Wireless Sensor Network Based Smart Grid

Communications: Challenges, Protocol Optimizations,

and Validation Platforms
Sana Rekik, Nouha Baccour, Mohamed Jmaiel, Khalil Drira

To cite this version:

Sana Rekik, Nouha Baccour, Mohamed Jmaiel, Khalil Drira. Wireless Sensor Network Based Smart
Grid Communications: Challenges, Protocol Optimizations, and Validation Platforms. Wireless Per-
sonal Communications, Springer Verlag, 2017, 95 (4), pp.4025-4047. �10.1007/s11277-017-4038-1�.
�hal-01538943�

HAL Id: hal-01538943

https://hal.archives-ouvertes.fr/hal-01538943
Submitted on 14 Jun 2017

HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est

archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents
entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non,
lished or not. The documents may come from émanant des établissements d’enseignement et de
teaching and research institutions in France or recherche français ou étrangers, des laboratoires
abroad, or from public or private research centers. publics ou privés.
Noname manuscript No.
(will be inserted by the editor)

Wireless Sensor Network based Smart Grid

Communications: Challenges, protocol optimizations,
and validation platforms

Sana Rekik · Nouha Baccour · Mohamed

Jmaiel · Khalil Drira

Received: date / Accepted: date

Abstract Smart grids, the next generation of electric grids, require the deploy-
ment of sophisticated monitoring and control systems to enhance their operational
efficiency. Wireless Sensor Networks (WSNs) have been considered as a promising
communication technology for the monitoring and control of smart grid opera-
tion. They bring significant advantages such as, rapid deployment, low cost and
scalability. However, the deployment of WSNs in smart grids brought new chal-
lenges mainly due to the electric grid features. Consequently, traditional WSN
communication protocols have been shown inadequate and several recent research
efforts were dedicated for their optimization. This paper provides a comprehensive
survey on related literature, discusses the still-open research issues, and identifies
the most common validation platforms for experimenting WSN communications in
smart grid. We believe this survey will pave the way for the research community to
(i.) understand important concepts related to WSN-based smart grid communica-
tions, (ii.) identify gaps and make valuable contributions in this timely and exiting
field and (iii.) choose the convenient experimental platform for the validation of
proposed solutions.
Keywords Smart grid communications · Wireless sensor networks · WSN-based
smart grid communications · Protocol optimization · Validation platforms

1 Introduction

The need for energy conservation and environmental compliance has driven gov-
ernments and utility industry over the world to transform their existing electrical
grid into smart electrical grid or shortly ”smart grid”. Indeed, existing electrical

S. Rekik (B) · N. Baccour · M. Jmaiel

ReDCAD Laboratory, University of Sfax, National School of Engineers of Sfax, B.P. 1173, 3038
Sfax, Tunisia
E-mail: sana.rekik@redcad.org
K. Drira
CNRS, LAAS, 7 Avenue du Colonel Roche, F-31400 Toulouse, France
2 Sana Rekik et al.

grids have several shortcomings calling for their modernization: First, they face the
challenge of limited power resources and growing power demand. Power generation
in traditional electrical grids relies on power plants which mainly use fossil fuels to
produce electricity. These energy resources are not only non-renewable but also en-
vironmentally unfriendly since they contribute significantly to gas emissions and
climate changes. Secondly, traditional electrical grids suffer from the imbalance
between power demand and supply. The amount of power to produce is generally
overestimated (i.e., exceeds real power demand) in order to prevent outage in the
power. This results in significant energy waste as the non-used power is dissipated
due to the difficulty of its storage. Third, the legacy electrical grids lack advanced
communication, monitoring and diagnostic capabilities. Electric grid utilities are
not able to remotely monitor and diagnose the grid status and obtain a complete
picture of the grid in real time. This is due to (i.) the limited control and sensing
technologies (i.e., few wired sensors are deployed), and (ii.) the lack of modern
communication networks that transfer, in real time, monitoring information from
the different electric grid components to the utility.
The smart grid [23, 18] represents the next generation of electrical grid aiming
to resolves all the above mentioned problems with the use of advanced information
and communication technologies (ICT). For example, smart grids are intended to
massively incorporate renewable energy sources (e.g., solar and wind) and manage
their high time-variability, unpredictability and distributed generation. They are
also expected to customize power demands based on the available power supplies,
by carrying two-way flows of information: a flow of metering data about power
consumption from the customer premises to the electric service provider and a
flow of pricing information in the other direction to coerce the customers into
adapting their demands. It is also foreseen that smart grids will ensure smart
and real-time monitoring of the grid status through the deployment of advanced
sensing capabilities.
Communication networks play an important role in the transformation of the
current electric grid into smart grid [63, 25, 65, 32, 22, 43, 44]. They provide inter-
connections between all the grid components (e.g., generators, substations, energy
storage systems, smart meters, etc.), allowing for the exchange of a large amount
of information. For example, real-time information about components conditions
can be exchanged for control, monitoring, and maintenance purposes. These mon-
itoring information are of great importance as they allow avoiding the impact of
electric equipment failures and natural accidents leading to power disturbances
and outages.
Wireless Sensor Networks (WSNs) have been shown as a promising technology
for smart grid monitoring and control applications [16, 30]. They bring signifi-
cant advantages compared with traditional wireless communication technologies,
namely rapid deployment, large areas coverage and low cost. The potential moni-
toring applications enabled by WSNs in smart grids spread throughout the entire
electric grid network, from power generation to transmission, distribution and con-
sumption. For instance, in the power generation side, WSN applications include
monitoring of wind farms, photo-voltaic panels and distributed energy genera-
tion [6, 45]. As for the T&D side, the applications of WSNs include monitoring of
overhead transmission lines, underground power lines, poles and towers, and sub-
station equipments [64, 47, 46, 1, 42]. Automatic Meter Reading (AMR), Advanced
Wireless Sensor Network based Smart Grid Communications... 3

Metering Infrastructure (AMI) and home/building energy management are exam-

ples of WSN applications in the consumer side [21, 17].
The application of WSNs in smart grids brought new challenges due to (i.) the
electric grid features and (ii.) the different traffic patterns of WSN-based smart
grid applications. For example, field tests conducted in [30] reveal that environ-
mental conditions in the smart grid are complex and harsh (electromagnetic in-
terference, obstructions, fading, etc.) turning wireless links extremely dynamic
and even unpredictable. Further, WSN applications in smart grids have different
quality-of-service (QoS) requirements, especially in terms of latency and reliabil-
ity [32], requiring the design of differentiated communication protocols.
A fundamental question that would precede any investigation in this field is
”whether existing WSN communication protocols and standards are adequate for
smart grid applications?”. Prior research works were dedicated for providing re-
sponse elements to this question [11, 61, 68]. Several well-known technologies and
protocols were evaluated such as, ZigBee, IEEE 802.15.4, IEEE 802.11, AODV,
etc. As a result, all these technologies and protocols were found not sufficiently
adequate for smart grid applications.
Currently, several research efforts are devoted for the optimization of existing
WSN communication protocols in order to cope with the new constraints emerged
from WSN-based smart grid applications. This fact motivates us to provide a
survey that synthesizes and classifies the related literature and also explores the
still-open research issues in this promising and timely field. To achieve this goal,
we adopted the following methodology: After reviewing related work (Section 2),
we first present fundamental concepts of communication networks in the smart
grid, including architectures, technologies and requirements (Section 3). Then, we
discuss the design challenges raised by WSNs in smart grid control and monitoring
applications (Section 4). Next, we provide a taxonomy of the most recent advances
in the design of new or optimized communication protocols for WSN-based smart
grid communications and discuss the open issues (Section 5). Finally, we present
the most common validation platforms for the performance evaluation of WSN
communication protocols in smart grids (Section 6), followed by our conclusion
(Section 7).

2 Related work

Several surveys have been introduced to tackle the different aspects of smart grid
communications in general, such as the communication infrastructure [65], archi-
tecture [32], requirements [32], design challenges [53], energy efficiency [19], etc.
However, to our knowledge, except [20], no previous survey paper tackled in de-
tail WSN-based communication in smart grids. In [20], Fadel et al. surveyed the
wide range of WSN applications in the different smart grid sides (i.e., generation,
T&D and consumer sides). They also discussed the adequacy of traditional WSN
MAC and routing protocols (such as RAP, ReInforM, RT-MAC, QoS-MAC, etc)
for smart grid by analyzing their features and capabilities and checking whether
they meet smart grid constraints. The authors argued that these protocols are
inherently inadequate for smart grid applications.
This paper builds on the survey of Fadel et al. [20], and provides a taxonomy of
recent advances on the optimization and design of WSN communication protocols
4 Sana Rekik et al.

for smart grid applications while discussing the still-open research issues. Further,
it provides a useful guide for network designers to choose the convenient validation
platform for their introduced protocols.

3 Smart grid communications

From an architectural perspective, the smart grid consists of three main layers:
energy layer, communication layer and information technology layer [25, 32]. The
energy layer is comprised of power generation, transmission and distribution sys-
tems, and consumer premises. The information technology layer collects and man-
ages data coming from different components of the smart grid. The communication
layer presents the core of the whole smart grid infrastructure. Communication net-
works support two-way flows of information between all devices in the grid (e.g.,
generators, substations, energy storage systems, smart meters, etc.). Exchanged
information represents either (i.) energy consumption and price, allowing for a
balance between energy demand and supply or (ii.) the conditions of smart grid
components, allowing for their control, monitoring, and maintenance.
A variety of communication technologies can be exploited in smart grid com-
munications, ranging from wired, such as optical fiber and Power Line Commu-
nications (PLC), to wireless technologies, such as Wireless Mesh Networks, Wire-
less Sensor Networks, cellular communication systems and satellite communica-
tions [23, 31, 25, 63, 7]. Wired technologies have the disadvantage of incurring high
cost for deployment and maintenance, which limits the network scalability. Conse-
quently, wireless communication technologies have more widespread use in smart
grid networks, compared to wired technologies [34]. In particular, WSNs have been
shown as a key communication technology for the control and monitoring appli-
cations in smart grids. This is due to their intrinsic characteristics, such as low
cost, wide coverage and easy installation. However, WSNs cannot operate alone.
They are intended to cooperate with other network technologies (e.g. high speed
networks) to ensure better coverage [64, 24, 69]. For instance, the combination of
WSNs and wide area networks such as cellular networks was proposed in [24],
to meet the requirements of the transmission line monitoring application, namely
high bandwidth and low latency.
Generally, all involved communication networks in the smart grid (including
WSNs) have to cooperate together to allow the exchange of different types of data,
while meeting the main communication requirements of smart grid applications,
namely latency, reliability, bandwidth, security, privacy, scalability and interoper-
ability [31, 63, 65, 7, 38]. These requirements are elaborated next:
• Latency: In the smart grid, most of information must be delivered within hard
time bounds. When the communication delay exceeds a certain threshold, the
information becomes outdated, which may lead to a damage in the grid. The
International Electrotechnical Commission (IEC) and the Institute of Electrical
and Electronics Engineers (IEEE) specified expected values of communication
delays for different categories of exchanged information in the smart grid (e.g.,
maintenance information, protection information, video data streams, etc.) [63].
For example, in the circuit breaker protection application, sensed data must
be received within a delay of 3 ms in order to make the necessary protection
actions [63].
Wireless Sensor Network based Smart Grid Communications... 5

• Reliability: It generally refers to high end-to-end data delivery. In smart grid ap-
plications, exchanged messages have different levels of reliability: some of them
may occasionally tolerate losses, while others require a high level of reliability
(up to 99.99%) [63, 7]. Smart grid communication networks must allow applica-
tions to select between different levels of reliability.
• Bandwidth: The number of interconnected intelligent devices is anticipated to
grow rapidly with the evolution of the smart grid. Thus, the bandwidth demands
are also growing [32]. The bandwidth of smart grid communication networks
must increase as fast as the demand of the connected elements in the grid to
ensure low transmission latency and to reduce packet losses.
• Security and privacy: Providing communication security, especially protecting
the customers’ privacy, is of great importance for smart grid applications to
prevent vulnerabilities to the critical assets of the electric grid. This vulnerability
is the consequence of the high penetration and the variable nature of renewable
energy generators, the deployment of smart meters, and the large geographical
area covered by the smart grid [23, 7]. Without appropriate security mechanisms
and privacy protection methods, interested parties access to the electric system
and destabilize the grid (e.g., altering the information stored in smart meters).
• Scalability: Smart grid communication networks have to accommodate more and
more users and smart devices (e.g. smart sensors, smart meters). This raises the
issue of scalability in the smart grid, which may be handled with the use of
IP-based networks, web services, reliable protocols, etc [31, 65].
• Interoperability: Building an interoperable smart grid communication infras-
tructure is a major challenge. Indeed, smart grid incorporates diverse networks,
which are intended to cooperate and exchange information. The adoption of
standards and open network architectures (using for example IP-based net-
works), while avoiding being isolated into proprietary protocols; will play a
crucial role for interoperability at all the network layers [65, 7]. IP-based stan-
dards may be considered as a promising solution to achieve interoperability in
smart grid communications. For instance, WSN standards introduced by the
IETF, including 6LoWPAN and RPL, can ensure high interoperability (with
IPv6-based networks). In addition to the use of standardized protocols, stan-
dard interfaces and networking elements (e.g., gateways) may be introduced to
provide communications between different networks that use different standards.

4 Protocol design challenges in WSN-based smart grid applications

WSNs have unique characteristics that make them different from any other wireless
network. For example, sensor devices are designed with constrained resources in
terms of energy, memory, and processing capabilities. Further, they use low cost
and low-power radios, which turn communication links prone to noise, interference
and multi-path effects. These features have been well considered in the design of
major WSN communication protocols.
Smart grid monitoring applications represent a recent and emerging application
domain of WSNs. Consequently, the application of WSNs in smart grids arise
new constraints that should be managed by WSN protocols. As summarized in
Fig. 1, these constrains pertain to (i.) the harsh smart grid environments, (ii.)
the different QoS requirements of WSN-based smart grid applications (mainly in
6 Sana Rekik et al.

Fig. 1 Constraints of WSN-based smart grid applications.

terms of communication reliability and latency), and (iii.) the large-scale of the
electric grid. In this section, we discuss these constraints, while emphasizing their
impact on the design of efficient WSN communication protocols for smart grid
applications.

4.1 Harsh smart grid environments

Electric grid environments are typically characterized by highly corrosive con-

ditions (e.g. wind, rain, solar radiation, humidity, vibrations, dust, etc.), which
contributes to communication link unreliability and sensor nodes failures.
Low-power link unreliability in smart grid environments. Communication links
in WSNs (also referred as low-power links) have been shown to be extremely
unreliable: Their quality fluctuates over time and space, and connectivity is typi-
cally asymmetric [8, 9]. This is due to the low cost and low-power radios used in
WSNs. Unfortunately, the harsh smart grid environments make low-power links
even more unreliable. For example, Gungor et al. [30] argue that smart grid distri-
bution environments have variable link delivery and higher packet error rate due
to electromagnetic interference, obstructions, equipment’s noise, etc. Low-power
link unreliability in smart grid environments constitutes a major handicap for
communication protocols to maintain the network correct operation and ensure
acceptable end-to-end data delivery.
Link quality estimation enables network protocols to overcome low-power link
unreliability. For instance, some routing protocols rely on link quality estimation to
deliver data over links with high quality, which improves the network throughput
by limiting packet loss and maximizes its lifetime by minimizing the number of re-
transmissions and avoiding route re-selection triggered by links’ failure. Topology
control mechanisms also rely on link quality estimation to maintain the stability of
the topology. High-quality links are long-lived, therefore, efficient topology control
Wireless Sensor Network based Smart Grid Communications... 7

mechanisms aggregate high-quality links to maintain robust network connectivity

for long periods, thus avoiding unwanted transient topology breakdown.
Empirical characterization of low-power links through real-world measurements
in smart grid environments is also of paramount importance to mitigate low-power
link unreliability. It allows a deep and clear understanding of link behavior and
statistical properties, which (1) contributes to the design of efficient link quality
estimators, and (2) allows deriving accurate wireless channel models that reflect
radio propagation in real smart grid environments. For example, in [30], based
on extensive empirical link measurements the authors provided an estimation of
the parameters of the log-normal shadowing path loss model, for different power
distribution environments.
Sensor failures in harsh smart grid environments. The severe conditions of
electrical grids may adversely affect the performance of sensor nodes, causing a
portion of them to malfunction. Further, sensor devices can be deployed in loca-
tions that are tough to access and replace battery (e.g. sensors for underground
cables) [16].
Sensor failures and battery depletion impact the network topology and can be
overcome using adaptive protocols to maintain the network operation following
node failures. Energy harvesting techniques may be also used to extend the bat-
tery lifetime of sensor nodes. In some WSN-based smart grid applications, main
power may be available. For instance, in [26], the power supply to the sensor nodes
is obtained from power supplies in substations and power lines. However, the volt-
ages of the power lines are too high and should be stepped down before being used
by the sensor nodes.

4.2 Different Quality of Service (QoS) requirements

WSNs are expected to satisfy the main requirements of smart grid communica-
tions, namely reliability and latency. On the other hand, WSN-based smart grid
applications have different and often contradictory QoS requirements in terms
of reliability and latency, resulting in a heterogeneous data traffic patterns [32].
For example, overhead transmission line monitoring and substation automation
represent critical smart grid applications that need strict reliability and latency
requirements, while other applications, such as Demand Response and AMI, do
not expect critical latency. Some other smart grid applications, such as residential
energy management, can occasionally tolerate losses in data transfer.
The heterogeneity of traffic makes QoS provisioning in WSN-based smart grid
applications a non-trivial task. Data traffic should be classified into several classes,
according to the QoS requirements. Communication protocols are then expected
to provide differentiated services to handle different traffic classes originated from
a wide range of possible smart grid applications [5]. Service differentiation is dis-
cussed in Section 5.1.

4.3 Large-scale electric grid

The geographical span of the electric grid is anticipated to be very large, espe-
cially with its extension into the consumer premises [16]. The number of sensors
8 Sana Rekik et al.

employed to monitor the different smart grid components, from generation to con-
sumer side, is too high and the management of this large-scale network would be a
challenging task [65]. For this reason, the design of scalable communication proto-
cols for WSN-based smart grid applications is necessary (e.g., IP-based protocols).
Data aggregation techniques are also mandatory to avoid redundant data collected
by sensors close to each other.

5 On the optimization of communication protocols for WSN-based

smart grid applications

The application of WSNs in smart grids brought new challenges that should
be properly considered by communication protocols. A fundamental question is
”whether existing WSN communication protocols and standards are adequate for
smart grid applications?”. Several research efforts have been dedicated for pro-
viding response elements to this question. For example, in [11], Bilgin et al. in-
vestigated the performance of ZigBee in different smart grid environments. It was
shown that ZigBee is only adequate for low-data rate smart grid applications not
requiring high reliability and real time deadlines. Temel et al. [61] studied the per-
formance of different types of routing protocols, namely on-demand (AODV (Ad
hoc On-Demand Distance Vector) and DYMO (Dynamic MANET On-demand)),
table-driven (DSDV (Destination Sequenced Distance Vector)) and QoS-aware
(TUQR (Topology Unaware QoS Routing)) protocols. They demonstrated that
these routing protocols are unable to deliver satisfactory performance in smart
grid environments [61]. The performance of state-of-the art MAC protocols, namely
IEEE 802.15.4, IEEE 802.11, CSMA, TDMA and Z-MAC, was evaluated by Yigit
et al. [68], demonstrating that these MAC protocols can not deliver adequate per-
formance in smart grid environments when network traffic load increases [68]. As a
conclusion, traditional WSN protocols have been shown not sufficiently adequate
for smart grid applications having different QoS requirements.
Recently, several techniques and mechanisms have been introduced for the op-
timization of existing WSN communication protocols for smart grid applications,
especially (1) traffic/service differentiation, (2) interference mitigation and (3) link
quality estimation. In this Section, we classify the recent research works on the
optimization of WSN protocols for smart grid applications according to the above
mentioned optimization techniques as well as the targeted layer (i.e., layered and
cross-layer approaches). Table 1 illustrates the proposed taxonomy.

5.1 Traffic/service differentiation

Service differentiation is a mechanism allowing distinguishing different traffic classes

where each class has its specific QoS requirements and priorities (in terms of re-
liability, latency, throughput, etc.). As stated previously, WSN-based smart grid
applications are diverse which leads to various traffic patterns, each having its
own QoS requirements (e.g., smart meter readings, real time pricing information,
periodic monitoring of electric components, etc.). Therefore, QoS differentiation
in WSN-based smart grid applications is a fundamental building block for com-
Wireless Sensor Network based Smart Grid Communications... 9

munication protocols to ensure acceptable QoS with respect to each application

or traffic class.
QoS differentiation can be performed at different layers, including network [54,
50] and MAC [2, 3, 60, 4] layers. It can also be performed at several layers. It is
thus referred to as a cross-layer QoS optimization scheme, such as in [60]. The
proposed solution consists of a MAC and a routing algorithm with QoS support.
At the MAC layer, the existing CSMA/CA of IEEE 802.15.4 was improved by
providing differentiated services for data traffic with different priorities (low and
high). The introduced routing algorithm cooperates with physical and MAC layers
to select paths that satisfy the reliability and latency requirements.

5.1.1 Routing level QoS differentiation

The authors in [54] argued that multi-path routing, i.e., routing that establishes
multiple paths between the source and the destination, provides QoS differentia-
tion in WSN-based smart grid applications. To demonstrate their statement, the
authors performed a comparative performance study of two WSN routing proto-
cols: SPEED and MMSPEED. SPEED is a single-path routing protocol that does
not support QoS differentiation. However, it is a QoS-aware protocol as it provides
end-to-end communication deadline guarantees. MMSPEED extends SPEED by
providing QoS differentiation through multi-path routing. In addition to real-time
communication guarantees, MMSPEED guarantees different reliability require-
ments by enabling probabilistic multi-path forwarding.
Different from the previous work (which provides QoS differentiation through
multi-path routing), a QoS differentiation mechanism that exploits multi-topology
routing in RPL protocol, is presented in [50]. In RPL, a network topology consists
of one or several DODAGs (Destination-Oriented Directed Acyclic Graph), which
form together an RPL instance. Each instance is built based on an Objective
Function, which specifies how to select paths based on certain criteria (e.g., link
quality, latency, hop count, etc.). An RPL network may consist of multiple RPL
instances where each instance operates independently of the others. This feature
was exploited in [50] to provide QoS differentiation for RPL, deployed in smart
grid Neighbor Area Networks (NANs). The authors proposed to associate each
traffic type with a different RPL instance. Especially, Objective Functions based
on Expected Transmission Count (ETX) and Hop Count (HC) were associated
with critical traffic and periodic traffic, respectively.

5.1.2 MAC level QoS differentiation

In [2–4], QoS differentiation is performed at MAC level by optimizing the IEEE

802.15.4. In [2], the authors targeted high data rate smart grid monitoring appli-
cations, such as Partial Discharge (PD) monitoring. The adoption of the IEEE
802.15.4 standard for these applications can lead to excessive delays as it is de-
signed for low data rate applications. The authors in [2] introduced a QoS dif-
ferentiation scheme that reduces the end-to-end delay of high priority and delay
critical packets. The introduced solution is based on a delay optimization model of
IEEE 802.15.4 for cluster-tree WSNs. The model allows each Cluster Head (CH)
minimizing the inter-CH delay by adjusting two IEEE 802.15.4 MAC parameters:
the Contention Access Period (CAP) and the CFP (Contention Free Period).
10 Sana Rekik et al.

Another optimization of the IEEE 802.15.4 MAC is presented in [4], where

the authors introduced two adaptive delay reduction schemes for cluster-tree and
mesh WSNs. The proposed schemes can adaptively tune the Guaranteed Time
Slot (GTS) to provide QoS differentiation required by delay critical smart grid
applications.
To address latency and QoS differentiation requirements in smart grid mon-
itoring applications, the authors in [3] proposed a scheme that is based on data
prioritization and delay estimation procedures performed at the application layer.
Thus, service differentiation is provided to high priority traffic by modifying the
Clear Channel Assessment (CCA) duration in the IEEE 802.15.4.

5.2 Interference mitigation

Smart grid environments are commonly characterized by electromagnetic inter-

ference, equipment’s noise and multi-path effects, which negatively impact the
performance of WSN communications [30]. Two techniques have been proposed
to mitigate the impact of interference in smart grid environments: cognitive radio
technology and multi-channel communication.

5.2.1 Cognitive radio technology

It is relatively a new approach for utilizing and accessing spectrum (i.e., radio
frequencies) in smart ways. Cognitive radios have the ability to dynamically sense
the spectrum, determine the unused bands and exploit them in an opportunis-
tic manner. Consequently, they can identify potential troubles in communication
quality, such as interference and multi-path fading. These features make cognitive
radio technology a promising solution for WSN applications in smart grid envi-
ronments to overcome their hostile propagation conditions [28]. Cognitive radio
sensor networks can enhance the network performance in terms of reliability and
communication latency through the mitigation of the congested and noisy bands
in the spectrum. This allows the selection of high capacity links for wireless com-
munications in smart grids and hence improves the reliability and the latency of
the network. For example, the authors in [56, 55], exploited cognitive radio commu-
nications to circumvent the harsh propagation conditions in the smart grid. They
proposed a cognitive radio-based framework that supports the QoS requirements
of WSN-based smart grid applications, namely latency, data rate and reliability.
However, the benefits of cognitive radio come with significant additional energy
consumption mainly caused by spectrum sensing operations, which may affects the
network lifetime. The advantages and challenges of cognitive radio wireless sensor
networks in smart grid applications are discussed in detail in [10].

5.2.2 Multi-channel communication

Multi-channel mechanism allows several nodes to communicate at the same time,

using different channels [36, 59]. Hence, multi-channel WSNs allow to (i.) alleviate
the effects of interference on wireless medium and (ii.) achieve parallel transmis-
sions through multiple channels for better data delivery. Multi-channel communi-
cation has been argued to be an efficient solution for WSN-based smart grid appli-
Wireless Sensor Network based Smart Grid Communications... 11

cations to mitigate electromagnetic interference, equipment’s noise and multi-path

effects characterizing smart grid environments.
In [12], the performance of multi-channel MAC protocols was evaluated in dif-
ferent smart grid environments and compared against that of single channel WSN
MAC protocols. This study showed that multi-channel WSNs improve the network
capacity compared to single channel WSNs: Increasing the number of communi-
cation channels helps to increase the packet reception rate, reduce communication
delay and increase almost twice the network throughput.
 12
Table 1 Taxonomy of contributions related to the optimization of WSN communication protocols for smart grid applications
Works Layered approach Cross-layer Optimization technique
approach
MAC layer Routing layer Traffic/service Interference mit- Link quality esti-
differentiation igation mation
[2] Optimization X x x
of the IEEE
802.15.4 MAC
for high data
rate applications
[54] Use of multi- X x X
path routing LQE: PRR
protocol
[50] Use of multiple X x X
RPL instances LQE: ETX
[29] Impact of repre- x x X
sentative LQEs (QoS provi- LQEs: PRR,
on CTP routing sioning for 1 WMEWMA,
protocol traffic class) RNP, ETX and
four-bit
[52] Use of holistic x x X
link quality es- (QoS provi- LQE: Opt-
timation to im- sioning for 1 FLQE
prove the reli- traffic class)
ability of RPL
routing protocol
[67] Use of multi- x X X
channel MAC & (QoS provi- Technique: LQE: PRR
Tree routing sioning for 1 multi-channel

Sana Rekik et al.

traffic class) communication
[66] Use of multi- X X X
channel MAC & Technique: LQE: PRR
channel-aware multi-channel
routing communication
[56] [55] Design of a X X x
framework for Technique:
QoS support cognitive radio
(interaction of technology
routing, MAC
and physical
layers)
[60] An improved X x X
IEEE 802.15.4 LQE: PRR
MAC & Routing
algorithm with
QoS differentia-
tion
[3] [4] Adaptation of X x x
IEEE 802.15.4
MAC (interac-
tion of MAC
and application
layers)
Wireless Sensor Network based Smart Grid Communications... 13

Although multi-channel communications allows mitigating interference, recent

research studies show that the network capacity is still limited by the routing topol-
ogy [67, 66]. In [67], the authors proposed the use of both multi-channel communi-
cation together with an appropriate routing tree protocol to further enhance the
performance of WSNs in smart grid environments. They selected RBCA (Receiver-
Based Channel Assignment) as a multi-channel MAC algorithm. The network
performance was evaluated using different routing tree algorithms, demonstrating
that CMST (Capacitated Minimum Spanning Tree) with PRR routing tree shows
an improvement in the network capacity in terms of delay and throughput.
In [66], the authors modified the RBCA’s time slot assignment algorithm as well as
the transmission schedule in order to support data prioritization. They proposed a
priority and channel-aware scheduling algorithm for smart grid applications. The
proposed algorithm performs successively (i.) channel assignment, while consid-
ering interference between nodes, (ii.) time slot assignment and (iii.) delay-aware
data transmission, allowing packet transmission according to their traffic classes
and priorities. Three classes of traffic were considered: real-time, non-real-time and
best effort traffic, having high, medium and low priority respectively. The real-time
packets are transmitted first, then the non-real-time and best effort packets are
sent respectively.

5.3 Link quality estimation

Link quality estimation is a prerequisite for higher layer protocols to overcome

low-power links unreliability in smart grid environments [30]. The performance of
well-known Link Quality Estimators (LQEs) in typical smart grid environments
was studied in [29]. The evaluated LQEs are Packet Reception Ratio (PRR), Win-
dow Mean with Exponentially Weighted Moving Average (WMEWMA), Expected
Transmission Count (ETX), Required Number of Packet Transmissions (RNP) and
four-bit. The comparative simulation study is based on radio propagation param-
eters of typical smart grid environments, found experimentally in [30]. The perfor-
mance of LQEs is assessed by comparing their impact on the collection tree routing
protocol (CTP). The authors found that ETX and four-bit estimators showed the
best performance in different smart grid environments since they consider the link
asymmetry property.
The authors in [51] compared the performance of LQEs, using a different evalu-
ation methodology from that used in [29]. This methodology consists in analyzing
the statistical properties of LQEs, independently from any external factors like
MAC collisions or routing. Further, the authors in [51] devised using composite
LQEs (i.e., LQEs that combine several link metrics) to provide holistic and reli-
able link quality estimation for WSN-based smart grid applications. The evaluated
LQEs are ETX, four-bit and F-LQE (Fuzzy-Link Quality Estimator). It was found
that, F-LQE is more reliable and more stable than ETX and four-bit. However,
it lacks reactivity which makes it not sufficiently adequate for smart grid envi-
ronments characterized by excessive packet loss and link burstiness. An optimized
version of F-LQE, called Opt-FLQE, that improves the reactivity and reduces the
computational complexity of the original F-LQE is then proposed.
In [52], the authors used Opt-FLQE for reliable end-to-end communication
in RPL networks deployed in smart grids. They proposed an alternative routing
14 Sana Rekik et al.

metric for RPL based on Opt-FLQE, called Opt-FLQERM . The authors showed
that Opt-FLQERM improves RPL performance over traditional routing metrics,
including the RPL default metric, mainly in terms of packet loss ratio and end-
to-end delay.

5.4 Discussion and open issues

The application of WSNs in smart grids raises new constraints, pertaining to

both the harsh electrical grid nature and the diverse QoS requirements of smart
grid monitoring applications. Throughout this section, we have presented relevant
techniques that have been introduced to cope with these constraints, namely QoS
differentiation, interference mitigation (through multi-channel communication or
cognitive radio communication) and link quality estimation. Nevertheless, the de-
ployment of WSNs in smart grids still has several challenges and open issues as
detailed next:

• Energy efficiency of cognitive radio sensor networks in smart grid communi-

cations. Cognitive radio technology promises minimizing the inconvenient en-
vironmental effects of smart grids (e.g. interference) and then improving the
communication delay and reliability of WSNs in smart grids. However, the pro-
posed approaches based on cognitive radio technology [10, 56, 55] have not been
evaluated in terms of energy-efficiency. It is well known that, the spectrum
scanning operation performed by the cognitive radio is energy greedy [28] and
thus, achieving energy-efficient communication, especially in large scale smart
grid deployments, is not a trivial task. Therefore, there is a need for further
research efforts addressing the trade-off between energy consumption, reliability
and communication delay. On the other hand, the scarcity of WSN platforms
with cognitive radio capabilities justifies the fact that most related literature is
simulation-based. Common WSN radio chips (e.g. the CC1000 radio used for
Mica2 nodes and the CC2420 radio for MICAz and TelosB sensor nodes) are
not cognitive radios.
• Adoption of standard multi-channel MAC protocols for WSN-based smart grid.
Multi-channel communication technique is supported by the available radios,
such as the CC2420 radio used in MICAz and TelosB WSN platforms. These
radios can just be tuned to operate on multiple frequency channels. However, the
performance of WSN multi-channel protocols is not well investigated, especially
for standardized multi-channel MAC protocols that offer the ease of integration
and interoperability, as well as communication reliability.
The IEEE 802.15.4e standard [35] is a recent amendment of IEEE 802.15.4
MAC specification, that aims at meeting the requirements (reliability, ultra-low
power consumption and bounded latency) of industrial applications, including
smart grid applications. It enhances the old standard by introducing mecha-
nisms such as multi-channel communication, time slotted access and channel
hopping. The IEEE 802.15.4e defines several MAC operation modes, including
Time Slotted Channel Hopping (TSCH) and Deterministic and Synchronous
Multi-channel Extension (DSME) [49, 27]. Both TSCH and DSME MAC proto-
cols enhance communication reliability by employing channel diversity schemes
and time slotted access. To improve reliability, TSCH MAC protocol employs
Wireless Sensor Network based Smart Grid Communications... 15

channel hopping, while DSME MAC protocol employs two types of channel di-
versity schemes: channel hopping and channel adaptation. The suitability and
the adoption of TSCH and DSME MAC protocols for WSN-based smart grid
applications is still an open research issue.
• Accurate link quality estimators for WSN higher layer protocols in smart grid
applications. Link quality estimation is a prerequisite for higher layer protocols
to improve the end-to-end data delivery and overcome low-power link unreliabil-
ity in harsh smart grid environments. We notice that most approaches proposed
for WSN-based smart grid applications leverage on simple link quality estima-
tors, such as PRR, [54, 60, 67, 66]. However, it is now well known that PRR is
inaccurate as it is only able to provide one link aspect, link delivery, and ignores
other important aspects such as link asymmetry [8]. Thus, there is a need for
employing accurate link quality estimators which are able to reflect the real
link status in smart grid environments. Further, few of work (except [30]) tack-
led the empirical analysis of low-power links in smart grid environments. Such
studies are of paramount importance as they provide a better understanding
of low-power link characteristics, which contributes to the design of accurate
link quality estimators and realistic wireless channel models improving network
simulation as it will be discussed in the next Section.
• Real experimentation of WSNs in smart grid environments. Unfortunately, most
research studies around WSN-based smart grid applications are based on net-
work simulation, using channel parameters of typical smart grid distribution
environments derived in [30]. Definitely, the study in [30] represents a valuable
contribution to the state of the art. However, it would be interesting to con-
duct additional similar studies to derive channel parameters in other smart grid
environments which are more hostile, especially in the smart grid generation
side such as power plants. Generally, there is a clear gap in literature in what
concerns real experimentation of WSNs in smart grids.

6 Validation platforms for WSN-based smart grid solutions

Experimental evaluation through field trials is an efficient validation tool, as it

allows anticipating the deployment of the proposed WSN solutions in real smart
grid environments and then get performance results with high confidence. However,
conducting field trials in harsh smart grid environments is not a trivial task. For
instance, performing experiments at the generation and transmission sides of the
smart grid (e.g., power plants and substations), where the voltages are too high,
can endanger the human safety. Running real experiments in laboratory smart grid
Testbeds can be an interesting alternative. However, up to now, publicly accessible
smart grid Testbeds are still missing.
Due to the challenges raised by field trials, most research efforts on WSN-
based smart grid solutions use network simulation. However, to reflect smart grid
features and get meaningful results, simulation scenarios should be properly set
using channel parameters that reflect radio propagation behavior in smart grid en-
vironments. Generally, these parameters pertain to a particular radio propagation
model implemented in the simulator. For instance, the log normal shadowing path
loss model has been argued as an accurate radio propagation model [29, 67] and
has been widely adopted in WSN-based smart grid [11, 61, 54, 51]. The parameters
16 Sana Rekik et al.

of this model have been estimated in typical smart grid distribution environments
in [30], namely an outdoor 500 kV substation, an indoor main power control room
and an underground transformer vaults. Simulators such as NS-2 and TOSSIM
rely on this radio propagation model. Hence, performing simulation with these
simulators while setting the channel parameters to realistic values such as those
provided by Gungor et al. [30] contributes to rigorous simulation results.
The choice of the convenient network simulator to validate WSN-based smart
grid communication protocols depends not only on the available radio propagation
models, but also on the implemented MAC and routing protocols. In order to
help network designers to select the adequate WSN simulator, an overview and a
comparison of the potential simulators for WSN-based smart grid applications is
given next.

6.1 Potential simulators for WSN-based smart grid applications

In this section, we present the most representative WSN simulators that have been
used to validate WSN-based smart grid solutions. Table 2 presents some technical
features of these simulators.

• NS-2 (Network Simulator 2) [14] is a general network simulator (non-specific

for WSNs). It is implemented in C++ and the simulation scripts are written
in OTcl (an object-oriented extension of Tcl (Tool Command Language)). A
considerable number of communication protocols are already implemented in
NS-2, which make it a very popular simulator.
NS-2 is the most used simulator in the community of WSN-based smart grid
applications [11, 61, 68]. One of the reasons is the accuracy of its radio model -
namely the log normal shadowing path loss model.
• NS-3 [33] is a subsequent version of NS-2. It brings several improvements com-
pared to NS-2, regarding modularity, re-usability and extensibility. NS-3 is en-
tirely developed in C++, including simulation scripts, which can be also devel-
oped in Python. It is important to notice that NS-2 simulation scripts can not
be run in NS-3 as they are written in OTcl.
• TOSSIM [39] is a WSN simulator available as part of TinyOS operating system,
written in nesC (network embedded system C) programming language. The
same TinyOS applications, which are written in nesC, can be run on real WSN
nodes as well as simulated nodes (MICAz motes). The simulation scenario can be
defined in Python or C++. A configurable CSMA-like MAC protocol is used by
TOSSIM where several properties, such as backoffs and delays, can be adjusted.
To simulate the signal propagation between nodes, TOSSIM uses the log-normal
shadowing path loss model.
• COOJA [48] is a Java-based WSN simulator/emulator available as part of Con-
tiki operating system [15]. COOJA is included in Instant Contiki1 , an entire
development environment that can be easily downloaded as a virtual machine.
Like TOSSIM, the same Contiki application code, written in C, can run on both
real and simulated sensor nodes. Contiki provides a compact implementation of
the IPv6 communication stack optimized for embedded devices, called µIPv6.
1 http://www.contiki-os.org/start.html
Wireless Sensor Network based Smart Grid Communications... 17

This protocol stack includes the 6LoWPAN standard and the RPL routing pro-
tocol. One limitation of COOJA is that it supports simplistic radio models such
as the unit disk graph model.
• J-Sim [57, 58] is a network simulator entirely implemented in Java. J-Sim latest
release (version 1.3) defines simulation script in Jacl (a Java implementation
of Tcl script language). It has been extended to include frameworks for WSNs
simulation, including an implementation of AODV and DSR routing protocols.
J-Sim integrates simplistic radio propagation models, namely Free-space, Two-
ray ground and Irregular terrain models. The authors in [54] implemented the
log-normal shadowing path loss model in order to accurately model the wireless
channel in typical smart grid environments.
• QualNet [37] is a commercial network simulator based on C++. It is designed
for large-scale networks (i.e., thousands of nodes can be modeled). This feature
is provided by the use of Parsec programming language (an extension of C
language for parallel programming).
• OMNeT++ (Objective Modular Network Testbed in C++) [62] is a C++-based
simulator destined to model communication networks and distributed and/or
parallel systems. The network and topology descriptions in OMNeT++ are de-
veloped in Network Description (NED) language.
Castalia [13] is a simulation framework that extends OMNeT++ for WSNs.
Castalia provides a configurable MAC protocol (where some parameters can be
adjusted) and an implementation of T-MAC and S-MAC protocols. However,
it supports only simple routing protocols, namely Multipath Rings and simple
Tree.

6.2 Discussion and open issues

The design of network protocols tailored to the specific characteristics of the WSN-
based smart grid monitoring applications (presented in Section 4) is relatively a
new research area and the validation of such protocols is still a challenging task.
Simulation is a widely considered validation technique in the literature. Sophis-
ticated network simulators for WSN-based smart grid applications should basically
integrate (1) representative WSN protocols and standards, such as 6LoWPAN, Zig-
bee and IEEE 802.15.4, and also (2) accurate radio propagation models such as
the log normal shadowing path loss model [30, 67]. Further, simulation scenarios
should be properly set using channel parameters reflecting real WSN applications
is smart grid environments. Unfortunately, there is a lack of such sophisticated
network simulators. Therefore, an important and urgent open issue is to optimize
existing simulators to get convenient validation tools for WSN-based smart grid
applications. For example, COOJA is a well known simulator that has been ex-
tensively used in WSN research community thanks to its IPv6 protocol stack.
However, its radio model is very simplistic. Hence, there is a need to implement
the log normal shadowing path loss model for COOJA simulator.
The use of the co-simulation technique represents another potential valida-
tion tool for WSN protocols in smart grid applications. Co-simulation generally
refers to the cooperation of two or more simulators. In the case of smart grids,
power/network co-simulation mandates the integration of a power simulator and
a network simulator. Co-simulation is expected to allow studying and validating
18 Sana Rekik et al.

WSN-based smart grid solutions in a near real electric grid behavior. Different
smart grid co-simulation platforms were proposed, including [40, 70, 41]. For in-
stance, GECO (Global Event-driven CO-simulation platform) [41] is an example
of co-simulation platform where the power system is modeled using PSLF and inte-
grated with NS-2 network simulator. However, the different proposed co-simulators
are not yet widely used by researchers. This may be due to the fact that they
are just prototype tools (some implementations are not available to the research
community as an open-source). Further, the co-simulation configuration may be
a complex process that requires a deep knowledge about electrical components
modeling. Therefore, the usefulness of co-simulation as a validation tool for WSN
protocols in smart grids needs to be demonstrated.
 Wireless Sensor Network based Smart Grid Communications...
Table 2 Comparison of potential simulators for WSN-based smart grid applications
Simulator Programming/ Supported MAC Supported rout- Supported radio propagation
Simulation protocols ing protocols models
language
NS NS-2 C++ / OTcl IEEE 802.11 AODV, DSDV, Free space, Two-ray ground reflection,
MAC, TDMA, DSR, TORA Log normal shadowing
IEEE 802.15.4
MAC
NS-3 C++ / Phyton IEEE 802.11 MAC, AODV, DSDV, Fixed Received Signal Strength Propa-
CSMA DSR, OLSR gation, Matrix Propagation Loss, Ran-
dom Propagation Loss, Okumura Hata
Propagation, Friis Propagation, Log
Distance Propagation, Three Log Dis-
tance Propagation, Two Ray Ground
Propagation (ported from NS-2), Jakes
propagation, Nakagami propagation.
TOSSIM NesC / Configurable CTP, DYMO Log normal shadowing
Python, C++ CSMA-like MAC
COOJA Java / C IEEE RPL, AODV Unit Disk Graph, Directed graph,
802.15.4 MAC Multi-path ray tracing
(CSMA/CA),
NullMAC
J-Sim Java / Tcl IEEE 802.11 MAC AODV, DSR Free space model, Two-ray ground
model, Irregular terrain model
QualNet Parsec C++ Point-to-point, AODV, DSR, Free space model, two ray path
CSMA, IEEE LAR1, ODMRP loss model, Irregular Terrain Model,
802.11, IEEE 802.3 Ricean model, Rayleigh model, Con-
stant shadowing model, Log normal
shadowing model
OMNeT++/Castalia C++ / NED Configurable Simple Tree, Multi- Free space, Log normal shadowing,
MAC, T-MAC, path Rings Temporal variation
S-MAC

19
20 Sana Rekik et al.

7 Conclusion

The equipment failures and the limited monitoring and control capabilities in the
current electric grid are the main motivations for the migration to a smarter grid
with advanced communication and monitoring skills. WSNs represent a promising
technology for enabling the pervasive monitoring and control in smart grids. How-
ever, the application of WSN communication technology in smart grids brought
new challenges, that should be managed by WSN protocols.
In this paper, we identified and discussed these arisen challenges. Then, we
proposed a thorough taxonomy of recent research works on the optimization of
existing WSN communication protocols, to cope with the new constraints emerged
from the deployment of WSNs in smart grid applications. We also highlighted open
issues and discussed the potential validation platforms for WSN-based smart grid
applications.

References

1. Al-Anbagi, I.S., Erol-Kantarci, M., Mouftah, H.T.: A delay mitigation scheme for WSN-
based smart grid substation monitoring. In: IWCMC, pp. 1470–1475 (2013)
2. Al-Anbagi, I.S., Erol-Kantarci, M., Mouftah, H.T.: QoS-aware inter-cluster head schedul-
ing in WSNs for high data rate smart grid applications. In: 2013 IEEE Global Commu-
nications Conference, GLOBECOM 2013, Atlanta, GA, USA, December 9-13, 2013, pp.
2628–2634 (2013)
3. Al-Anbagi, I.S., Erol-Kantarci, M., Mouftah, H.T.: Priority- and delay-aware medium
access for wireless sensor networks in the smart grid. IEEE Systems Journal 8(2), 608–
618 (2014). DOI 10.1109/JSYST.2013.2260939
4. Al-Anbagi, I.S., Erol-Kantarci, M., Mouftah, H.T.: Delay critical smart grid appli-
cations and adaptive qos provisioning. IEEE Access 3, 1367–1378 (2015). DOI
10.1109/ACCESS.2015.2466077
5. Al-Anbagi, I.S., Erol-Kantarci, M., Mouftah, H.T.: A survey on cross-layer quality-of-
service approaches in WSNs for delay and reliability-aware applications. IEEE Communi-
cations Surveys and Tutorials 18(1), 525–552 (2016). DOI 10.1109/COMST.2014.2363950
6. Al-Anbagi, I.S., Mouftah, H.T., Erol-Kantarci, M.: Design of a delay-sensitive WSN for
generation monitoring in the smart grid. In: CCECE, pp. 1370–1373 (2011)
7. Ancillotti, E., Bruno, R., Conti, M.: The role of communication systems in smart grids:
Architectures, technical solutions and research challenges. Computer Communications
36(17-18), 1665–1697 (2013)
8. Baccour, N., Koubaa, A., Mottola, L., Zamalloa, M.A.Z., Youssef, H., Boano, C.A., Alves,
M.: Radio link quality estimation in wireless sensor networks: A survey. TOSN 8(4), 34
(2012)
9. Baccour, N., Koubaa, A., Noda, C., Fotouhi, H., Alves, M., Youssef, H., Zuniga, M., Boano,
C.A., Römer, K., Puccinelli, D., Voigt, T., Mottola, L.: Radio Link Quality Estimation in
Low-Power Wireless Networks. Springer Briefs in Electrical and Computer Engineering.
Springer (2013). DOI 10.1007/978-3-319-00774-8
10. Bicen, A., Akan, O., Gungor, V.: Spectrum-aware and cognitive sensor networks for
smart grid applications. Communications Magazine, IEEE 50(5), 158–165 (2012). DOI
10.1109/MCOM.2012.6194397
11. Bilgin, B.E., Çagri Güngör, V.: Performance evaluations of zigbee in different
smart grid environments. Computer Networks 56(8), 2196–2205 (2012). DOI
10.1016/j.comnet.2012.03.002
12. Bilgin, B.E., Gungor, V.C.: On the performance of multi-channel wireless sensor networks
in smart grid environments. In: ICCCN, pp. 1–6 (2011)
13. Boulis, A.: Castalia: revealing pitfalls in designing distributed algorithms in wsn. In:
Proceedings of the 5th international conference on Embedded networked sensor systems,
pp. 407–408. ACM (2007)
14. Downard, I.T.: Simulating sensor networks in NS-2. Tech. rep., DTIC Document (2004)
Wireless Sensor Network based Smart Grid Communications... 21

15. Dunkels, A., Gronvall, B., Voigt, T.: Contiki-a lightweight and flexible operating system
for tiny networked sensors. In: Local Computer Networks, 2004. 29th Annual IEEE Inter-
national Conference on, pp. 455–462. IEEE (2004)
16. Erol-Kantarci, M., Mouftah, H.T.: Wireless multimedia sensor and actor networks for the
next generation power grid. Ad Hoc Networks 9(4), 542–551 (2011)
17. Erol-Kantarci, M., Mouftah, H.T.: Wireless sensor networks for cost-efficient residential
energy management in the smart grid. IEEE Trans. Smart Grid 2(2), 314–325 (2011)
18. Erol-Kantarci, M., Mouftah, H.T.: Smart grid forensic science: applications, challenges,
and open issues. IEEE Communications Magazine 51(1), 68–74 (2013)
19. Erol-Kantarci, M., Mouftah, H.T.: Energy-efficient information and communication infras-
tructures in the smart grid: A survey on interactions and open issues. IEEE Communica-
tions Surveys and Tutorials 17(1), 179–197 (2015). DOI 10.1109/COMST.2014.2341600
20. Fadel, E.A., Gungor, V.C., Nassef, L., Akkari, N., Malik, M.G.A., Almasri, S., Akyildiz,
I.F.: A survey on wireless sensor networks for smart grid. Computer Communications 71,
22–33 (2015). DOI 10.1016/j.comcom.2015.09.006
21. Fan, Z., Kalogridis, G., Efthymiou, C., Sooriyabandara, M., Serizawa, M., McGeehan,
J.: The new frontier of communications research: Smart grid and smart metering. In:
Proceedings of the 1st International Conference on Energy-Efficient Computing and
Networking, e-Energy ’10, pp. 115–118. ACM, New York, NY, USA (2010). DOI
10.1145/1791314.1791331
22. Fan, Z., Kulkarni, P., Gormus, S., Efthymiou, C., Kalogridis, G., Sooriyabandara, M.,
Zhu, Z., Lambotharan, S., Chin, W.H.: Smart grid communications: Overview of research
challenges, solutions, and standardization activities. IEEE Communications Surveys and
Tutorials 15(1), 21–38 (2013)
23. Fang, X., Misra, S., Xue, G., Yang, D.: Smart grid - the new and improved power grid: A
survey. IEEE Communications Surveys and Tutorials 14(4), 944–980 (2012)
24. Fateh, B., Govindarasu, M., Ajjarapu, V.: Wireless network design for transmission line
monitoring in smart grid. IEEE Trans. Smart Grid 4(2), 1076–1086 (2013)
25. Gao, J., Xiao, Y., Liu, J., Liang, W., Chen, C.L.P.: A survey of communication/networking
in smart grids. Future Generation Comp. Syst. 28(2), 391–404 (2012)
26. Grilo, A., Casaca, A., Pereira, P.R., Buttyán, L., Gonçalves, J., Fortunato, C.: A wireless
sensor and actuator network for improving the electrical power grid dependability. In:
NGI, pp. 71–78 (2012)
27. Guglielmo, D.D., Brienza, S., Anastasi, G.: IEEE 802.15.4e: A survey. Computer Com-
munications 88, 1–24 (2016). DOI 10.1016/j.comcom.2016.05.004
28. Gungor, V., Sahin, D.: Cognitive radio networks for smart grid applications: A promising
technology to overcome spectrum inefficiency. Vehicular Technology Magazine, IEEE 7(2),
41–46 (2012). DOI 10.1109/MVT.2012.2190183
29. Gungor, V.C., Korkmaz, M.K.: Wireless link-quality estimation in smart grid environ-
ments. International Journal of Distributed Sensor Networks 2012 (2012)
30. Gungor, V.C., Lu, B., Hancke, G.P.: Opportunities and challenges of wireless sensor net-
works in smart grid. IEEE Transactions on Industrial Electronics 57(10), 3557–3564
(2010)
31. Gungor, V.C., Sahin, D., Kocak, T., Ergüt, S., Buccella, C., Cecati, C., Hancke, G.P.:
Smart grid technologies: Communication technologies and standards. IEEE Trans. Indus-
trial Informatics 7(4), 529–539 (2011)
32. Gungor, V.C., Sahin, D., Kocak, T., Ergüt, S., Buccella, C., Cecati, C., Hancke, G.P.:
A survey on smart grid potential applications and communication requirements. IEEE
Trans. Industrial Informatics 9(1), 28–42 (2013)
33. Henderson, T.R., Lacage, M., Riley, G.F., Dowell, C., Kopena, J.: Network simulations
with the NS-3 simulator. SIGCOMM demonstration 15, 17 (2008)
34. Ho, Q., Gao, Y., Rajalingham, G., Le-Ngoc, T.: Wireless Communications Networks for
the Smart Grid. Springer Briefs in Computer Science. Springer (2014). DOI 10.1007/978-
3-319-10347-1
35. std IEEE 802.15.4e: Part. 15.4: Low-rate wireless personal area networks (lr-wpans)
amendment 1: Mac sublayer) (2012)
36. Incel, O.D.: A survey on multi-channel communication in wireless sensor networks. Com-
puter Networks 55(13), 3081–3099 (2011)
37. Jaikaeo, C., Shen, C.C.: Qualnet tutorial. Retrieved Jan 6 (2005)
38. Khan, R.H., Khan, J.Y.: A comprehensive review of the application characteristics and
traffic requirements of a smart grid communications network. Computer Networks 57(3),
825 – 845 (2013). DOI http://dx.doi.org/10.1016/j.comnet.2012.11.002
22 Sana Rekik et al.

39. Levis, P., Lee, N., Welsh, M., Culler, D.: Tossim: Accurate and scalable simulation of entire
tinyos applications. In: Proceedings of the 1st International Conference on Embedded
Networked Sensor Systems, SenSys ’03, pp. 126–137. ACM, New York, NY, USA (2003).
DOI 10.1145/958491.958506
40. Liberatore, V., Al-Hammouri, A.: Smart grid communication and co-simulation. In: En-
ergytech, 2011 IEEE, pp. 1–5 (2011). DOI 10.1109/EnergyTech.2011.5948542
41. Lin, H., Veda, S.S., Shukla, S.K., Mili, L., Thorp, J.S.: GECO: Global event-driven co-
simulation framework for interconnected power system and communication network. IEEE
Trans. Smart Grid 3(3), 1444–1456 (2012)
42. Long, X., Dong, M., Xu, W., Li, Y.W.: Online monitoring of substation grounding grid
conditions using touch and step voltage sensors. IEEE Trans. Smart Grid 3(2), 761–769
(2012)
43. Ma, R., Chen, H.H., Huang, Y.R., Meng, W.: Smart grid communication: Its chal-
lenges and opportunities. Smart Grid, IEEE Transactions on 4(1), 36–46 (2013). DOI
10.1109/TSG.2012.2225851
44. Mahmood, A., Javaid, N., Razzaq, S.: A review of wireless communications for smart grid.
Renewable and Sustainable Energy Reviews 41, 248–260 (2015)
45. Majumder, R., Bag, G., Kim, K.H.: Power sharing and control in distributed generation
with wireless sensor networks. IEEE Trans. Smart Grid 3(2), 618–634 (2012)
46. Matta, N., Rahim-Amoud, R., Merghem-Boulahia, L., Jrad, A.: A wireless sensor network
for substation monitoring and control in the smart grid. In: GreenCom, pp. 203–209 (2012)
47. Nasipuri, A., Cox, R., Conrad, J., der Zel, L.V., Rodriguez, B., McKosky, R.: Design
considerations for a large-scale wireless sensor network for substation monitoring. In:
LCN, pp. 866–873 (2010)
48. Osterlind, F., Dunkels, A., Eriksson, J., Finne, N., Voigt, T.: Cross-level sensor network
simulation with cooja. In: Local Computer Networks, Proceedings 2006 31st IEEE Con-
ference on, pp. 641–648 (2006). DOI 10.1109/LCN.2006.322172
49. Palattella, M., Accettura, N., Vilajosana, X., Watteyne, T., Grieco, L., Boggia, G., Dohler,
M.: Standardized protocol stack for the internet of (important) things. Communications
Surveys Tutorials, IEEE 15(3), 1389–1406 (2013). DOI 10.1109/SURV.2012.111412.00158
50. Rajalingham, G., Gao, Y., Ho, Q., Le-Ngoc, T.: Quality of service differentiation for
smart grid neighbor area networks through multiple RPL instances. In: Q2SWinet’14,
Proceedings of the 10th ACM Symposium on QoS and Security for Wireless and Mo-
bile Networks, Montreal, QC, Canada, September 21-22, 2014, pp. 17–24 (2014). DOI
10.1145/2642687.2642695
51. Rekik, S., Baccour, N., Jmaiel, M., Drira, K.: Low-Power link quality estimation in smart
grid environments. In: IWCMC 2015 Wireless Sensor Symposium (IWCMC 2015 Wireless
Sensor Symposium). Dubrovnik, Croatia (2015)
52. Rekik, S., Baccour, N., Jmaiel, M., Drira, K.: Holistic link quality estimation-based routing
metric for RPL networks in smart grids. In: 2016 IEEE 27th Annual IEEE International
Symposium on Personal, Indoor and Mobile Radio Communications - (PIMRC): Mobile
and Wireless Networks (IEEE PIMRC2016 Mobile and Wireless) (2016)
53. Sabbah, A.I., Mougy, A.H.E., Ibnkahla, M.: A survey of networking challenges and routing
protocols in smart grids. IEEE Trans. Industrial Informatics 10(1), 210–221 (2014). DOI
10.1109/TII.2013.2258930
54. Sahin, D., Gungor, V.C., Kocak, T., Tuna, G.: Quality-of-service differentiation in single-
path and multi-path routing for wireless sensor network-based smart grid applications. Ad
Hoc Networks 22, 43–60 (2014)
55. Shah, G., Gungor, V., Akan, O.: A cross-layer QoS-aware communication framework in
cognitive radio sensor networks for smart grid applications. Industrial Informatics, IEEE
Transactions on 9(3), 1477–1485 (2013). DOI 10.1109/TII.2013.2242083
56. Shah, G.A., Gungor, V.C., Akan, Ö.B.: A cross-layer design for QoS support in cognitive
radio sensor networks for smart grid applications. In: ICC, pp. 1378–1382 (2012)
57. Sobeih, A., Hou, J.C., Kung, L., Li, N., Zhang, H., Chen, W., Tyan, H., Lim, H.: J-sim:
a simulation and emulation environment for wireless sensor networks. IEEE Wireless
Commun. 13(4), 104–119 (2006). DOI 10.1109/MWC.2006.1678171
58. Sobeih, A., Viswanathan, M., Marinov, D., Hou, J.C.: J-sim: An integrated environment
for simulation and model checking of network protocols. In: IPDPS, pp. 1–6 (2007)
59. Soua, R., Minet, P.: Multichannel assignment protocols in wireless sensor networks:
A comprehensive survey. Pervasive and Mobile Computing 16, 2–21 (2015). DOI
10.1016/j.pmcj.2014.04.004
Wireless Sensor Network based Smart Grid Communications... 23

60. Sun, W., Wang, J.: Cross-layer QoS optimization of wireless sensor network for smart grid.
International Journal of Distributed Sensor Networks (2014)
61. Temel, c., Gungor, V.c., Koçak, T.: Routing protocol design guidelines for smart grid
environments. Comput. Netw. 60, 160–170 (2014). DOI 10.1016/j.bjp.2013.11.009
62. Varga, A., et al.: The omnet++ discrete event simulation system. In: Proceedings of the
European simulation multiconference (ESM2001), vol. 9, p. 65. sn (2001)
63. Wang, W., Xu, Y., Khanna, M.: A survey on the communication architectures in smart
grid. Computer Networks 55(15), 3604–3629 (2011)
64. Wu, Y.C., Cheung, L.F., Lui, K.S., Pong, P.W.T.: Efficient communication of sensors
monitoring overhead transmission lines. IEEE Trans. Smart Grid 3(3), 1130–1136 (2012)
65. Yan, Y., Qian, Y., Sharif, H., Tipper, D.: A survey on smart grid communication infras-
tructures: Motivations, requirements and challenges. IEEE Communications Surveys and
Tutorials 15(1), 5–20 (2013)
66. Yigit, M., Gungor, V.C., Fadel, E.A., Nassef, L., Akkari, N., Akyildiz, I.F.: Channel-aware
routing and priority-aware multi-channel scheduling for wsn-based smart grid applications.
J. Network and Computer Applications 71, 50–58 (2016). DOI 10.1016/j.jnca.2016.05.015
67. Yigit, M., Incel, Ö.D., Çagri Güngör, V.: On the interdependency between multi-channel
scheduling and tree-based routing for wsns in smart grid environments. Computer Net-
works 65, 1–20 (2014)
68. Yigit, M., Yoney, E., Gungor, V.: Performance of mac protocols for wireless sensor net-
works in harsh smart grid environment. In: Communications and Networking (Black-
SeaCom), 2013 First International Black Sea Conference on, pp. 50–53 (2013). DOI
10.1109/BlackSeaCom.2013.6623380
69. Zaker, N., Kantarci, B., Erol-Kantarci, M., Mouftah, H.T.: Smart grid monitoring with ser-
vice differentiation via EPON and wireless sensor network convergence. Optical Switching
and Networking 14, 53–68 (2014). DOI 10.1016/j.osn.2014.01.010
70. Zhu, K., Chenine, M., Nordstrom, L.: ICT architecture impact on wide area monitoring
and control systems’ reliability. Power Delivery, IEEE Transactions on 26(4), 2801–2808
(2011). DOI 10.1109/TPWRD.2011.2160879