, “The Social Internet [41] J. S. Vardakas, N. Zorba, and C. V. Verikoukis, “A Survey on Demand
of Things (SIoT)–When Social Networks Meet the Internet of Things: Response Programs in Smart Grids: Pricing Methods and Optimization
Concept, Architecture and Network Characterization,” Computer Net- Algorithms,” IEEE Communications Surveys & Tutorials, vol. 17, no. 1,
works, vol. 56, no. 16, pp. 3594–3608, 2012. pp. 152–178, 2015.
[21] A. M. Ortiz, D. Hussein, S. Park, S. N. Han, and N. Crespi, “The [42] K. Reddy, M. Kumar, T. Mallick, H. Sharon, and S. Lokeswaran, “A
Cluster Between Internet of Things and Social Networks: Review and Review of Integration, Control, Communication and Metering (ICCM)
Research Challenges,” IEEE Internet of Things Journal, vol. 1, no. 3, of Renewable Energy based Smart Grid,” Renewable and Sustainable
pp. 206–215, 2014. Energy Reviews, vol. 38, pp. 180–192, 2014.
[22] K. Sood, S. Yu, and Y. Xiang, “Software Defined Wireless Networking [43] A. A. Khan, M. H. Rehmani, and M. Reisslein, “Cognitive Radio for
Opportunities and Challenges for Internet of Things: A Review,” IEEE Smart Grids: Survey of Architectures, Spectrum Sensing Mechanisms,
Internet of Things Journal, in Print, 2016. and Networking Protocols,” IEEE Communications Surveys & Tutori-
[23] C.-W. Tsai, C.-F. Lai, M.-C. Chiang, and L. T. Yang, “Data Mining als, vol. 18, no. 1, pp. 860–898, 2016.
for Internet of Things: A Survey,” IEEE Communications Surveys & [44] A. Sajid, H. Abbas, and K. Saleem, “Cloud-assisted IoT-Based SCADA
Tutorials, vol. 16, no. 1, pp. 77–97, 2014. systems security: A review of the state of the art and future challenges,”
[24] M. Mehaseb, Y. Gadallah, A. Elhamy, and H. El-Hennawy, “Classifi- IEEE Access, in Print, 2016.
cation of LTE Uplink Scheduling Techniques: An M2M Perspective,” [45] C. Wang, X. Li, Y. Liu, and H. Wang, “The Research on Development
IEEE Communications Surveys & Tutorials, vol. pp, no. 99, pp. 1–1, Direction and Points in IoT in China Power Grid,” in International
2015. Conference on Information Science, Electronics and Electrical Engi-
[25] C. Perera, A. Zaslavsky, P. Christen, and D. Georgakopoulos, “Context neering (ISEEE), vol. 1, 2014, pp. 245–248.
Aware Computing for the Internet of Things: A Survey,” IEEE Com- [46] X. Chen, L. Sun, H. Zhu, Y. Zhen, and H. Chen, “Application
munications Surveys & Tutorials, vol. 16, no. 1, pp. 414–454, 2014. of Internet of Things in Power-Line Monitoring,” in International
[26] M. R. Palattella, N. Accettura, X. Vilajosana, T. Watteyne, L. A. Conference on Cyber-Enabled Distributed Computing and Knowledge
Grieco, G. Boggia, and M. Dohler, “Standardized Protocol Stack for Discovery (CyberC), 2012, pp. 423–426.
the Internet of (important) Things,” IEEE Communications Surveys & [47] A. Zanella, N. Bui, A. Castellani, L. Vangelista, and M. Zorzi, “Internet
Tutorials, vol. 15, no. 3, pp. 1389–1406, 2013. of Things for Smart Cities,” IEEE Internet of Things Journal, vol. 1,
[27] Z. Sheng, S. Yang, Y. Yu, A. Vasilakos, J. Mccann, and K. Leung, “A no. 1, pp. 22–32, 2014.
Survey on the IETF Protocol Suite for the Internet of Things: Stan- [48] S. Tonyali, O. Cakmak, K. akkaya, M. Mahmoud, and I. Guvenc,
dards, Challenges, and Opportunities,” IEEE Wireless Communications, “Secure data obfuscation scheme to enable privacy-preserving state
vol. 20, no. 6, pp. 91–98, 2013. estimation in smart grid AMI networks,” IEEE Internet of Things
[28] A. Gluhak, S. Krco, M. Nati, D. Pfisterer, N. Mitton, and T. Razafind- Journal, in Print, 2016.
ralambo, “A survey on Facilities for Experimental Internet of Things [49] M. Yun and B. Yuxin, “Research on the Architecture and Key
Research,” IEEE Communications Magazine, vol. 49, no. 11, pp. 58– Technology of Internet of Things (IoT) Applied on Smart Grid,” in
67, 2011. International Conference on Advances in Energy Engineering (ICAEE),
[29] Z. Fan, P. Kulkarni, S. Gormus, C. Efthymiou, G. Kalogridis, 2010, pp. 69–72.
M. Sooriyabandara, Z. Zhu, S. Lambotharan, and W. H. Chin, “Smart [50] J. Liu, X. Li, X. Chen, Y. Zhen, and L. Zeng, “Applications of Internet
Grid Communications: Overview of Research Challenges, Solutions, of Things on Smart Grid in China,” in 13th International Conference
and Standardization Activities,” IEEE Communications Surveys & on Advanced Communication Technology (ICACT), 2011, pp. 13–17.
Tutorials, vol. 15, no. 1, pp. 21–38, 2013. [51] M. A. Zaveri, S. K. Pandey, and J. S. Kumar, “Collaborative Service
[30] X. Fang, S. Misra, G. Xue, and D. Yang, “Smart Grid - The New and Oriented Smart Grid using the Internet of Things,” in International
Improved Power Grid: A Survey,” IEEE Communications Surveys & Conference on Communication and Signal Processing (ICCSP), 2016,
Tutorials, vol. 14, no. 4, pp. 944–980, 2012. pp. 1716–1722.
[31] I. Al-Anbagi, M. Erol Kantarci, and H. T. Mouftah, “A Survey [52] K. Samarakoon, J. Ekanayake, and N. Jenkins, “Reporting Available
on Cross-layer Quality of Service Approaches in WSNs for Delay Demand Response,” IEEE Transactions on Smart Grid, vol. 4, no. 4,
and Reliability Aware Applications,” IEEE Communications Surveys pp. 1842–1851, 2013.
Tutorials, vol. 18, no. 1, pp. 525–552, 2016. [53] A. A. Khan and H. T. Mouftah, “Web Services for Indoor Energy
[32] M. Erol-Kantarci and H. T. Mouftah, “Energy-Efficient Information Management in a Smart Grid Environment,” in 22nd International
and Communication Infrastructures in the Smart Grid: A Survey Symposium on Personal Indoor and Mobile Radio Communications
on Interactions and Open Issues,” IEEE Communications Surveys & (PIMRC), 2011, pp. 1036–1040.
Tutorials, vol. 17, no. 1, pp. 179–197, 2015. [54] F. Benzi, N. Anglani, E. Bassi, and L. Frosini, “Electricity Smart
[33] J. Gao, Y. Xiao, J. Liu, W. Liang, and C. P. Chen, “A Survey Meters Interfacing the Households,” IEEE Transactions on Industrial
of Communication/Networking in Smart Grids,” Future Generation Electronics, vol. 58, no. 10, pp. 4487–4494, 2011.
Computer Systems, vol. 28, no. 2, pp. 391–404, 2012. [55] Q. Hu and F. Li, “Hardware Design of Smart Home Energy Manage-
[34] H. Liang, A. K. Tamang, W. Zhuang, and X. S. Shen, “Stochastic In- ment System with Dynamic Price Response,” IEEE Transactions on
formation Management in Smart Grid,” IEEE Communications Surveys Smart Grid, vol. 4, no. 4, pp. 1878–1887, 2013.
& Tutorials, vol. 16, no. 3, pp. 1746–1770, 2014. [56] F. Viani, F. Robol, A. Polo, P. Rocca, G. Oliveri, and A. Massa,
[35] Y. Yan, Y. Qian, H. Sharif, and D. Tipper, “A Survey on Cyber Security “Wireless Architectures for Heterogeneous Sensing in Smart Home
for Smart Grid Communications,” IEEE Communications Surveys & Applications: Concepts and Real Implementation,” Proceedings of the
Tutorials, vol. 14, no. 4, pp. 998–1010, 2012. IEEE, vol. 101, no. 11, pp. 2381–2396, 2013.
 28

[57] W. Shu-wen, “Research on the Key Technologies of IOT Applied on [79] L. Yu, T. Jiang, Y. Cao, and Q. Qi, “Carbon-aware Energy Cost Min-
Smart Grid,” in International Conference on Electronics, Communica- imization for Distributed Internet Data Centers in Smart Microgrids,”
tions and Control (ICECC), 2011, pp. 2809–2812. IEEE Internet of Things Journal, vol. 1, no. 3, pp. 255–264, 2014.
[58] Y. Wang, W. Lin, T. Zhang, and Y. Ma, “Research on Application [80] M. M. Rana and L. Li, “Microgrid State Estimation and Control
and Security Protection of Internet of Things in Smart Grid,” in Inter- for Smart Grid and Internet of Things Communication Network,”
national Conference on Information Science and Control Engineering Electronics Letters, vol. 51, no. 2, pp. 149–151, 2015.
(ICISCE), 2012. [81] V. Balijepalli, V. Pradhan, S. Khaparde, and R. Shereef, “Review
[59] A. Basit, G. A. S. Sidhu, A. Mahmood, and F. Gao, “Efficient of Demand Response under Smart Grid Paradigm,” in IEEE PES
and autonomous energy management techniques for the future smart Innovative Smart Grid Technologies (ISGT), 2011, pp. 236–243.
homes,” IEEE Transactions on Smart Grid, in Print, 2016. [82] P. Y. Kong, “Wireless neighborhood area networks with qos support
[60] D. Mashima and W.-P. Chen, “Residential Demand Response System for demand response in smart grid,” IEEE Transactions on Smart Grid,
Framework Leveraging IoT Devices,” in IEEE International Confer- in Print, 2016.
ence on Smart Grid Communications (SmartGridComm), 2016, pp. [83] P. Siano, “Demand Response and Smart Grids-A Survey,” Renewable
514–520. and Sustainable Energy Reviews, vol. 30, pp. 461–478, 2014.
[61] R. O. Aburukba, A. Al-Ali, T. Landolsi, M. Rashid, and R. Hassan, [84] M. Jaradat, M. Jarrah, Y. Jararweh, M. Al-Ayyoub, and A. Bousselham,
“IoT based Energy Management for Residential Area,” in IEEE In- “Integration of Renewable Energy in Demand-side Management for
ternational Conference on Consumer Electronics-Taiwan (ICCE-TW), Home Appliances,” in International Renewable and Sustainable Energy
2016, pp. 1–2. Conference (IRSEC), 2014, pp. 571–576.
[85] S. Mohanty, B. N. Panda, and B. S. Pattnaik, “Implementation of a
[62] H. Mortaji, O. S. Hock, M. Moghavvemi, and H. A. Almurib, “Smart
Web of Things based Smart Grid to Remotely Monitor and Control
Grid Demand Response Management using Internet of Things for
Renewable Energy Sources,” in Students’ Conference on Electrical,
Load Shedding and Smart-direct Load Control,” in IEEE Industry
Electronics and Computer Science (SCEECS), 2014, pp. 1–5.
Applications Society Annual Meeting, 2016, pp. 1–7.
[86] M. Khan, N. B. Silva, and H. Kijun, “Internet of Things based Energy
[63] Z. Zhu, S. Lambotharan, W. H. Chin, and Z. Fan, “Overview of Aware Smart Home Control System,” IEEE Access, in Print, 2016.
Demand Management in Smart Grid and Enabling Wireless Communi- [87] R. Mukhopadhyay and I. Mukhopadhyay, “Home Automation and Grid
cation Technologies,” IEEE Wireless Communications, vol. 19, no. 3, Mapping Technology using IoT,” in IEEE 7th Annual Information
pp. 48–56, 2012. Technology, Electronics and Mobile Communication Conference (IEM-
[64] D. Bandyopadhyay and J. Sen, “Internet of Things: Applications and CON), 2016, pp. 1–5.
Challenges in Technology and Standardization,” Wireless Personal [88] S. K. Viswanath, C. Yuen, W. Tushar, W.-T. Li, C.-K. Wen, K. Hu,
Communications, vol. 58, no. 1, pp. 49–69, 2011. C. Chen, and X. Liu, “System Design of the Internet of Things
[65] J. Swetina, G. Lu, P. Jacobs, F. Ennesser, and J. Song, “Toward a for Residential Smart Grid,” IEEE Wireless Communications, vol. 23,
standardized common m2m service layer platform: Introduction to no. 5, pp. 90–98, 2016.
onem2m,” IEEE Wireless Communications, vol. 21, no. 3, pp. 20–26, [89] I. A. Shah, F. A. Malik, and S. A. Ahmad, “Enhancing Security in
2014. IoT based Home Automation using Reed Solomon Codes,” in Inter-
[66] “oneM2M - Standards for M2M and the Internet of Things,” accessed: national Conference on Wireless Communications, Signal Processing
May 2016. [Online]. Available: http://onem2m.org/ and Networking (WiSPNET), 2016, pp. 1639–1642.
[67] “OMA LightweightM2M ,” accessed: May 2016. [90] D. Minoli, K. Sohraby, and B. Occhiogrosso, “IoT Considerations,
[Online]. Available: http://technical.openmobilealliance.org/ Requirements, and Architectures for Smart Buildings–Energy Opti-
Technical/technical-information/release-program/current-releases/ mization and Next Generation Building Management Systems,” IEEE
oma-lightweightm2m-v1-0 Internet of Things Journal, vol. 4, no. 1, pp. 269 – 283, 2017.
[68] A. Aijaz and A. H. Aghvami, “Cognitive Machine-to-Machine Commu- [91] L. Yu, T. Jiang, and Y. Zou, “Distributed Online Energy Management
nications for Internet-of-Things: A Protocol Stack Perspective,” IEEE for Data Centers and Electric Vehicles in Smart Grid,” IEEE Internet
Internet of Things Journal, vol. 2, no. 2, pp. 103–112, 2015. of Things Journal, 2016.
[69] S. L. Keoh, S. S. Kumar, and H. Tschofenig, “Securing the Internet [92] A. Ghasempour and T. K. Moon, “Optimizing the Number of Collectors
of Things: A Standardization Perspective,” IEEE Internet of Things in Machine-to-Machine Advanced Metering Infrastructure Architecture
Journal, vol. 1, no. 3, pp. 265–275, 2014. for Internet of Things-based Smart Grid,” in IEEE Green Technologies
[70] J. Bhatt, V. Shah, and O. Jani, “An Instrumentation Engineer’s Review Conference (GreenTech), 2016, pp. 51–55.
on Smart Grid: Critical Applications and Parameters,” Renewable and [93] Y. Ma, F. Liu, X. Zhou, and Z. Gao, “Key Technologies of Smart
Sustainable Energy Reviews, vol. 40, pp. 1217–1239, 2014. Distribution Grid,” in IEEE International Conference on Mechatronics
[71] V. C. Gungor, D. Sahin, T. Kocak, S. Ergut, C. Buccella, C. Cecati, and and Automation (ICMA), 2016, pp. 2639–2643.
G. P. Hancke, “Smart Grid Technologies: Communication Technologies [94] Z. Jingchao, H. Jinhua, M. Linchang, S. Ningxi, and L. Xiaolei, “The
and Standards,” IEEE Transactions on Industrial informatics, vol. 7, Implementation of the Internet of Things Technology in Henan Smart
no. 4, pp. 529–539, 2011. Distribution Network Demonstration Project,” in China International
[72] K. H. Chang, “Interoperable nan standards: a path to cost-effective Conference on Electricity Distribution (CICED), 2016, pp. 1–5.
smart grid solutions,” IEEE Wireless Communications, vol. 20, no. 3, [95] Y. E. Song, Y. Liu, S. Fang, and S. Zhang, “Research on Applications
pp. 4–5, 2013. of the Internet of Things in the Smart Grid,” in 7th International
Conference on Intelligent Human-Machine Systems and Cybernetics
[73] H. Wang, Y. Qian, and H. Sharif, “Multimedia communications over
(IHMSC), vol. 2, 2015, pp. 178–181.
cognitive radio networks for smart grid applications,” IEEE Wireless
[96] Q. Ou, Y. Zhen, X. Li, Y. Zhang, and L. Zeng, “Application of Internet
Communications, vol. 20, no. 4, pp. 125–132, 2013.
of Things in Smart Grid Power Transmission,” in 3rd FTRA Interna-
[74] Y. Zhen, X. Li, Y. Zhang, L. Zeng, Q. Ou, and X. Yin, “Transmission tional Conference on Mobile, Ubiquitous, and Intelligent Computing
Tower Protection System based on Internet of Things in Smart Grid,” in (MUSIC), 2012, pp. 96–100.
7th IEEE International Conference on Computer Science & Education [97] L. Zheng, S. Chen, S. Xiang, and Y. Hu, “Research of Architecture
(ICCSE), 2012, pp. 863–867. and Application of Internet of Things for Smart Grid,” in Computer
[75] X. Chen, J. Liu, X. Li, L. Sun, and Y. Zhen, “Integration of IoT Science & Service System (CSSS), 2012 International Conference on,
with Smart Grid,” in International Conference on Communication 2012, pp. 938–941.
Technology and Application (ICCTA), 2011, pp. 723–726. [98] U. R. Committee, “A National Green Building Research Agenda,” US
[76] F. K. Santoso and N. C. Vun, “Securing IoT for smart home system,” Green Building Council (USGBC), 2007.
in IEEE International Symposium on Consumer Electronics (ISCE). [99] J. Pan, R. Jain, S. Paul, T. Vu, A. Saifullah, and M. Sha, “An Internet of
IEEE, 2015, pp. 1–2. Things Framework for Smart Energy in Buildings: Designs, Prototype,
[77] M. Jaradat, M. Jarrah, A. Bousselham, Y. Jararweh, and M. Al- and Experiments,” IEEE Internet of Things Journal, vol. 2, no. 6, pp.
Ayyoub, “The Internet of Energy: Smart Sensor Networks and Big Data 527–537, 2015.
Management for Smart Grid,” Procedia Computer Science, vol. 56, pp. [100] E. Spano, L. Niccolini, S. Di Pascoli, and G. Iannacconeluca, “Last-
592–597, 2015. Meter Smart Grid Embedded in an Internet-of-Things Platform,” IEEE
[78] X. Li, R. Lu, X. Liang, X. Shen, J. Chen, and X. Lin, “Smart Com- Transactions on Smart Grid, vol. 6, no. 1, pp. 468–476, 2015.
munity: An Internet of Things Application,” IEEE Communications [101] E. Spano, S. Di Pascoli, and G. Iannaccone, “An Intragrid Implemen-
Magazine, vol. 49, no. 11, pp. 68–75, 2011. tation Embedded in an Internet of Things Platform,” in IEEE 18th
 29

International Workshop on Computer Aided Modeling and Design of [127] S. Jain, N. Kumar, A. Paventhan, V. K. Chinnaiyan, V. Arnachalam,
Communication Links and Networks (CAMAD), 2013, pp. 134–138. and M. Pradish, “Survey on Smart Grid Technologies-Smart Metering,
[102] P3 International, “Kill-a-watt Product Series,” accessed: January IoT and EMS,” in Students’ Conference on Electrical, Electronics and
2016. [Online]. Available: http://www.p3international.com/products/ Computer Science (SCEECS), 2014, pp. 1–6.
p4400.html [128] W. Luan, D. Sharp, and S. Lancashire, “Smart Grid Communication
[103] Belkin, “WeMo switch,” accessed: January 2016. [Online]. Available: Network Capacity Planning for Power Utilities,” in IEEE PES Trans-
http://www.belkin.com/us/wemo mission and Distribution Conference and Exposition, 2010, pp. 1–4.
[104] ARM, “LPC1768 Datasheet,” accessed: January 2016. [Online]. Avail- [129] “Z-Wave,” accessed: September 2016. [Online]. Available: http:
able: http://www.keil.com/dd/docs/datashts/philips/lpc176x ds.pdf //www.z-wave.com/
[105] ——, “CMSIS OS,” accessed: January 2016. [On- [130] S. S. I. Samuel, “A Review of Connectivity Challenges in IoT-Smart
line]. Available: http://www.arm.com/products/processors/cortex-m/ Home,” in 3rd MEC International Conference on Big Data and Smart
cortex-microcontroller-software-interface-standard.php City (ICBDSC), 2016, pp. 1–4.
[106] G. Iannaccone, “Fast Prototyping of Network Data Mining Applica- [131] M. Kuzlu, M. Pipattanasomporn, and S. Rahman, “Review of Com-
tions,” in Passive and Active Measurement Conference, 2006. munication Technologies for Smart Homes/Building Applications,” in
IEEE Innovative Smart Grid Technologies-Asia (ISGT ASIA), 2015, pp.
[107] L. Peterson, S. Muir, T. Roscoe, and A. Klingaman, “Planetlab Archi-
1–6.
tecture: An Overview,” PlanetLab Consortium May, vol. 1, no. 15, pp.
[132] C. Withanage, R. Ashok, C. Yuen, and K. Otto, “A Comparison of the
4–1, 2006.
Popular Home Automation Technologies,” in IEEE Innovative Smart
[108] Tornado, “Tornado Documentation Release 3.1.1,” accessed: January Grid Technologies-Asia (ISGT ASIA), 2014, pp. 600–605.
2016. [Online]. Available: http://www.tornadoweb.org/en/stable/ [133] “SIGMA Designs,” accessed: September 2016. [Online]. Available:
[109] M. Otto and J. Thornton, “Twitter Bootstrap,” 2013. [Online]. http://www.sigmadesigns.com/
Available: http://getbootstrap.com/ [134] M. Z. Huq and S. Islam, “Home area network technology assessment
[110] S. C. M. et al., “Interfacing power system and ICT simulators: for demand response in smart grid environment,” in Universities Power
Challenges, state-of-the-art, and case studies,” IEEE Transactions on Engineering Conference (AUPEC), 2010 20th Australasian. IEEE,
Smart Grid, in Print, 2016. 2010, pp. 1–6.
[111] I. Group, “Managing Big Data for Smart Grids and Smart Meters,” [135] “11 Internet of Things (IoT) Protocols You Need to
IBM Corporation, whitepaper, 2012. Know About,” accessed: September 2016. [Online]. Available:
[112] X. He, Q. Ai, R. C. Qiu, W. Huang, L. Piao, and H. Liu, “A big data http://www.rs-online.com/designspark/electronics/knowledge-item/
architecture design for smart grids based on random matrix theory,” eleven-internet-of-things-iot-protocols-you-need-to-know-about
IEEE Transactions on Smart Grid, in Print, 2016. [136] G. Montenegro, N. Kushalnagar, J. Hui, and D. Culler, “Transmission
[113] Y. Sun, H. Song, A. Jara, and R. Bie, “Internet of Things and Big of IPv6 Packets over IEEE 802.15.4 Networks,” Internet Requests
Data Analytics for Smart and Connected Communities,” IEEE Access, for Comments, IETF, RFC 4944, 2007. [Online]. Available:
vol. 4, pp. 766–773, 2016. http://www.rfc-editor.org/rfc/rfc4944.txt
[114] Y. Jararweh, M. Jarrah, Z. Alshara, M. N. Alsaleh, M. Al-Ayyoub [137] “LoRA Alliance,” accessed: September 2016. [Online]. Available:
et al., “CloudExp: A Comprehensive Cloud Computing Experimental https://www.lora-alliance.org
Framework,” Simulation Modelling Practice and Theory, vol. 49, pp. [138] H. Schroder Filho, J. Pissolato Filho, and V. Moreli, “The Adequacy of
180–192, 2014. LoRaWAN on Smart Grids: A Comparison with RF Mesh Technology,”
[115] F. Luo, J. Zhao, Z. Y. Dong, Y. Chen, Y. Xu, X. Zhang, and K. P. Wong, in IEEE International Smart Cities Conference (ISC2), 2016, pp. 1–6.
“Cloud-based information infrastructure for next-generation power [139] J. Petajajarvi, K. Mikhaylov, A. Roivainen, T. Hanninen, and M. Pet-
grid: Conception, architecture, and applications,” IEEE Transactions tissalo, “On the Coverage of LPWANs: Range Evaluation and Channel
on Smart Grid, in Print, 2016. Attenuation Model for LoRa Technology,” in 14th International Con-
[116] Z. Cao, J. Lin, C. Wan, Y. Song, Y. Zhang, and X. Wang, “Optimal ference on ITS Telecommunications (ITST), 2015, pp. 55–59.
cloud computing resource allocation for demand side management,” [140] P. Yi, A. Iwayemi, and C. Zhou, “Developing ZigBee Deployment
IEEE Transactions on Smart Grid, in Print, 2016. Guideline under WiFi Interference for Smart Grid Applications,” IEEE
[117] Y. Saleem, F. Salim, and M. H. Rehmani, “Integration of Cognitive Transactions on Smart Grid, vol. 2, no. 1, pp. 110–120, 2011.
Radio Sensor Networks and Cloud Computing: A Recent Trend,” [141] R. P. Lewis, P. Igic, and Z. Zhou, “Assessment of Communication
Cognitive Radio Sensor Networks: Applications, Architectures, and Methods for Smart Electricity Metering in the UK,” in IEEE PES/IAS
Challenges: Applications, Architectures, and Challenges, p. 288, 2014. Conference on Sustainable Alternative Energy (SAE), 2009, pp. 1–4.
[118] D. S. Markovic, D. Zivkovic, I. Branovic, R. Popovic, and [142] H. Zhang, G. Guan, and X. Zang, “The Design of Insulation Online
D. Cvetkovic, “Smart Power Grid and Cloud Computing,” Renewable Monitoring System based on Bluetooth Technology and IEEE1451. 5,”
and Sustainable Energy Reviews, vol. 24, pp. 566–577, 2013. in International Power Engineering Conference (IPEC). IEEE, 2007,
[119] A. Meloni and L. Atzori, “A Cloud-based and RESTful Internet of pp. 1287–1291.
Things Platform to Foster Smart Grid Technologies Integration and [143] A. Yarali, “Wireless Mesh Networking Technology for Commercial
Re-usability,” in IEEE International Conference on Communications and Industrial Customers,” in Canadian Conference on Electrical and
Workshops (ICC), 2016, pp. 387–392. Computer Engineering (CCECE), 2008, pp. 000 047–000 052.
[144] M. Maier, “Fiber-Wireless Sensor Networks (Fi-WSNs) for Smart
[120] A. Sajid, H. Abbas, and K. Saleem, “Cloud-assisted IoT-based SCADA
Grids,” in 13th International Conference on Transparent Optical Net-
Systems Security: A Review of the State of the Art and Future
works (ICTON), 2011, pp. 1–4.
Challenges,” IEEE Access, vol. 4, pp. 1375–1384, 2016.
[145] L. Jianming, Z. Bingzhen, and Z. Zichao, “The Smart Grid Multi-
[121] F. Bonomi, R. Milito, J. Zhu, and S. Addepalli, “Fog Computing and Utility Services Platform based on Power Fiber to the Home,” in
its Role in the Internet of Things,” in Proceedings of the first edition IEEE International Conference on Cloud Computing and Intelligence
of the MCC workshop on Mobile cloud computing. ACM, 2012, pp. Systems (CCIS), 2011, pp. 17–22.
13–16. [146] E. Ancillotti, R. Bruno, and M. Conti, “The Role of Communication
[122] F. Jalali, A. Vishwanath, J. de Hoog, and F. Suits, “Interconnecting Systems in Smart Grids: Architectures, Technical Solutions and Re-
Fog Computing and Microgrids for Greening IoT,” in IEEE Innovative search Challenges,” Computer Communications, vol. 36, no. 17, pp.
Smart Grid Technologies-Asia (ISGT-Asia), 2016, pp. 693–698. 1665–1697, 2013.
[123] Hadoop MapReduce, accessed: January 2016. [Online]. Available: [147] J. Steiner, M. Blakeley, and A. Miller, “Estimating the battery life of
http://hadoop.apache.org/ a wireless occupancy sensor,” Lutron Electronics Co., Inc. IP/N 367-
[124] Gartner, “Stream Processing is The New Black for Smart Grid 2437 REV A, Tech. Rep., 2014.
Analytics,” accessed: January 2016. [Online]. Available: http://blogs. [148] C. H. Hauser, D. E. Bakken, I. Dionysiou, K. H. Gjermundrod, V. Irava,
gartner.com/chet-geschickter/smart-grid-analytics-stream-processing/ J. Helkey, and A. Bose, “Security, Trust, and QoS in Next-Generation
[125] V. Gungor, D. Sahin, T. Kocak, and S. Ergut, “Smart Grid Communi- Control and Communication for Large Power Systems,” International
cations and Networking,” Turk Telecom, 2011. Journal of Critical Infrastructures, vol. 4, no. 1-2, pp. 3–16, 2008.
[126] P. P. Parikh, M. G. Kanabar, and T. S. Sidhu, “Opportunities and [149] Z. M. Fadlullah, M. M. Fouda, N. Kato, A. Takeuchi, N. Iwasaki, and
Challenges of Wireless Communication Technologies for Smart Grid Y. Nozaki, “Toward Intelligent Machine-to-Machine Communications
Applications,” in IEEE Power and Energy Society General Meeting, in Smart Grid,” IEEE Communications Magazine, vol. 49, no. 4, pp.
2010, pp. 1–7. 60–65, 2011.
 30

[150] C. Bekara, “Security Issues and Challenges for the IoT-based Smart
Grid,” Procedia Computer Science, vol. 34, pp. 532–537, 2014.
[151] S. Ciavarella, J. Y. Joo, and S. Silvestri, “Managing contingencies in
smart grids via the Internet of things,” IEEE Transactions on Smart
Grid, in Print, 2016.
[152] A. K. Koundinya, G. Sharvani, and K. U. Rao, “Calibrated Security
Measures for Centralized IoT Applications of Smart Grids,” in Interna-
tional Conference on Computation System and Information Technology
for Sustainable Solutions (CSITSS), 2016, pp. 153–157.
[153] W. Mesbah, “Securing smart electricity meters against customer at-
tacks,” IEEE Transactions on Smart Grid, in Print, 2016.
[154] F. Dalipi and S. Y. Yayilgan, “Security and Privacy Considerations for
IoT Application on Smart Grids: Survey and Research Challenges,” in
IEEE International Conference on Future Internet of Things and Cloud
Workshops (FiCloudW), 2016, pp. 63–68.