Energy efficient information and communication infrastructures in the smart grid a survey on interactions and open issues 1

IEEE COMMUNICATION SURVEYS & TUTORIALS, VOL. 17, NO. 1, FIRST QUARTER 2015 179
Energy-Efficient Information and Communication
Infrastructures in the Smart Grid: A Survey
on Interactions and Open Issues
Melike Erol-Kantarci, Member, IEEE, and Hussein T. Mouftah, Fellow, IEEE
Abstract—Smart grid has modernized the way electricity is
generated, transported, distributed, and consumed by integrating
advanced sensing, communications, and control in the day-to-day
operation of the grid. Electricity is a core utility for the func-
tioning of society and for the services provided by information
and communication technologies (ICTs). Several concepts of the
smart grid, such as dynamic pricing, distributed generation, and
demand management, have significantly impacted the operation
of ICT services, in particular, communication networks and data
centers. Ongoing energy-efficiency and operational expenditures
reduction efforts in communication networks and data centers
have gained another dimension with those smart grid concepts.
In this paper, we provide a comprehensive survey on the smart
grid-driven approaches in energy-efficient communications and
data centers, and the interaction between smart grid and infor-
mation and communication infrastructures. Although the studies
on smart grid, energy-efficient communications, and green data
centers have been separately surveyed in previous studies, to this
end, research that falls in the intersection of those fields has not
been properly classified and surveyed yet. We start our survey by
providing background information on the smart grid and continue
with surveying smart grid-driven approaches in energy-efficient
communication systems, followed by energy, cost and emission
minimizing approaches in data centers, and the corresponding
cloud network infrastructure. We discuss the open issues in smart
grid-driven approaches in ICTs and point some important re-
search directions such as the distributed renewable energy genera-
tion capability-coupled communication infrastructures, optimum
energy-efficient network design for the smart grid environment,
the impact of green communication techniques on the reliability
and latency requirements of smart grid data, workload consolida-
tion with smart grid-awareness, and many more.
Index Terms—Data centers, distributed generation, dynamic
pricing, energy-efficiency, green communications, renewable en-
ergy, smart grid, sustainability.
I. INTRODUCTION
INCREASING electricity prices, diminishing fossil fuels and
rising concerns on GreenHouse Gas (GHG) emissions have
called for the modernization of the power grid, in which Infor-
mation and Communication Technologies (ICTs) are playing
the key role. Smart grid adopts the cutting-edge technology in
Manuscript received November 10, 2013; revised May 17, 2014; accepted
July 15, 2014. Date of publication July 22, 2014; date of current version
March 13, 2015.
The authors are with the School of Electrical Engineering and Computer
Science, University of Ottawa, Ottawa, ON K1N 6N5, Canada (e-mail: melike.
erolkantarci@uottawa.ca; mouftah@uottawa.ca).
Digital Object Identifier 10.1109/COMST.2014.2341600
generation, transmission, delivery, consumption and storage of
electricity. ICTs bring significant innovations in these areas, as
sense, communicate and control functionalities are becoming
the standard practice in smart grid deployments. Meanwhile,
modernization of the grid brings forward new concepts such as
Time of Use (ToU), real-time pricing, distributed generation,
demand management which impact the Operational Expendi-
tures (OPEX) and emissions of the information and commu-
nication infrastructure. In particular, communication networks
and data centers, being the largest power consumers and GHG
emitters among other ICTs, benefit from smart grid-driven
techniques to enhance energy savings and emission reductions.
For the past decade, telecommunication companies have
been seeking ways to reduce their electricity bills as well as
their energy consumption and emissions. The research in the
field of energy-efficient communications targets communica-
tion networks with less OPEX, less power consumption and
less emissions with minimal service degradation. To this end,
vast amount of research has been performed in energy-efficient
wireless, wireline and optical communications. Nevertheless,
ToU, demand management, renewable-dominated supply selec-
tion or briefly smart grid-driven techniques can further bring
down the bills, reduce the amount of energy consumption and
emission release of the operators.
Besides telecommunication companies, data center opera-
tors are impacted by high electricity prices and carbon costs.
Combined electricity demand of telecommunication networks
and data centers is reported to be equivalent to the demand of
the fifth largest country [1]. Therefore, green data centers have
been another major research track in the past several years. To
this end, smart grid-driven techniques have been implemented
through price- or emission-aware workload consolidation and
migration as well as renewable powered data centers.
Implementation of wireless communications and data cen-
ters in connection with the smart grid is illustrated in Fig. 1
including the power flow in the grid. Electricity is transported
to the consumers over the transmission and distribution system
similar to the legacy grid, while in addition, consumers can
become power generators and sell power the grid according
to the novel distributed generation practices of the smart grid
[2]. In the smart grid, all of the entities are connected through
communications. In the figure only wireless technologies are
presented but as will be discussed later on in this paper, using
optical and wireline communications is also possible. In the
figure, data collected from sensors and meters are fed into the
utility headquarters and stored in cloud data centers.
1553-877X © 2014 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission.
See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
www.redpel.com+917620593389

180 IEEE COMMUNICATION SURVEYS & TUTORIALS, VOL. 17, NO. 1, FIRST QUARTER 2015
Fig. 1. Smart grid, wireless wide area solutions and Internet data centers. (Some links and flows have been omitted for the sake of clarity.).
In this paper, we take a closer look at the smart grid-driven
research in energy-efficient information and communication
infrastructures, in particular communication networks and data
centers. The reason for focusing on communication networks
and data centers is their large power intake that positions them
as significant players in the power grid. Our survey differenti-
ates from the other works in the literature, being the initial sur-
vey on smart grid-driven techniques in energy-efficient ICTs.
There are a number of surveys solely on smart grid communi-
cations, each of which focus on different aspects of the topic.
A comprehensive review of smart grid technologies in terms of
infrastructure, management and protection is presented in [3].
In [4] the authors present an informative tutorial on data sensing
technologies for the smart grid. In [5], the authors provide a
comprehensive review on the applications of wireless technolo-
gies in the smart grid while [6], [7] discuss the challenges and
feasibility of state-of-the-art wireless technologies in the smart
grid domain. A broad review of communication technologies
in smart grid can be found in [8]. Smart grid application
requirements in terms of delay, reliability, bandwidth or more
generally Quality of Service (QoS) and the ability of existing
communication technologies to meet those requirements have
been discussed in detail in [9], [10]. Furthermore, recent efforts
on standardization can be found in [11] and [12]. In addition,
surveys that solely focus on energy-efficient communications or
green data centers are also available in the literature. Interested
readers can refer to [13]–[16] for energy efficient wireless
communications, [17], [18] for energy efficient optical commu-
nications, and [19], [20] for green data centers. In this survey,
we focus on the works that enhance the energy efficiency of
the information and communication infrastructures by making
use of the concepts that have come under the spotlight with
the emergence of the smart grid; such as dynamic pricing,
distributed renewable energy generation, fine-tuned fault and
disturbance monitoring and so on. We also point out open re-
search directions, as we believe utilizing smart grid concepts for
enhancing the energy-efficiency of communications and data
centers is a research topic with significant potential. Looking
from another perspective, communications is needed for the
smart grid and energy-efficiency techniques that are employed
in green communications are anticipated to impact the latency
and the reliability of smart grid traffic. Further research is
needed to analyze and quantify those impacts.
The rest of the paper is organized as follows. Section II
provides background on the smart grid and introduces its funda-
mental concepts. Section III surveys the studies that use smart
grid concepts to enhance energy-efficiency of communication
networks including wireless, wireline and optical communica-
tions. Section IV surveys the works that benefit from price,
renewable energy or emission awareness to improve the oper-
ational practices at data centers. Section V discusses the open
issues and lessons learned in both fields with a taxonomy of
surveyed papers. Finally Section VI concludes the paper.
II. AN OVERVIEW OF SMART GRID
Smart grid integrates advanced sensing, communication and
control functionalities in the power grid’s operation, for the
purpose of enhanced efficiency, reliability, security and reduced
emissions [21]. The legacy grid involved sensing, communica-
tion and control functionalities to some extend, however the
advances in ICTs have necessitated bottom-up modernization
of the cyber and the physical elements of the power grid. Smart
grid aims to increase its efficiency by integrating renewables
into the energy mix, employing smart demand management
and by adopting techniques to reduce the losses over the power
lines [22]. Meanwhile, smart grid aims to enhance its reliability
without falling into the pitfalls of the traditional approaches,

EROL-KANTARCI AND MOUFTAH: INFORMATION AND COMMUNICATION INFRASTRUCTURES IN THE SMART GRID 181
Fig. 2. A typical AMI network.
i.e. over provisioning of the resources. Reliability corresponds
to 100% availability of the electricity as well as protection
against power quality fluctuations and operation in isolated
mode in unexpected circumstances such as natural disasters
or terrorist attacks [23]. In addition, smart grid desires to
have improved security. Security of the physical smart grid
infrastructure refers to the security of the utility assets such as
substations, overhead power lines, towers, generators, batteries,
as well as the customer equipment, e.g. smart meters, smart
thermostats that may be controlled by the utility while security
in the cyber domain refers to secure communications and secure
software tools that preserve the corporate, infrastructure and
customer information. The readers who are interested on a
detailed survey on smart grid security are referred to [24], [25].
Advanced communications enable smart grid to device vari-
able pricing policies such as ToU and real-time pricing. The
variable pricing of electricity can be either based on historical
demand data or real-time demand and supply condition. In
either options, electricity has higher price during high demand
than the price set for low demand periods. ToU uses the
seasonal and daily demand information collected over the years
and real-time pricing uses the hour-to-hour market price of
electricity [26]. With the installation of smart meters, variable
pricing and corresponding billing actions have become feasible
in the smart grid.
Variable pricing policies lead to price-following demand
management. In legacy demand response programs, utility op-
erators used to call large-scale customers such as industrial
plant operators or building operators and ask them reduce their
consumption during critical grid conditions by offering some
incentives. This type of demand response is not convenient for
small-scale customers nor for dynamic pricing, therefore in the
smart grid price-following demand management has become
the common practice. In this context, smart grid adopts diverse
communication technologies to support remote smart meter
reading and price-following demand management.
The network of smart meters is known as the Advanced
Metering Infrastructure (AMI). AMI may be implemented us-
ing a wide variety of communication technologies that allow
remote configuration, meter reading and appliance control. For
instance, IEEE 802.15.4g is being developed for the utility-
to-meter communications while IEEE 802.11b/g is another
standard available in many smart meters. Cellular, Worldwide
Interoperability for Microwave Access (WIMAX) and narrow-
band wireless communications are also among the utilized tech-
nologies for AMI. Furthermore, cognitive radio using TV white
spaces has emerged as an alternative to costly cellular com-
munications, especially for coverage in rural areas [27]–[32].
In addition, the technique known as Power Line Communi-
cations (PLC) emerges as a strong candidate for the AMI
[33], [34]. A typical AMI using cellular communications is
presented in Fig. 2. In a broader context, AMI is seen as an ideal
application of Machine-to-Machine (M2M) communications
while M2M communications will also become useful for self-
organization of microgrids [35], remote control of home appli-
ances, interaction of distributed renewable energy generation
resources and so on [36], [37].
As for price-following demand management, the past several
years have seen a tremendous interest in the development of
residential demand management techniques. In general, based
on prediction of consumption patterns, consumer demands are
shifted to less expensive timeslots to reduce energy expenses
of the consumers and to reduce the peak-to-average ratio of the
grid. Several studies apply optimization models [38]–[40] while
others use neural networks [41], [42] or game-theory [43]–[46].
Furthermore, sensor networking technology has been consid-
ered as an integral part of residential demand management [47].
As communication technologies are getting more and more
integrated in the smart grid, reliability of communication and its
impact of smart grid applications has been a relevant research
direction as well [48], [49].
The novelties included in the developing smart grid is not
limited to pricing policies and demand management. Energy
flow from the customers to the grid [50] or energy transactions
among communities [51]–[54] or advanced monitoring solu-
tions for the transmission and distribution systems [55]–[59]

are among the new concepts of the smart grid. For instance,
wide area measurement and situational awareness components
are gaining large deployment opportunities. In the U.S., there
is a momentum for dense deployment of phasor measurement
units (PMUs) and other wide area measurement instruments in
the power grid1
[60]. Close to 1000 PMUs are planned to be up
and running in substations and power plants in the U.S. in the
near future [61].
Efficient integration of renewable energy, in parallel with the
governments’ targets of reducing GHG emissions, is another
driving factor for an enhanced grid. The energy sector is one of
the major GHG emitting sectors, therefore utilizing resources
such as solar, wind, tidal or biomass will significantly reduce its
emissions. Among those renewable resources, solar and wind
power are available in most of the parts of the world while
their contribution to the energy mix of today’s power grids
is fairly low. This is due to their intermittent nature since,
without efficient storage or backup capacity, fluctuation in their
generation output may risk the operation of the grid. According
to a US-based independent system operator (ISO), 9000 MW of
wind generation requires an additional of 1150 MW of capacity
for regulation and ramping purposes [62]. Once again, advance
sensing, communication and control functionalities help the in-
tegration of such intermittent energy resources [63]. For exam-
ple, sensors collect weather-related data, feed them to the grid
operators through near-real-time communications and control
actions are implemented for dispatching more power from other
sources, ramping or load shedding. Besides, the challenges
of renewable energy generation, management of large-scale,
distributed generators emerges as a significant problem where
researchers from the NSF future renewable electric energy de-
livery and management (FREEDM) systems center seek solu-
tions for [64], [65]. In relation to renewable energy integration,
convenient energy storage is essential where advanced storage
technologies are becoming an integral part of the smart grid
[66]. Distribution automation is another important feature that
will transform the legacy grid into a smart grid [67].
Finally, the smart grid technologies are closely related to
Electric Vehicle Charging Infrastructure (EVCI) since Electric
Vehicles (EVs) require a reliable and digitized grid for battery
charging. The past couple of years have seen a significant
amount of studies addressing the challenges of charging EVs
from the power grid, as well as using EV batteries for the
benefit of smart grid. The former is known as Grid-to-Vehicle
(G2V) energy flow while the latter is called Vehicle-to-Grid
(V2G) energy flow [68]. EV charging is challenging because
it is expected to be aligned with the evening peak demand,
adding to the stress on the grid that is already operating close
to its operational limits. Thus, uncontrolled vehicle charging
is overload transformers and harm utility equipment, degrade
power quality, increase power losses and cause voltage prob-
lems [62]. The coordination of vehicle loads, and utilization
1PMUs provide timestamped and synchronized phasor measurements which
demonstrate the direction of power flow in the grid. Monitoring the power
flow is essential since excessive flow towards one direction may unstabilize
the power grid while time synchronization provides a complete view of the grid
at a specific time for interconnections covering multiple time zones.
of vehicle batteries for ancillary services have been studied in
several works such as [69]–[74]. For a more complete survey of
electrification of transportation interested readers are referred
to [75]. The interaction of smart grid with the transportation
industry is out of the scope of this survey.
III. ENERGY-EFFICIENT COMMUNICATIONS
AND THE SMART GRID
Energy-efficiency has been an integral part of networks that
consist of battery-run nodes such as sensor nodes or mobile
phones while energy-efficiency of network equipment powered
from mains such as base stations, switches, routers has not been
under the spotlight until recently. In parallel to the increasing
number of subscribers and their diverse demands, electricity
bills of network operators have been skyrocketing. As a result, a
significant amount of academic and industry efforts have been
put into reducing the energy consumption of core and access
network equipment. Along with the traditional electricity costs,
large amount of GHG emissions of communication networks
is foreseen to add more costs to the operators with the forth-
coming carbon taxes and caps. Price-following demand man-
agement of smart grid can be employed by the communications
infrastructure to reduce electricity bills where this can be also
extended to emission-following load management of the equip-
ment. On one side smart grid-driven approaches impact the way
energy-efficient communication technologies are implemented,
on the other side smart grid involves dense communications and
it is impacted from energy-efficiency techniques. This section
summarizes the works that study the mutual impacts of smart
grid and energy-efficient communications. We have grouped
these studies under wireless, wireline and optical networks and
for each type of network, we focus on the appropriate smart grid
domain. Energy-efficiency of battery-run, hand-held devices
and Wireless Sensor Networks (WSNs) are beyond the scope of
this survey since their energy-efficiency aims at increasing the
lifetime of the network or the device considering their limited
battery power.
The smart grid can be roughly divided into three domains
in terms of communication coverage and functionality; Smart
Grid Home Area Network (SG-HAN), Smart Grid Neighbor-
hood Area Network (SG-NAN) and Smart Grid Wide Area
Network (SG-WAN). SG-HAN is a single residential unit
with smart appliances, an energy display, power consumption
control tools, storage, solar panels, small-scale wind turbines,
electric vehicles and a smart meter. A SG-NAN corresponds
to a group of houses possibly fed by the same transformer.
Advanced Metering Infrastructure (AMI) collects smart grid
data from the premises in a SG-NAN and aggregates the meter
data before they are sent to SG-WAN which connects SG-NANs
to the utility operator. From the utility perspective, besides
metering, equipment in the field needs to be monitored and con-
trolled. Hence the equipment in the field are managed by a sepa-
rate network which is called as Smart Grid Field Area Network
(SG-FAN). The geographical scale of a SG-FAN is similar to
SG-NAN therefore similar communication technologies are ap-
plicable for both. These network domains can be implemented
using a variety of communication technologies. For instance,

due to their wide coverage fiber optic, UMTS, LTE/LTE-A may
be more convenient for SG-WAN while IEEE 802.11, IEEE
802.15.4, PLC would be more suitable for SG-NAN and SG-
HAN. In [76], the authors provide a comprehensive survey on
routing techniques and applications in each of those domains.
In the following sections, we will provide a survey of energy-
efficient wireless, wireline and optical communications tailored
for each smart grid domain.
A. Energy-Efficient Wireless Communications
Wireless communications can be employed in a wide range
of smart grid applications including, meter data collection,
demand management, substation and power line monitoring
and protection. For the sake of clarity, in the next three sections,
we present wireless communications that are used or planned to
be used in SG-WAN, SG-NAN, and SG-HAN, respectively.
1) Energy-Efficient Wireless Communications for SG-WAN:
In wireless communications, energy-efficiency is generally
quantified by the “bits-per-joule” metric which corresponds
to throughput versus unit energy consumption [13]. Here, the
definition of throughput may include the whole transmitted bits
or it may only refer to the actual data that excludes headers,
signaling bits, duplicate packets and transmission errors. Fur-
thermore, the majority of the studies in the domain of energy-
efficiency focus on optimizing transmit power and ignore the
circuit power and power needed for cooling the equipment. Al-
though a holistic metric and an approach is not available in the
literature, even focusing only on energy-efficient transmission
has resulted in significant power savings.
Third generation (3G) and fourth generation (4G) cellular
networks are expected to be densely used in the SG-WAN. They
provide the convenience of using the existing infrastructure and
wide coverage with high data rates. For instance, KEPCO KDN
Co. [77] offers a transmission line surveillance system that uses
laser scans to monitor the surroundings of power lines and
sends an SMS notification in case of an alarm. Another industry
example comes from Qualcomm Inc. and Echelon Inc. who
are using Verizon’s and T-mobile’s infrastructures, respectively,
to provide cellular connectivity for smart meters [78], [79].
Furthermore, with the adoption of high data rate Long Term
Evolution (LTE) and LTE Advanced (LTE-A) technologies, the
cellular networks will be able to carry the high-volume PMU
data. The peak data rates for LTE is around 300 Mbps at the
downlink and 80 Mbps at the uplink with 20 MHz channel
bandwidth and 4 × 4 Multiple Input Multiple Output (MIMO)
antennas. On the other hand, with 70 MHz channel bandwidth
and 4 × 4 MIMO antennas, LTE-A’s targeted peak downlink
transmission rate is 1 Gbps and the uplink transmission rate
is 500 Mbps. In addition to their high speed, LTE and LTE-A
employs the relaying technique to extend the typical cell di-
ameter. Relaying technique is specifically useful for connecting
geographically spread smart meters in rural areas.
Relying technique is considered to be inherently energy-
efficient since the transmitter and the receiver are close to
each other, incurring less transmission power and interference
[13]. Relaying allows for multiple links between a source and
a destination pair which have different channel conditions.
This diversity gain also contributes to energy-efficiency [14].
In particular energy savings increase for environments with
higher path loss [80] making relaying technique ideal for
enhancing the energy-efficiency of SG-WAN communications.
In addition, using mobile relays instead of direct transmission
under delay-tolerant data has been shown to be more energy
efficient [81]. In smart grid, during off-peak hours smart meter
data can be collected in a delay-tolerant manner since demand
management will be less of an issue. In this sense, mobile
relays become useful in SG-WAN to provide meter connec-
tivity. Similarly, cooperative communications provide energy
efficiency because they benefit from channel diversity [13].
Therefore, cooperative communications emerges as another
alternative for energy efficient communication of smart meters.
It is worth to mention that the potential savings from relaying
and cooperative communications have not been exploited in
previous research efforts.
In energy-efficient wireless communication literature,
putting Base Stations (BSs) to sleep during low traffic volume
has been the common approach to save energy. This is relevant
since BSs account for 60–80% of the whole power consumption
in a wireless network [82]. Gong et al. have proposed to switch
on and off a BS based on a blocking probability threshold and
holding the state of a BS for a certain amount of time to avoid
the ping-pong effect [83]. Although this technique has not
been employed in a smart grid environment, similar BS sleep/
wake-up schemes can be useful in the smart grid. Nevertheless,
the nature of the smart grid traffic will be different than human
generated traffic and sleep/wake-up schemes need to take this
into account. Besides traffic volume, QoS requirements of
smart grid applications can interfere with energy-efficiency
techniques.
In [84], the authors have proposed an energy-efficient re-
source allocation scheme for Code Division Multiple Access
(CDMA)-based systems which maximizes each users’ utility
in terms of transmit power and rate while satisfying delay
requirements. In the smart grid, asset monitoring and control of
protection switches require strict delay guarantees while data
collection from the smart meters during off-peak hours can
tolerate delay to a certain extend. For this reason, coordination
of smart meter reports with BS station sleep schedules can yield
further energy savings. However, those coordination schemes
are only convenient for off-peak hours because smart meters
need to report more frequently during peak hours as the energy
consumption data becomes crucial for demand management,
protection and dispatch operations. Although a number of stud-
ies have considered putting BSs to sleep, its impact on smart
grid communications is still an open issue.
Coordinated Multi-Point (CoMP) technology can be also
utilized in SG-WAN applications. Energy-efficiency of Coordi-
nated Multi-Point (CoMP) technology has been studied in the
literature [85]. CoMP relies on separating the conventional BS
into two parts as Baseband unit (BBU) and Remote Radio Unit
(RRU). BBU is responsible for network management, resource
allocation and so on. Multiple RRUs are scattered in the cell
and they offer transceivers close to the users. BBUs and RRUs
are connected via fiber. In [85], the authors explore the optimal
placement of RRUs in terms of energy-efficiency. Furthermore,

[86] investigates energy-efficiency of CoMP and relaying under
an average outage constraint. A use case of CoMP in the smart
grid has been provided in [87]–[90]. In [88], [89], the authors
blend smart grid provided real-time electricity price informa-
tion in the decision of shutting down BSs. Once the selected
BSs are shut off, CoMP is used to attain coverage for active
users. In [90], a Heterogeneous Network (HetNet) environment
has been considered using a dual cognitive approach where
both the spectrum and smart grid resources are allocated in a
cognitive manner. In [87] along with [91] are the seminal works
that integrate smart grid concepts in energy-efficient communi-
cations. In [91], the authors discuss the utilization of renewable
energy in wireless BSs and the implications regarding the
uncertainties. The authors have proposed an adaptive power
management scheme that optimizes the power consumption of
the BS by taking renewable generation output, electricity price
and wireless load profiles into consideration.
In addition to 3G and 4G networks, WIMAX stands as a
strong candidate for SG-WAN communications. To this end,
energy efficiency of Orthogonal Frequency-Division Multi-
ple Access (OFDMA) which is the common multiple access
scheme for all of those networks, has been studied in several
works. In [92], the authors have proposed an energy-efficient
rate adaptation and resource allocation scheme. Another tech-
nique that is common in 3G, 4G and WIMAX is Multiple
Input Multiple Output (MIMO). MIMO increases throughput,
however the circuit power required for multiple antennas incur
high energy consumption. In [93], the authors claim that over
short distances Single Input Single Output (SISO) provides
higher energy-efficiency than MIMO. To this end, adaptively
changing the number of active antennas has been proposed
to provide energy efficiency for MIMO [94]. For instance,
voice and data traffic volume is low during overnight hours,
so is the electricity consumption data fed by the smart meters
and protection related data fed by the sensors since at off-
peak hours it is less likely to have overloaded transformers
and circuits, hence sensors are expected to report less events.
Thus, switching to SISO for overnight hours and then back
to MIMO for daytime may reduce circuit power in SG-WAN
communications. The implications of adaptively changing the
number of antennas in a smart grid environment has not been
explored yet. Its impacts on the QoS requirements of smart grid
applications remains as an open problem.
Finally, cognitive radio technology as described in the IEEE
802.22 standard defines the access of unlicensed users to TV
white spaces. According to the cognitive radio concept, a
secondary user (unlicensed user) with a cognitive radio has the
ability to sense the unused spectrum, use it and then vacate it as
soon as the primary user (licensed user) arrives. The use of cog-
nitive radio in SG-WAN has been motivated in [27] where the
range of IEEE 802.22 standard is between 33 km and 100 km,
which makes it suitable for the SG-WAN, especially for meter
readings from remote regions. Furthermore, in [29], the authors
have proposed a cognitive network that dually senses the radio
spectrum and the smart grid environment. Operating cost of
the cognitive network is optimized under real-time pricing
where macro and femto cell base station adjust their power
consumption based on electricity prices. Although energy con-
servation has not been the main focus of those studies, they
are either proposing to use cognitive radio in the smart grid
or utilizing smart grid concepts in cognitive networks. To this
end, energy efficiency of CR has been studied in [95]. The
authors focus on the joint optimization of the medium access
and physical layers to maximize energy efficiency. Nonetheless,
using cognitive radio as a part of smart grid-driven energy-
efficiency efforts has not been considered in previous studies.
Open issues in smart grid-driven approaches in energy-efficient
wireless communications and the impacts of energy-efficiency
techniques on the QoS of smart grid data will be discussed in
more detail in Section V.
2) Energy-Efficient Wireless Communications for SG-NAN:
SG-NAN is one of the crucial domains that glue utility net-
works with the customer networks. It carries a large volume
of heterogenous data and supports a large number of devices
[96]. Similar to SG-WAN, 3G, 4G and WIMAX can be utilized
in the SG-NAN. In addition, IEEE 802.11 family of standards
offers promising deployments for urban SG-NANs.
Recently, several IEEE 802.11 standards have adopted power
saving mode (PSM) in their operations [97]. IEEE 802.11b and
IEEE 802.11s utilize PSM to allow wireless nodes sleep when
they are not receiving or transmitting. IEEE 802.11b is widely
used for residential premises therefore it is also preferred for
SG-HANs while IEEE 802.11s, the mesh standard has been
considered as a promising candidate for the electric vehicle
charging network. In [73], the authors study the performance
of IEEE 802.11s in an admission control scenario for elec-
tric vehicles. PSM can be considered for IEEE 802.11 based
SG-NAN communications however the drawback of the PSM is
the additional delay which is the common tradeoff in all sleep-
based schemes [98]. Delay-tolerant SG-NAN data may not be
impacted from PSM however, communications for controlling
real-time operations may experience service degradation. For
instance, in an electric vehicle charging network connected via
IEEE 802.11s, increased delay due to sleeping nodes can cause
severe damages to the distribution system if a control command
from the utility that restricts charging happens to be delivered
with extensive delay. For this reason, the impacts of PSM over
the smart grid operation needs to be further explored.
3) Energy-Efficient Wireless Communications for SG-HAN:
Zigbee is one of the widely adopted smart home and smart
energy standards in SG-HANs. Presently, there are various
Zigbee certified products for home automation and several
smart meter vendors such as Landis+Gyr, Itron and Elster have
manufactured Zigbee-enabled smart meters. ZigBee Alliance
has also developed Smart Energy Profile (SEP) to support the
needs of smart metering and AMI, and provide communication
among utilities and household devices.
Zigbee is a short-range, low-data rate wireless technology
that is based on the IEEE 802.15.4 standard. Zigbee has been
initially designed for power-constrained sensor networks, there-
fore energy efficiency is an intrinsic property of Zigbee. It em-
ploys duty-cycling to increase the network lifetime. In [47], the
authors utilize a Zigbee-based WSN for demand management
in the SG-HAN.
Wi-Fi is another strong candidate for wireless SG-HAN
communications. Utilization of Wi-Fi based sensors in the

smart grid has been studied in a recent work [99]. Particularly,
newly emerging ultra low power Wi-Fi chips are expected to
increase the adoption of Wi-Fi for WSNs and increase their in-
teroperability at SG-HAN. Energy-efficiency in WSNs has been
studied vastly in the literature, however those works are out of
the scope of this survey since they are independent of smart grid
concepts such as dynamic pricing, demand management, etc.
In addition to Zigbee and Wi-Fi, cellular technologies may
provide connectivity to residential premises with the deploy-
ment of small cells. Indoor picocells and femtocells may be
useful in reducing the energy consumption of the transmitters
in SG-HAN as they are less prone to path loss than macrocells
[100]. It has been shown that joint deployment of macro- and
publicly accessible residential picocells can reduce the total
energy consumption of a network by up to 60% in urban areas
[101]. Furthermore, energy-efficiency increases as the density
of small cells increase [102]. Thus, implementing SG-HANs
with small cells may contribute to the overall energy-efficiency
of the cellular network. To this end, optimal energy efficient
network design in a smart grid environment has not been
explored yet. Furthermore, a comparison of SG-HAN technolo-
gies in terms of their energy-efficiency has not been made.
B. Energy-Efficient Wireline Communications
Wireline communication technologies may be used to im-
plement a SG-HAN and SG-NAN. In this context, PLC and
Energy Efficient Ethernet (EEE) are among promising wireline
communications.
PLC uses low voltage power lines as the communication
medium for data. PLC is already being used by some utilities
for load control and remote metering. Energy efficiency of the
PLC has been studied in [103] and a spectrum sensing scheme
has been introduced to reduce its power consumption. In [104],
the authors have extended this work to include resource allo-
cation. Transmission parameters have been selected carefully
to mitigate interference in contrast to the former approach of
passively avoiding interference. Interference mitigation relies
on a signal detection function whose durations are optimized
with a limitation on power budget. Another energy-efficient
PLC scheme has been proposed in [105] where the authors
have energy-efficiency as a bound for infinite transmissions
and introduce a green resource allocation scheme. The scheme
optimizes the quantity of information allocated to each chan-
nel when parallel channels are available. As a PLC standard
tailored for smart grid applications, HomePlug Alliance has
recently adopted power saving mode within its Green PHY 1.1
definition. A central coordinator is responsible for sleep/wake
up schedules of connected devices in order to avoid devices
waking up at different times [106].
Energy Efficient Ethernet is also among the wireline tech-
nologies that are being considered for SG-HANs. As described
in the IEEE 802.3az-2010 standard, EEE uses low-power cycles
to save energy. During the low-power mode, short periodic
refresh intervals are defined to maintain alignment between
the transmitter and the receiver. For a low link utilization of
25%, it has been shown that EEE reduces power consumption
by 25% [107].
Although wireline technologies are promising and in partic-
ular PLC is considered to be a natural communication medium
for smart grid, smart grid-driven techniques have not found its
way into energy-efficiency approaches in PLC. PLC is known
to be impacted by the disturbances in low-voltage network
[108]. For instance, energy-efficient resource allocation with
disturbance awareness has not been explored yet.
C. Energy-Efficient Optical Networks
Fiber optical networks offer high speed, large bandwidth and
high degree of reliability that has enabled them to be widely
deployed as the basic physical network infrastructure. Optical
networks have been traditionally implemented in a hierarchical
structure consisting of core, metro and access domains. In
general, a core network provides nationwide coverage with a
few hundreds to a few thousands kilometers of optical fibers. A
metro network provides coverage for a metropolitan area with
optical fibers of a few tens to a few hundreds of kilometers
length. Meanwhile access networks reach the customer and run
between the service provider and the customer which typically
uses optical fibers of a few kilometers in length. Energy-
efficiency has found its way into all of those three domains
including energy-efficient components, transmission, network
design or applications. Optical communications may be used in
SG-WAN or SG-NAN where the former is implemented by core
technologies and the latter with metro or access technologies.
SG-HANs can be implemented by visible light technology
which is a fairly new, high data rate communication technology
[109]. However, this is out of the scope of this survey since
energy-efficiency of visible light communications nor its use in
smart grid have been explored so far.
1) Energy-Efficient Optical Communications for SG-WAN:
A single core network equipment is one of the most power
hungry elements of the telecom optical networks [17]. A core IP
router port consumes approximately 1000 W, whereas the max-
imum power consumption of an optical network unit (ONU)
operating in the access network is around 5 W [110]. Therefore
significant research efforts have been put into addressing the
energy efficiency of the core network.
In smart grid, information flow between ISOs in distant
regions utilizes the core network infrastructure. For instance,
wide area measurements taken under the control of an ISO, can
be delivered to another ISO for protection purposes. Although
smart grid data will be carried over the same core infrastructure
with all other traffic and it is difficult to have smart grid specific
energy-efficiency measures, energy-efficiency techniques will
have direct impact on the QoS of smart grid applications. For
this reason, this section will cover general energy-efficiency
techniques in core optical networks and include several studies
that consider the renewable energy and ToU-based energy-
efficiency improvements. Since utilization of renewable energy
and ToU are part of the big smart grid picture, those works stand
out as examples of the interaction of smart grid and energy-
efficient core optical communications.
In the core networks, there are several opportunities to reduce
energy consumption of network equipment. Various studies
have proposed turning off idle network equipment, such as

switches, line cards or the links, during low traffic volume
[111], [112]. The fundamental challenge in turning the equip-
ment off is guaranteeing the same QoS with the case when
all equipment are operational, as well as responding to the
variations in traffic volume in a fast and an efficient manner.
Recent research has shown that network design is one of the
significant factors impacting the energy-efficiency of the core
networks. Considering this, in [113], the authors have proposed
optimizing the physical topology to increase energy-efficiency.
In IP-over-WDM networks, it has been shown that lightpath
bypass technique is more energy-efficient than lightpath non-
bypass technique [114]. In [115], the authors have addressed
provisioning survivable demands with minimized power con-
sumption in a Wavelength-division multiplexing (WDM) net-
work using lightpath bypass approach. In [116], the authors
have proposed a multi-path selection approach to minimize the
energy consumption of the optical core network. Moreover,
energy-efficient traffic grooming has been shown to reduce the
core network energy consumption [117] while multi-granular
switching has been claimed to provide further energy savings
[118]. Finally, energy-aware routing has been proposed as an
alternative to the traditional shortest path routing in [119]. The
fundamental challenge of the previously proposed schemes is
that they consider human-generated traffic when for instance
turning off network equipment or taking energy-efficient rout-
ing decisions whereas smart grid traffic is mostly machine-
generated and it may suffer from undesired QoS degradation.
Although more research efforts are needed to investigate the
impact of energy-efficient core optical networks on the SG-
WAN traffic, smart grid related concepts such as ToU and
distributed renewable energy generation have made their way
through core optical networks.
The interaction of smart grid with energy-efficient communi-
cations appear explicitly in studies that make use of renewable
energy and ToU. In [120], the authors have formed a Linear
Programming (LP) model to optimize the use of renewable en-
ergy in the IP-over-WDM network which can be considered as a
power grid-driven approach since it allows selecting the source
of electricity. The authors’ findings show that non-renewable
energy consumption is further reduced when renewable energy
is employed in the nodes at the center of the network rather
than the nodes at the edge of the network. Similarly, in [121]
network equipment is powered from renewable resources to
reduce GHG emissions while routing is also performed ac-
cordingly. Nodes with better access to renewable energy are
put on the preferred routes. Integration of distributed renew-
able energy generators to core optical networks and renewable
energy-aware planning and routing has considerably reduced
the amount of emissions [120].
In addition to utilizing renewable energy, reducing OPEX of
the operators using TOU electricity prices is another aspect of
the interaction between the smart grid and the optical networks
networks. TOU-aware routing and wavelength assignment have
been considered initially in [122]. The authors have proposed
an ILP formulation to reduce the electricity bills of a large
network covering multiple time zones. ToU-aware routing and
wavelength assignment directly impact the OPEX of network
operators and provide an opportunity to reduce their electricity
bills. On the other hand, ToU-awareness implies using off-peak
electricity generators which are generally less GHG emitting
than on-peak generators. In power grids, base load supplying
generators are a mix of nuclear, hydro or in some places coal
while peak load supplying resources are diesel generators or
alikes which respond fast to load variations but have high GHG
emissions. It is worth to mention that ToU and energy-efficiency
is loosely coupled, i.e. ToU-aware approaches shift the loads to
off peak hours rather than conserving energy or using renewable
energy. In this sense, ToU-awareness provides reduced electric-
ity bills and it can also lead to reduced emissions if off-peak
generators are cleaner than on-peak generators.
2) Energy-Efficient Optical Communications for SG-NAN:
Most of the smart grid communications is within metro and
access networks reach. Utility headquarters and offices are
already connected via fiber optical networks while with the
fast adoption of Fiber-to-the-Home (FTTH) technology, the
number of residential subscribers are increasing. In addition,
hybrid technologies are emerging as promising candidates for
metro and access convergence. Thus, metro and access op-
tical networks emerge as a strong alternative for SG-NAN
communications.
Although a single access network equipment consumes much
less power than a core router port, access equipment are large
in quantity, hence they consume significant amount of power
in total. According to [18], access networks contribute to 70%
of power consumption of all networks. For this reason, energy-
efficiency of access networks has been an active research area
in the past several years. According to [123], Passive Optical
Networks (PONs) and point-to-point optical networks have
been found to be the most energy-efficient access technology in
comparison to digital subscriber line (DSL), hybrid fiber coax-
ial (HFC) networks, fiber to the node (FTTN), Universal Mobile
Telecommunications System (UMTS) using wideband wide-
band code-division multiple access (W-CDMA) and WiMAX.
Further energy savings in Ethernet Passive Optical Networks
(EPONs) have been studied in several works. The general
approach has been putting ONUs to sleep similar to putting BSs
to sleep in wireless networks. ONUs can be put to sleep either
depending on the traffic volume or link speeds between ONUs
and Optical Line Terminal (OLT) [124]. Gigabit PON (GPON)
which is another PON technology, have seen energy-efficiency
efforts as well, which have lead to standardization. Power
conservation standard in [125], describes the implementation of
sleep mode for ONUs. Furthermore, for a more general PON, a
modified ONU architecture has been proposed in [126] which
reduces the clock recovery overhead when waking up from
sleep mode. EPON has been utilized for smart grid monitoring
applications in [127]–[129] which will be covered later in
this section.
Long-reach PON (LR-PON) technology which is another
recent optical technology in metro and access consolidation,
can provide a backbone for most high speed utility networking
needs. Energy-efficiency of LR-PON has been initially studied
in [130] where the authors have proposed to put transmitters
to sleep that become idle upon a dynamic wavelength allo-
cation scheme. In summary, putting an access unit to sleep
completely or partially have formed the fundamental focus

Fig. 3. Fiber-WSN Architecture [129].
of energy-efficiency in access networks in optical domain.
Although the use of EPON in SG-NAN has been studied [129]
before, to this end, the impact of energy-efficiency techniques,
such as putting ONUs to sleep, on the SG-NAN traffic has
not been researched. Similar to wireless networks and core
optical technologies, more research is needed to explore how
the machine-generated SG-NAN traffic is impacted under those
energy-efficiency techniques.
In addition to pure optical networks, optical-wireless net-
works are also becoming strong candidates for SG-NAN com-
munications. Optical-wireless networks combine the speed and
the reliability of optical networks with the ubiquity, coverage
and flexibility of wireless networks. In this context, Fiber-
Wireless (Fi-Wi) networks are expected to be adopted in the
SG-NAN. A comprehensive survey on energy-efficient Fi-Wi
networks can be found in [110]. An interesting approach in
Fi-Wi networks proposes to put ONU modules of a joint ONU-
BS node to sleep, while the BS module forwards data to a peer
ONU-BS node when the traffic volume is light [131]. Mean-
while the downlink traffic towards the standby ONU is buffered
at the peer ONU-BS node. The authors have considered
human activity profile while optimizing the sleep schedules
of ONU-BS nodes, however SG-NAN traffic is machine-
generated although coupled with human activity as well. Thus,
it still remains as an open problem to study the degree of
energy-efficiency under different types of smart grid traffic.
Very recently, Fiber-WSN technology has been explored for
possible uses in the smart grid environment. Wireless sensor
networking is a proven technology for monitoring indoor and
outdoor facilities while fiber provides fast and reliable con-
nectivity. According to a proposed architecture in [127], smart
grid data, such as user energy profiles, equipment status, etc.
are transmitted to the operators through a broadband access
network, i.e. EPON. In the EPON architecture, each distribution
fiber connects an ONU to the OLT where an ONU is either
serving to a Fiber To The Home/Building/Curb (FTTX) user
or a group of residential premises which are monitored by
WSNs. Thus, the Fiber-WSN network provides a combined
solution for SG-HAN broadband access as well as WSN-based
smart grid monitoring in the SG-NAN. Fig. 3 presents a typical
Fiber-WSN network deployment in the smart grid. In [128],
[129], the authors propose a gateway design for the Fiber-WSN
network that allows data prioritization under urgent and non-
urgent smart grid data. Energy efficiency of the integrated Fi-Wi
architecture is challenging because synchronization among
sleeping ONUs and PSM mode of wireless devices require care-
ful design. In addition, energy-efficiency techniques are closely
coupled with QoS of data. If undesired service degradation is
anticipated or a traffic class with higher priority arrives, then
energy-efficiency techniques may be opted out, however in an
integrated network it is a significant problem to choose which
of the networks will opt out, i.e. fiber, wireless or both.
In [132], the authors consider synchronization of energy-
efficiency procedures and propose a smart Access Point (AP)-
ONU synchronization strategy where the ONU sleeps between
two AP beacons. In the smart grid environment, AP beacon
intervals need to comply with smart grid traffic, thus ONU sleep
mode. Dynamically tuning beacon intervals will require novel
signaling schemes. Energy-efficiency of optical networks have
been covered in a wider scope in [133], [134] which analyzes
and models the energy consumption of transport and network
equipment, and in [17] which provides a comprehensive survey
on the studies in energy-efficient optical communications.
IV. GREEN DATA CENTERS AND THE SMART GRID
Information infrastructure along with communication infras-
tructure is a significant consumer of electricity globally [1].

In proportion to growing user demands, emerging video and
data-intensive applications, as well as the increasing popularity
of cloud computing, power consumption of data centers is
increasing at remarkable rates. According to recent reports
[135], on average, annual energy cost of a 400 W server is
around $800 and a data center houses thousands of those servers
while these numbers are expected to escalate with the growing
popularity of applications that utilize cloud data centers [136].
Energy-efficient or green data centers2
become essential
for reducing the energy expenses of the data center service
providers as well as for reducing the environmental impact of
the data centers. Earlier this year, cap-and-trade rules became in
effect in the state of California which allow large electric power
plants and large industrial plants sell and buy carbon allowances
[137]. The carbon market is expected to grow in the upcoming
years and data center operators will be eventually faced with
the extra costs of carbon emissions. In addition, data centers put
stress on the power grid and the utility that they are being served
by. Therefore, green design of a data center is also important
from the power grid perspective.
Research on green data centers make use of smart grid
concepts such as the option to choose a greener supply, work-
load migration towards greener data centers, electricity price-
based and emission-aware workload migration and so on. In
this section, we first introduce the fundamental terminology in
energy-efficient data centers and then survey the techniques that
use smart grid concepts to help greening the data centers.
A. Energy-Efficiency Metrics for Data Centers
Data center energy efficiency is generally measured by
Power Usage Efficiency (PUE) that is given by:
PUE =
IT Equipment energy + Facility Overhead Energy
IT Equipment energy
(1)
The ideal PUE is 1.0 which corresponds to zero facility
overhead [139]. Facility overhead is related to cooling, lighting
and so on. As an example, Google claims that their data centers
incur around 85% less facility energy consumption and around
15% less IT equipment energy consumption than a typical data
center. This implies that there is still room for more energy-
efficient IT operations. On the other hand, PUE metric is not
enough to represent the energy-efficiency of IT equipment.
Power consumption of IT equipment can be measured by the
data collected from server inlets, power consumption levels
for rack servers, blade servers, and power distribution units
(PDUs), as well as the uninterrupted power supplies (UPSs).
Recently, WSNs have been used for data center monitoring
2In this paper, we use the term Green Data Centers (GDC) to refer to
data centers that adopt energy-efficiency techniques or those that make use of
renewable energy. However, energy-efficiency may not always mean green or
low-emission data center. For instance, an energy-efficient data center powered
from the Texas power grid (which is dominated by coal and natural gas) may
have more emissions than a non-energy-efficient data center in Denmark that is
powered by wind energy. To give an idea on the amount of emissions, according
to [138], migrating services of a 48-core multiprocessor server system from
Alberta’s fossil dominant power grid to a greener region may save 28.5 tons of
GHG emissions annually.
Fig. 4. Energy consumption of a data center.
in the Microsoft Genome Project [140]. Measurements from
sensors are used to plot a thermal map of the data center. In
[141], the authors collect sensor data to perform thermal and
energy aware resource provisioning.
Recently, standardization efforts around sustainable data cen-
ters are focusing on defining key performance indicators (KPI)
for data centers. ISO 30134 standard which is expected to be
released by the end of this year, will define the KPIs for data
centers. Along with PUE, IT Equipment Utilization (ITEU),
IT Equipment Energy Efficiency (ITEE) and Green Energy
Coefficient (GEC) are among the considered metrics to measure
the energy-efficiency of a data center [142].
ITEE refers to the use of IT equipment with higher energy-
efficiency and is defined as:
ITEE =
IT equipment rated work capacity
Rated Energy Consumption of IT equipment
(2)
ITEU refers to the efficient use of IT equipment through con-
solidation and virtualization and is given by:
ITEU =
Measured energy consumption of IT equipment
Rated Energy Consumption of IT equipment
(3)
GEC corresponds to the utilization of renewable energy re-
sources which are generated on-site by the data center operators
rather than the green energy supplied by the grid. GEC is
defined as:
GEC =
Energy input from renewable resources
Energy consumption of data center
(4)
These metrics can be combined into Data center Performance
Per Energy (DPPE) metric which is computed as:
DPPE
=
IT equipment work rate x IT equipment work capacity
Energy consumption of data center − Green energy
(5)
In the context of smart grid, ITEU, GEC and DPPE seem to
be the relevant metrics since price- or emission-aware workload
consolidation and renewable energy usage impact only those.
Approximate energy consumption of a data center is presented
in Fig. 4 for illustrative purposes. The data is derived from [143]
and [144]. As it is seen from the figure, servers in a data center
incur the lion’s share in power consumption.

In addition to the abovementioned metrics, eBay has devel-
oped a metric to quantify its buy/sell transactions. According to
a white paper from eBay that is published in March 2013, Digi-
tal Service Efficiency (DSE) metric computes power consump-
tion per transaction [145]. DSE can be seen as the equivalent
of the “bits-per-joule” metric used in energy-efficient commu-
nications. This metric can be extended to Green Digital Service
Efficiency (GDSE) to compute the amount of renewable power
per transaction once total renewable energy input to a data
center is known.
B. Smart Grid-Aware Energy-efficiency Techniques in
Data Centers
Data centers benefit from ToU prices, distributed renewable
energy generation and similar smart grid related concepts to
reduce their energy intake, expenses and emissions. In this
section, for the sake of presentation clarity, we group the
energy-efficiency efforts in data centers under three titles and
later discuss their interaction with the smart grid:
• Energy-efficient facility management: The actual fa-
cility that houses the servers, UPS, cooling equipment,
operations room and so on, is known as the data center
facility. Most of the power consumed by a data center
facility is spent for cooling the servers and the UPS. The
rule of thumb for facility management is to organize hot
isles and cold isles appropriately. Cold air intake of two
rows of servers, i.e. rack fronts, faces each other forming a
cold isle while hot air exhaust of two rows of servers face
each other forming hot isles. Those isles are isolated by
containments such as plastic covers. Raising floors, run-
ning cooling equipment at higher temperatures have been
among the other techniques that provide energy savings.
In addition, placing UPS in a separate containment has
been also implemented at Google’s data centers [139]. To
analyze the effectiveness of the implemented techniques
Computational Fluid Dynamics (CFD) is used to model
the airflow and the heat transfer in data centers. In general,
energy-efficiency of a data center facility is considered
independent of smart grid concepts. Only recently, several
ideas have emerged from the industry advocating to use
eco-modes of UPSs during high electricity prices or to
recharge UPSs during overnight hours similar to electric
vehicle charging [146], [147].
• Energy-efficiency techniques for IT equipment:
Energy-efficiency of IT equipment is implemented
through energy-efficient server hardware as well as
virtualization and dynamic workload consolidation [19].
Although data centers contain switches for intra data
center networking, the total power consumption of core,
aggregation and access switches is less than half of
server power consumption approximately as seen in
Fig. 4. Therefore most of the efforts in the literature
focus on energy-efficiency of servers. One of the primary
energy-efficiency techniques for server hardware is the
Dynamic Frequency and Voltage Scaling (DFVS). DFVS
dynamically adjusts the frequency of the CPU of a
server to save power [148]. Virtualization is a concept
that helps running multiple user demands on a single
physical host by placing more than one virtual machine
on a server. With the combination of virtualization and
dynamic workload consolidation, some servers may be
left idle and put to sleep. This technique is known as
Dynamic Power Management (DPM). To this end, smart
grid-driven DFVS or DPM still remain as open problems.
Since energy efficiency of facility management and IT
equipment has not been investigated from the perspective
of smart grid, we do not include those works and refer the
interested readers to [20] for a comprehensive survey on
energy-efficient facility management, IT equipment and
intra-data center network.
• Energy-efficiency techniques for multiple data centers:
Those techniques are designed for geographically-diverse,
large-scale Internet data centers such as the ones owned
and operated by Google, Yahoo, Microsoft, Amazon,
eBay, etc. The general idea is to route user demands to-
wards one or more data centers either considering heating
and cooling constraints or to minimize cost or to minimize
emissions. In this survey, we are primarily focusing on the
energy-efficiency of such large-scale Internet data centers
since smart grid-driven techniques have been tailored for
those in most of the cases. Energy-efficiency techniques
for multiple data centers can be further grouped under
three categories: 1) Minimizing electricity bills of data
centers; 2) Minimizing energy consumption of inter data
center network and 3) Data centers powered by renewable
energy.
In the following sections, we will focus on energy-efficiency
techniques for multiple data centers and their corresponding
transport networks. Furthermore, we will look into the adoption
of data centers for hosting smart grid services to give a complete
picture of the interactions between the smart grid and data
centers.
1) MinimizingElectricityBillsofDataCenters: Qureshiet al.
have exploited the temporal and the geographic diversity of
electricity prices in the smart grid to reduce the OPEX of data
center operators [149]. When user demand is below the peak,
the authors have proposed to migrate workloads towards data
centers that can take advantage of less expensive electricity
according to local ToU tariffs. During workload mitigation,
user demands may suffer from increased delay, especially if the
assigned data center is geographically distant and overloaded
due to forwarded demands from other data centers. This issue
has been addressed in [150]–[152]. In [150] and [151], load
balancing of Internet data centers has been considered. For this
purpose, workload constraints, a multi-electricity market model
and QoS constraints have been formulated in an optimization
problem. In [152], the authors augment [150] with additional
cost savings resulting from server level design. Furthermore, in
[153] a holistic cost minimization approach has been proposed,
considering DPM, DFVS, workload migration as well as the
cooling systems. The authors claim that 20% of energy ex-
penses could be cut down even under peak workloads by using
their proposed scheme.
In [154], the authors take an interesting approach and con-
sider minimizing operation costs under uncertainty in demand

and electricity price. According to [154], data center operators
buy electricity beforehand (when the price for electricity is low)
to compensate for the fluctuations in workload and electricity
prices. The authors propose a two-level optimization model
which first determines the amount of electricity to buy via
forward contract for each data center and then solves a mixed
integer nonlinear programming problem with the objective of
minimizing the unit electricity cost at each data center as well
as balancing load among data centers. Furthermore, in [155],
the interaction between electricity prices in deregulated markets
and data center energy demand has been explored. The authors
have assumed that the price is a piecewise linear function of
the demand where the price of electricity rises steeply after a
certain demand threshold. Considering this price model, the
authors have aimed to identify an optimum operating regime
for data centers. Workload migration is performed in order
to minimize electricity bills under the assumed smart grid
pricing scheme. All of the above summarized works, use smart
grid pricing concepts to reduce bills, therefore they are ideal
examples of studies that reveal mutual interactions between the
smart grid and green data centers.
Although pricing related workload migration has been ex-
tensively studied, workload migration and its impact on the
local grid is not well understood. In [156], the authors propose
a two stage optimization that first balances the load on the
power grid with data center load and then data center operators
migrate workloads in order to reduce their energy costs. The
interesting approach of this paper is power balancing decisions
of utilities play a role in workload migration in addition to
cost minimization decisions of data center operators. Power
grid interaction with data centers have also led to the use
of data centers as controllable loads in [157]. Data centers
commit to curtail their loads as a response to utility requests.
Load curtailment occurs in the form of migrating workloads to
other data centers that do not need to curtail their loads. When
responding to utility requests workloads need to be migrated in
a timely manner since power grid operates in real-time. Thus
the load on the transport network is taken into account in order
to curtail load within delay bounds. In [157], the authors have
formulated an optimization problem considering the power grid
requirements and network load conditions.
2) Minimizing Energy Consumption of Inter Data Center
Network: With the growing popularity of cloud computing,
computational complexity is shifting from end users towards
the data centers which is reducing the power consumption of
end users but at the same time increasing the power consump-
tion of the data centers as well as the transport network [158].
In this section, we focus on the works that aim to minimize
the energy consumption of the data centers as well as inter
data center networks. The latter is handled by the works in the
previous section.
In [159], the authors study the impact of ToU-awareness on
the OPEX and energy-efficiency of the cloud transport network
infrastructure. The TOU tariff in different locations and traffic
profile of users are exploited in order to minimize the network
and the data center power consumption. The proposed Mixed
Integer Linear Programming (MILP) model helps to migrate
user demands over network nodes that incur less OPEX. In
[160], the authors extend this work to include inter-data center
workload sharing where each data center is allowed to migrate
a certain amount of its workload to several destination data
centers. Delay has been identified as a fundamental tradeoff in
power minimizing or cost minimizing schemes.
In [161]–[163], the authors aim to minimize power consump-
tion as well as make the best use out of renewable energy when
transporting data between data centers. First, the authors deter-
mine the optimal locations of data centers in an IP-over-WDM
network that minimizes the network’s power consumption.
Then, they investigate further savings when renewable energy is
utilized either by generating on-site or transporting through the
smart grid. The authors have reported 73% savings when multi-
hop bypass routing with renewable energy is used together with
a popularity-based replication scheme. Note that, the studies
surveyed in this section, consider using renewable energy at
the network equipment. In the next section, we summarize the
works that use renewable energy at the data centers.
3) Data Centers Powered by Renewable Energy: Islanding
data centers from the utility grid and powering them from
renewable resources whenever sun penetration or wind power
is adequate, offer reduced emissions and OPEX. The GreenStar
Network (GSN) testbed of Canada is a network of renewable-
powered data centers [164]. The GSN testbed aims to explore
the feasibility of powering a network of data centers with
solar and wind power. When there is a shortfall of renewable
energy at one data center virtual machines are migrated towards
another data center if that data center is capable of working
on renewable power and QoS requirements of users can be
met. The virtual machines are transported using the high-speed
Canarie network to tackle latency tradeoff.
Mahdi et al. [165] optimize workload distribution based
on the availability of renewable energy, QoS constraints and
electricity price. They propose the “behind-the-meter” notion
to refer to renewable resources operated at the data center
facilities. The main purpose of those generators is to supply
electricity to the data center, however if surplus power is
generated it can be sold to the grid as well. In [166], the authors
extend their work to take service rates into consideration and
they propose an optimization for profit maximization. When
investing for on-site renewable energy generation is costly, data
center operators can also sign up with a local renewable energy
generator. In [167], the authors have exploited the concept of
futures contracts to allow data centers exchange workloads.
With futures contracts a customer commits to purchase an
amount of renewable power ahead of time. When there is
excessive user demand and the data center utilizes all the futures
contract limit, its workload is migrated to a data center that has
under used renewable energy capacity.
Besides energy-efficiency, emission awareness has been an-
other dimension of the smart grid-driven research in green
data centers. In [168], the authors device a request distribu-
tion policy by taking Service Level Agreements (SLAs) and
“brown energy”3
consumption of the data centers into consider-
ation. The authors incorporate carbon market dynamics in their
3The term brown energy is used to refer to a mix of green and fossil-fuel
based energy.

Fig. 5. FNET Architecture.
optimization model and minimize the overall cost per service.
In [169], the authors have considered various ways of including
renewable energy in data center operations such as renewable
energy certificates and power purchase agreements in addition
to the widely used on-site generation. The authors have con-
cluded that on-site generation can provide carbon reduction to
some extend however off-site generation and power purchase
through agreements is more cost-efficient.
In addition to price, emission or renewable energy aware
workload migration, grid status awareness has been considered
in [170]. The authors have looked into the interaction of the
smart grid and cloud computing from a different perspec-
tive and proposed grid-aware cloud computing that considers
workload migration in the event of power grid failure risk.
For this purpose, the authors have formulated an optimization
problem to solve demand routing jointly with power flow
analysis.
C. Data Centers for Hosting Smart Grid Services
Besides smart grid-driven energy-efficiency techniques for
greening the data centers, the interaction of smart grid and data
centers include the use of data centers at the service of the
power grid. As the amount of data generated by smart grid ap-
plications is increasing rapidly, storage and processing of those
data are shifting from utility servers to the cloud or Internet
data centers. A typical example of a smart grid application with
high volume of data is the power system frequency monitoring
network (FNET). FNET collects timestamped measurements
from the grid using Frequency Disturbance Recorder (FDR)
and provide functionalities such as real-time event alerts, ac-
curate event location estimation, animated event visualization,
and post event analysis [5], [60]. FDR is a type of PMU that
allows single phase measurements. Measurement data from the
FNET are managed in the data center as seen in Fig. 5. Energy-
efficiency techniques employed at the data centers may impact
the access patterns of utilities, yet these interactions have not
been explored yet.
V. OPEN ISSUES AND FUTURE RESEARCH DIRECTIONS
This survey takes a closer look at the smart grid-driven
research in green information and communication systems, in
particular communication networks and data centers. In this
section, we summarize the lessons learned and point out the
open research issues in those fields.
A. Lessons Learned
There is an increasing interest from the academia and the
industry to reduce the OPEX, energy consumption and emis-
sions of ICTs. With the development of smart grid these efforts
gained another dimension as to include smart grid concepts to
improve the overall energy-efficiency and cleanness of com-
munication and information infrastructures. Most of the ideas
have evolved around distributed renewable energy generation,
price dependent and emission aware operation. In the commu-
nication infrastructure putting high energy consuming network
equipment to sleep has been one of the fundamental approaches
while in information infrastructures, more specifically in data
centers, putting less busy servers to sleep has been considered.
Putting an equipment to sleep either during low user load or
high electricity price has brought forward QoS challenges. It is
natural to expect performance degradation when some nodes on
the network or servers in the data center are sleeping. Most of
the approaches have focused on researching operational limits
of latency and packet loss and energy-efficiency. Another fun-
damental approach has been using renewable energy to power
either network equipment or data centers. Those works have
mostly considered using on-site generated energy, usually by
the behind-the-meter generators. Only recently grid integrated
generators or off-site generation has been considered. In the
former data centers are able sell surplus energy and in the
latter case they do not generate energy themselves but they
purchase it from a generator. The latter has been found more
economically feasible in [169]. Another less explored direction
has been responsiveness of ICTs to grid events such as failures,
voltage drops and interaction with other heavy loads like EVs.

In general, researchers’ perspective for smart grid’s interac-
tion with green ICTs has been inclined towards utilizing smart
grid related concepts for the benefit of ICTs. However, another
less explored direction is the impacts of greening ICTs on smart
grid communications and applications. Smart grid is a vast
Machine-to-Machine (M2M) network and it involves intense
communications with devices on utility assets as well as homes.
Energy-efficiency of smart grid communications and its impact
on reliability and latency is one of the open problems.
B. Open Issues in Smart Grid-Driven Research on
Green Communications
In green wireless communications, relaying, cooperative
communications and small cells are recent techniques that are
developed for extending the cell coverage and providing high
data rates for the users within a cell. On the other hand, they are
considered to be energy efficient since they can serve the users
who are far from the macro cell BS, incurring energy savings
for the macro BS. Although they are promising techniques for
the AMI and WAN communications of the smart grid, their
deployment in the smart grid environment is not well explored.
Furthermore, femtocells and picocells are energy efficient alter-
natives to macrocells, however when they are densely deployed,
they may reduce the efficiency of the macrocell as the primary
BS will operate under low traffic [16]. A relevant question is;
what is the optimum network design for cellular smart grid
communications? Since smart grid traffic is machine-generated
and previous research on optimal network design is based
on human-generated traffic, those results may not be blindly
applicable to smart grid domain. Moreover, depending on the
price of electricity utilization of macro cell BS can vary and
alternative techniques may used to provide coverage to the cell
users, as such in [87]. Yet, there are only a few works that
investigate price-awareness in BS operation and corresponding
coverage issues. Therefore, the mutual impact of sleeping BSs
on the QoS requirements of smart grid data needs to be further
investigated. Meanwhile, mean and variance of the traffic load
and the density of BSs is know to impact energy savings under
sleep mode [171]. Thus, how does the traffic generated by smart
grid applications impact the energy efficiency of communi-
cation systems? Smart grid traffic is heterogenous since it is
fed by a variety of devices. For instance, smart meter data is
collected every 10–15 minutes while sensor data is collected in
the order of seconds. Particularly, transformer or capacitor bank
monitoring requires near-real-time data delivery. Under these
traffic types what is an ideal sleep schedule for a BS is another
open problem.
MIMO is an efficient technique to increase throughput,
however it incurs high energy consumption. To overcome this
challenge previous studies have proposed to use SISO during
low traffic intensity. Adaptively changing the number of active
antennas in the smart grid environment is another interesting
research direction. To this end, the implications of adaptively
configuring MIMO in a smart grid environment have not been
explored yet.
In a wireless network, spectral efficiency, cost, cover-
age and QoS are generally conflicting requirements with
energy-efficiency. QoS guidelines for smart grid communica-
tions and their tradeoffs with energy-efficiency techniques need
to be explored. Particularly, SG-NAN carries various types of
data with a large spectrum of QoS requirements as we have
mentioned before. Future research needs to explore the impacts
of energy-efficiency techniques on the QoS of different types
of smart grid data, as well as the bounds for energy-efficiency
under various QoS classes of smart grid data.
In the optical domain, the impact of sleeping ONUs on the
QoS requirements of smart grid data is an interesting research
problem. In addition, in the integrated Fi-Wi architecture de-
veloping efficient mechanisms for sleeping ONUs and APs so
that smart grid data requirements is not violated is an open
problem. Although coordination of ONUs and APs according
to beacon intervals have been considered in [132], scheduling
of intervals in the smart grid environment remains as an open
problem. Furthermore, when there is QoS degradation concern
determining which of the integrated networks, i.e. optical or
wireless, will opt out sleeping is a further research issue.
Finally, in PLC energy-efficient resource allocation schemes
consider ideal channel conditions, however PLC channel is
impacted by the disturbances in the low voltage circuit. Thus,
energy efficient physical and medium access under power dis-
turbances needs to be studied.
C. Open Issues in Smart Grid Motivated Research on Green
Data Centers
In green data centers, workload migration is usually done
based on either price or emission or renewable energy awareness.
Emission aware workload migration is relatively recent and has
seen less attention because carbon market is newly shaping with
the new carbon cap-and-trade regulations. In addition, emission
awareness has been considered only for data centers and not for
the transport network that carry data center traffic. Workload
migration for jointly minimizing emissions of data centers and
their transport network is still an open problem. Furthermore,
the impact of power grid events such as failures, disturbances,
voltage drips has been less explored. Thus, inter-data center
load distribution based on power grid events is an open issue.
Smart grid and data centers are large-scale cyber-physical
systems. As electric vehicles are becoming a conventional
demand for the grid, their heavy loads may cause service
degradation. In that case, load-following data center operations
may be necessary in order to avoid putting too much stress
on the distribution transformers. Those interactions have not
been addressed in the literature, therefore future research needs
to explore green data center operation under local distribution
system loading. In Table I, we provide a taxonomy of all
the surveyed smart grid-driven approaches in this paper. We
decidedly left out the energy-efficiency approaches that are not
related to smart grid. Thus, all of the approaches included in
the table are load or price or emission or supply-aware where
some have additional links to smart grid. Price-awareness and
utilization of renewable energy has often found its way to
energy-efficient ICTs frequently. On the other hand, emission
and grid-awareness have been less explored. Meanwhile, QoS
has been addressed in a considerable amount of works.

TABLE I
A TAXONOMY OF RESEARCH THAT USES SMART GRID CONCEPTS IN ENERGY EFFICIENT ICTs
VI. CONCLUSION
Information and Communication Infrastructures (ICTs) are
among the major power consumers and GreenHouse Gas
(GHG) emitters. In particular, the power consumption of com-
munication networks and data centers in total is comparable
to consumption of large countries. Consequently, the Opera-
tional Expenditures (OPEX) of communication networks and
data centers are growing with increasing electricity prices and
emerging carbon caps and taxes. For this reason, recently vast
amount of efforts have been put into enhancing the energy-
efficiency of ICTs. In addition to traditional approaches, smart
grid-driven techniques have found their way to provide OPEX
reductions. Dynamic pricing schemes, distributed generation,
demand management, fine-tuned monitoring of faults and dis-
turbances are new concepts brought by the smart grid and they
can be effectively utilized to reduce the bills, consumption and
emissions of ICTs.
In this survey paper, we have focused on the studies that
investigate the challenges and opportunities arising from the in-
teraction of the smart grid with green ICTs. We have provided a
comprehensive survey of the smart grid-driven works in energy-
efficient communication networks and data centers. We have
first given a brief introduction to the smart grid and introduced
the fundamental concepts that have emerged with the smart
grid. Next, we have elaborated on the use of energy-efficient
wireless, wireline and optical communications in different do-
mains of the smart grid such as Home Area Networks (HAN),
Neighborhood Area Networks (NAN) and Wide Area Networks
(WAN). Then, we have surveyed the studies that take smart
grid factors into consideration while managing data center’s
power consumption, emissions and electricity bills. Finally,
we have outlined the lessons learned and the open research
issues regarding smart-grid awareness in energy-efficient ICTs.
In addition, we have pointed out new research directions.
REFERENCES
[1] How clean is your cloud? [Online]. Available: http://www.greenpeace.
org/international/Global/international/publications/climate/2012/iCoal/
HowCleanisYourCloud.pdf
[2] G. Zimon, “Integrating distributed generation into the smarter grid,”
IEEE Smart Grid, Dec. 2013. [Online]. Available: http://smartgrid.
ieee.org/december-2013/1013-integrating-distributed-generation-into-
the-smarter-grid
[3] X. Fang, S. Misra, G. Xue, and D. Yang, “Smart grid—The new and
improved power grid: A survey,” IEEE Commun. Surveys Tuts., vol. 14,
no. 4, pp. 944–980, 2012.
[4] D. Niyato, N. Kayastha, E. Hossain, and Z. Han, “Smart grid sensor data
collection, communication, networking: A tutorial,” Wireless Commun.
Mobile Comput., vol. 14, no. 11, pp. 1055–1087, Aug. 2014.
[5] R. Ma, H.-H. Chen, Y.-R. Huang, and W. Meng, “Smart grid communi-
cation: Its challenges and opportunities,” IEEE Trans. Smart Grid, vol. 4,
no. 1, pp. 36–46, Mar. 2013.
[6] P. P. Parikh, M. G. Kanabar, and T. S. Sidhu, “Opportunities and chal-
lenges of wireless communication technologies for smart grid appli-
cations,” in Proc. IEEE Power Energy Soc. Gen. Meet., Jul. 2010,
pp. 1–7.
[7] V. Aravinthan, B. Karimi, V. Namboodiri, and W. Jewell, “Wire-
less communication for smart grid applications at distribution level—
Feasibility and requirements,” in Proc. IEEE Power Energy Soc. Gen.
Meet., Jul. 2011, pp. 1–8.
[8] H. T. Mouftah and M. Erol-Kantarci, “Smart grid communications:
Opportunities and challenges,” in Handbook of Green Information and
Communication Systems, M. S. Obaidat, A. Anpalagan, and I. Woun-
gang, Eds. Amsterdam, The Netherlands: Elsevier, 2012, pp. 631–663.
[9] Y. Yan, Y. Qian, H. Sharif, and D. Tipper, “A survey on smart grid com-
munication infrastructures: Motivations, requirements and challenges,”
IEEE Commun. Surveys Tuts., vol. 15, no. 1, pp. 5–20, 2013.
[10] V. Gungor et al., “A survey on smart grid potential applications and
communication requirements,” IEEE Trans. Ind. Informat., vol. 9, no. 1,
pp. 28–42, Feb. 2013.
[11] Z. Fan et al., “Smart grid communications: Overview of research chal-
lenges, solutions, standardization activities,” IEEE Commun. Surveys
Tuts., vol. 15, no. 1, pp. 21–38, 2013.
[12] P. Chatzimisios, D. G. Stratogiannis, G. I. Tsiropoulos, and
G. Stavrou, “Survey on smart grid communications: From an archi-
tecture overview to standardization activities,” in Handbook of Green
Information and Communication Systems, M. S. Obaidat, A. Anpalagan,
and I. Woungang, Eds. Amsterdam, The Netherlands: Elsevier, 2012,
pp. 665–389.
[13] D. Feng et al., “A survey of energy-efficient wireless communications,”
IEEE Commun. Surveys Tuts., vol. 15, no. 1, pp. 167–178, 2013.

[14] G. Y. Li et al., “Energy-efficient wireless communications: Tutorial,
survey, open issues,” IEEE Wireless Commun., vol. 18, no. 6, pp. 28–35,
Dec. 2011.
[15] Y. Chen, S. Zhang, S. Xu, and G. Y. Li, “Fundamental trade-offs on green
wireless networks,” IEEE Commun. Mag., vol. 49, no. 6, pp. 30–37,
Jun. 2011.
[16] Z. Hasan, H. Boostanimehr, and V. K. Bhargava, “Green cellular net-
works: A survey, some research issues and challenges,” IEEE Commun.
Surveys Tuts., vol. 13, no. 4, pp. 524–540, 2011.
[17] Y. Zhang, P. Chowdhury, M. Tornatore, and B. Mukherjee, “Energy
efficiency in telecom optical networks,” IEEE Commun. Surveys Tuts.,
vol. 12, no. 4, pp. 441–458, 2010.
[18] R. Bolla, R. Bruschi, F. Davoli, and F. Cucchietti, “Energy efficiency
in the future Internet: A survey of existing approaches and trends in
energy-aware fixed network infrastructures,” IEEE Commun. Surveys
Tuts., vol. 13, no. 2, pp. 223–244, 2011.
[19] D. Cavdar and F. Alagoz, “A survey of research on greening data cen-
ters,” in Proc. IEEE GLOBECOM, 2012, pp. 3237–3242.
[20] B. Kantarci and H. T. Mouftah, “Energy-efficiency in cloud data
centers,” in Communication Infrastructures for Cloud Computing.
Hershey, PA, USA: IGI Global, 2013.
[21] S. M. Amin and B. F. Wollenberg, “Toward a smart grid: Power delivery
for the 21st century,” IEEE Power Energy Mag., vol. 3, no. 5, pp. 34–41,
Sep./Oct. 2005.
[22] M. Erol-Kantarci and H. T. Mouftah, “Wireless multimedia sensor and
actor networks for the next generation power grid,” Ad Hoc Netw. J.,
vol. 9, no. 4, pp. 542–511, Jun. 2011.
[23] M. Erol-Kantarci and H. T. Mouftah, “Smart grid forensic science:
Applications, challenges and open issues,” IEEE Commun. Mag.,
vol. 51, no. 1, pp. 68–74, Jan. 2013.
[24] Y. Yan, Y. Qian, H. Sharif, and D. Tipper, “A survey on cyber security
for smart grid communications,” IEEE Commun. Surveys Tuts., vol. 14,
no. 4, pp. 998–1010, 2012.
[25] S. M. Amin and A. M. Giacomoni, “Smart grid- safe, secure, self-
healing: Challenges and opportunities in power system security, re-
siliency, privacy,” IEEE Power Energy Mag., vol. 10, no. 1, pp. 33–40,
Jan./Feb. 2012.
[26] C. M. Ruibal and M. Mazumdar, “Forecasting the mean and the variance
of electricity prices in deregulated markets,” IEEE Trans. Power Syst.,
vol. 23, no. 1, pp. 25, 32, Feb. 2008.
[27] A. Ghassemi, S. Bavarian, and L. Lampe, “Cognitive radio for smart grid
communications,” in Proc. IEEE Int. Conf. SmartGridComm, Oct. 2010,
pp. 297–302.
[28] M. Brew et al., “UHF white space network for rural smart grid com-
munications,” in Proc. IEEE Int. Conf. SmartGridComm, Oct. 2011,
pp. 138–142.
[29] J. Huang, H. Wang, Y. Qian, and C. Wang, “Priority-based traffic
scheduling and utility optimization for cognitive radio communication
infrastructure-based smart grid,” IEEE Trans. Smart Grid, vol. 4, no. 1,
pp. 78–86, Mar. 2013.
[30] R. Deng et al., “Sensing-performance tradeoff in cognitive radio en-
abled smart grid,” IEEE Trans. Smart Grid, vol. 4, no. 1, pp. 302–310,
Mar. 2013.
[31] Y. Zhang et al., “Cognitive machine-to-machine communications:
Visions and potentials for the smart grid,” IEEE Netw., vol. 26, no. 3,
pp. 6–13, May/Jun. 2012.
[32] R. C. Qiu et al., “Cognitive radio network for the smart grid: Experimen-
tal system architecture, control algorithms, security, microgrid testbed,”
IEEE Trans. Smart Grid, vol. 2, no. 4, pp. 724–740, Dec. 2011.
[33] S. Galli, A. Scaglione, and Z. Wang, “For the grid and through the grid:
The role of power line communications in the smart grid,” Proc. IEEE,
vol. 99, no. 6, pp. 998–1027, Jun. 2011.
[34] S. Güzelgöz, H. Arslan, A. Islam, and A. Domijan, “A review of wireless
and PLC propagation channel characteristics for smart grid environ-
ments,” J. Elect. Comput. Eng., vol. 2011, pp. 154040-1–154040-12,
Jan. 2011.
[35] R. H. Lasseter, “MicroGrids,” in Proc. IEEE Power Eng. Soc. Winter
Meet., 2002, pp. 305–308.
[36] Z. M. Fadlullah et al., “Toward intelligent machine-to-machine commu-
nications in smart grid,” IEEE Commun. Mag., vol. 49, no. 4, pp. 60–65,
Apr. 2011.
[37] M. Erol-Kantarci and H. T. Mouftah, “Machine-to-machine communi-
cations in the smart grid,” in Machine-To-Machine Communications—
Architectures, Technology, Standards, Applications, J. Misic and V. Misic,
Eds. New York, NY, USA: Taylor & Francis, 2013, pp. 227–259.
[38] A.-H. Mohsenian-Rad, V. W. S. Wong, J. Jatskevich, and R. Schober,
“Optimal and autonomous incentive-based energy consumption schedul-
ing algorithm for smart grid,” in Proc. ISGT, Jan. 2010, pp. 1–6.
[39] A.-H. Mohsenian-Rad and A. Leon-Garcia, “Optimal residential load
control with price prediction in real-time electricity pricing environ-
ments,” IEEE Trans. Smart Grid, vol. 1, no. 2, pp. 120–133, Sep. 2010.
[40] M. A. A. Pedrasa, T. D. Spooner, and I. F. MacGill, “Coordinated
scheduling of residential distributed energy resources to optimize smart
home energy services,” IEEE Trans. Smart Grid, vol. 1, no. 2, pp. 134–
143, Sep. 2010.
[41] A. Molderink, V. Bakker, M. G. C. Bosman, J. L. Hurink, and
G. J. M. Smit, “Domestic energy management methodology for opti-
mizing efficiency in Smart Grids,” in Proc. IEEE Bucharest Conf. Power
Technol., Jun. 2009, pp. 1–7.
[42] A. Molderink, V. Bakker, M. G. C. Bosman, J. L. Hurink, and
G. J. M. Smit, “Management and control of domestic smart grid tech-
nology,” IEEE Trans. Smart Grid, vol. 1, no. 2, pp. 109–119, Sep. 2010.
[43] A.-H. Mohsenian-Rad, V. W. S. Wong, J. Jatskevich, R. Schober, and
A. Leon-Garcia, “Autonomous demand-side management based on
game-theoretic energy consumption scheduling for the future smart
grid,” IEEE Trans. Smart Grid, vol. 1, no. 3, pp. 320–331, Dec. 2010.
[44] H. K. Nguyen, J. B. Song, and Z. Han, “Demand side management
to reduce peak-to-average ratio using game theory in smart grid,” in
Proc. IEEE INFOCOM Workshop Commun. Control Sustainable Energy
Syst.—Green Netw. Smart Grids, Mar. 2012, pp. 91–96.
[45] C. Ibars, M. Navarro, and L. Giupponi, “Distributed demand manage-
ment in smart grid with a congestion game,” in Proc. IEEE Int. Conf.
Smart Grid Commun., Oct. 2010, pp. 495–500.
[46] M. Fahrioglu, M. Fern, and F. Alvarado, “Designing cost effective de-
mand management contracts using game theory,” in Proc. IEEE Power
Eng. Soc. Winter Meet., Jan. 1999, pp. 427–432.
[47] M. Erol-Kantarci and H. T. Mouftah, “Wireless sensor networks for cost-
efficient residential energy management in the smart grid,” IEEE Trans.
Smart Grid, vol. 2, no. 2, pp. 314–325, Jun. 2011.
[48] D. Niyato, P. Wang, Z. Han, and E. Hossain, “Impact of packet loss on
power demand estimation and power supply cost in smart grid,” in Proc.
IEEE WCNC, Mar. 2011, pp. 2024–2029.
[49] D. Niyato, D. Qiumin, P. Wang, and E. Hossain, “Optimizations of power
consumption and supply in the smart grid: Analysis of the impact of
data communication reliability,” IEEE Trans. Smart Grid, vol. 4, no. 1,
pp. 21–35, Mar. 2013.
[50] Ontario Micro-FIT program. [Online]. Available: http://microfit.
powerauthority.on.ca/
[51] W. Saad, Z. Han, and H. V. Poor, “Coalitional game theory for co-
operative micro-grid distribution networks,” in Proc. IEEE Int. Conf.
Commun., Kyoto, Japan, May 2011, pp. 1–5.
[52] M. Erol-Kantarci, B. Kantarci, and H. T. Mouftah, “Reliable overlay
topology design for the smart microgrid network,” IEEE Netw.—Special
Issue Commun. Infrastruct. Smart Grid, vol. 25, no. 5, pp. 38–43,
Sep./Oct. 2011.
[53] M. Erol-Kantarci, B. Kantarci, and H. T. Mouftah, “Cost-aware smart
microgrid network design for a sustainable smart grid,” in Proc. IEEE
GLOBECOM Int. Workshop Smart Grid Commun. Netw., Dec. 2011,
pp. 1223–1227.
[54] H.Liang,B.J.Choi,A.Abdrabou,W.Zhuang, andX.Shen, “Decentralized
economic dispatch in microgrids via heterogeneous wireless networks,”
IEEE J. Sel. Areas Commun., vol. 30, no. 6, pp. 1061–1074, Jul. 2012.
[55] A. Nasipuri et al., “Design considerations for a large-scale wireless
sensor network for substation monitoring,” in Proc. IEEE 35th Conf.
LCN, 2010, pp. 866–873.
[56] I. Al-Anbagi, M. Erol-Kantarci, and H. T. Mouftah, “Priority and delay
aware medium access for wireless sensor networks in the smart grid,”
IEEE Syst. J., vol. 8, no. 1, pp. 608–618, Jun. 2014.
[57] I. Al-Anbagi, M. Erol-Kantarci, and H. T. Mouftah, “A reliable IEEE
802.15.4 model for cyber physical power grid monitoring systems,” IEEE
Trans. Emerging Topics Comput., vol. 1, no. 2, pp. 258–272, Dec. 2013.
[58] I. Al-Anbagi, M. Erol-Kantarci, and H. T. Mouftah, “A delay-aware
medium access scheme for WSN-based partial discharge measurement,”
IEEE Trans. Instrum. Meas., vol. 63, no. 12, pp. 3045–3057, Dec. 2014.
[59] K. Hung et al., “On wireless sensors communication for overhead trans-
mission line monitoring in power delivery systems,” in Proc. 1st IEEE
Int. Conf. SmartGridComm, Oct. 2010, pp. 309–314.
[60] Z. Yingchen et al., “Wide-area frequency monitoring network (FNET)
architecture and applications,” IEEE Trans. Smart Grid, vol. 1, no. 2,
pp. 159–167, Sep. 2010.
[61] M. Patel et al., “Real-time application of synchrophasors for improv-
ing reliability,” North Amer. Elect. Rel. Corp., Princeton, NJ, USA,
Oct. 2010. Last accessed on Jan. 2013. [Online]. Available: http://www.
nerc.com/filez/rapirtf.html
[62] A. Ipakchi and F. Albuyeh, “Grid of the future,” IEEE Power Energy
Mag., vol. 7, no. 2, pp. 52, 62, Mar./Apr. 2009.

[63] F. R. Yu, P. Zhang, W. Xiao, and P. Choudhury, “Communication systems
for grid integration of renewable energy resources,” IEEE Netw., vol. 25,
no. 5, pp. 22, 29, Sep./Oct. 2011.
[64] A. Q. Huang, M. L. Crow, G. T. Heydt, J. P. Zheng, and S. J. Dale, “The
future renewable electric energy delivery and management (FREEDM)
system: The energy Internet,” Proc. IEEE, vol. 99, no. 1, pp. 133, 148,
Jan. 2011.
[65] X. Lu, W. Wang, and J. Ma, “An empirical study of communication
infrastructures towards the smart grid: Design, implementation, evalu-
ation,” IEEE Trans. Smart Grid, vol. 4, no. 1, pp. 170, 183, Mar. 2013.
[66] NETL Storage Whitepaper. [Online]. Available: http://www.netl.doe.
gov/smartgrid/referenceshelf/whitepapers/Energy%20Storage_2009_
10_02.pdf
[67] E. M. Lightner and S. E. Widergren, “An orderly transition to a trans-
formed electricity system,” IEEE Trans. Smart Grid, vol. 1, no. 1, pp. 3,
10, Jun. 2010.
[68] J. A. P. Lopes, F. J. Soares, and P. M. R. Almeida, “Integration of
electric vehicles in the electric power system,” Proc. IEEE, vol. 99, no. 1,
pp. 168–183, Jan. 2011.
[69] K. Mets, T. Verschueren, W. Haerick, C. Develder, and F. De Turck,
“Optimizing smart energy control strategies for plug-in hybrid electric
vehicle charging,” in Proc. IEEE/IFIP Netw. Oper. Manage. Symp. Work-
shops, Apr. 2010, pp. 293–299.
[70] K. Clement, E. Haesen, and J. Driesen, “Coordinated charging of mul-
tiple plug-in hybrid electric vehicles in residential distribution grids,” in
Proc. IEEE Power Syst. Conf. Expo., Mar. 2009, pp. 1–7.
[71] S. Sojoudi and S. H. Low, “Optimal charging of plug-in hybrid electric
vehicles in smart grids,” in Proc. IEEE Power Energy Soc. Gen. Meet.,
Jul. 2011, pp. 1–6.
[72] E. Sortomme, M. M. Hindi, S. D. J. MacPherson, and S. S. Venkata,
“Coordinated charging of plug-in hybrid electric vehicles to minimize
distribution system losses,” IEEE Trans. Smart Grid, vol. 2, no. 1,
pp. 198–205, Mar. 2011.
[73] M. Erol-Kantarci, J. H. Sarker, and H. T. Mouftah, “Communication-
based plug-in hybrid electrical vehicle load management in the smart
grid,” in Proc. IEEE Symp. Comput. Commun., Jun. 2011, pp. 404–409.
[74] C. Wu, H. Mohsenian-Rad, and J. Huang, “Vehicle-to-aggregator in-
teraction game,” IEEE Trans. Smart Grid, vol. 3, no. 1, pp. 434–442,
Mar. 2012.
[75] W. C. Su, H. Rahimi-Eichi, W. T. Zeng, and M. Y. Chow, “A survey on
the electrification of transportation in a smart grid environment,” IEEE
Trans. Ind. Informat., vol. 8, no. 1, pp. 1–10, Feb. 2012.
[76] N. Saputro, K. Akkaya, and S. Uludag, “A survey of routing protocols for
smart grid communications,” Comput. Netw. J., vol. 56, no. 11, pp. 2742–
2771, Jul. 2012.
[77] Power Line Surveillance Tool, KEPCO KDN, Seoul, Korea. Last ac-
cessed: Mar. 2013. [Online]. Available: https://www.kdn.com/home3/
we/gs/WEGS_0305.jsp
[78] Qualcomm Smart Energy. [Online]. Available: http://www.qualcomm.
com/solutions/ioe/smart-energy
[79] Echelon and T-Mobile Alliance. Last accessed: Mar. 2013. [On-
line]. Available: https://www.echelon.com/company/news-room/2009/
tmobilealliance.htm
[80] A. Radwan and H. Hassanein, “Does multi-hop communication extend
the battery life of mobile terminals?” in Proc. IEEE Global Telecommun.
Conf., 2006, pp. 1–5.
[81] P. Kolios, V. Friderikos, and K. Papadaki, “Store carry and forward
relay aided cellular networks,” in Proc. 7th ACM Int. Workshop Veh.
Internetw., 2010, pp. 71–72.
[82] M. A. Marsan, L. Chiaraviglio, D. Ciullo, and M. Meo, “Optimal energy
savings in cellular access networks,” in Proc. IEEE ICC Workshops,
2009, pp. 1–5.
[83] J. Gong, S. Zhou, Z. Niu, and P. Yang, “Traffic-aware base station
sleeping in dense cellular networks,” in Proc. IWQoS, 2010, pp. 1–2.
[84] F. Meshkati, H. Poor, S. Schwartz, and R. Balan, “Energy-efficient
resource allocation in wireless networks with quality-of-service con-
straints,” IEEE Trans. Commun., vol. 57, no. 11, pp. 3406–3414,
Nov. 2009.
[85] C. Zhang, T. Zhang, Z. Zeng, L. Cuthbert, and L. Xiao, “Optimal loca-
tions of remote radio units in comp systems for energy efficiency,” in
Proc. IEEE VTC–Fall, 2010, pp. 1–5.
[86] D. Cao, S. Zhou, C. Zhang, and Z. Niu, “Energy saving performance
comparison of coordinated multi-point transmission and wireless relay-
ing,” in Proc. IEEE GLOBECOM, 2010, pp. 1–5.
[87] S. Bu, F. R. Yu, Y. Cai, and X. P. Liu, “When the smart grid meets energy-
efficient communications: Green wireless cellular networks powered by
the smart grid,” IEEE Trans. Wireless Commun., vol. 11, no. 8, pp. 3014–
3024, Aug. 2012.
[88] S. Bu, F. R. Yu, Y. Cai, and X. P. Liu, “Energy efficient cellular net-
works with CoMP communications and smart grid,” in Proc. IEEE ICC,
Jun. 2012, pp. 5921, 5925.
[89] S. Bu, F. R. Yu, Y. Cai, and X. P. Liu, “Dynamic operation of BSs in
green wireless cellular networks powered by the smart grid,” in Proc.
IEEE GLOBECOM, Dec. 3–7, 2012, pp. 5201, 5205.
[90] S. Bu, F. R. Yu, and Y. Qian, “Energy-efficient cognitive heterogeneous
networks power by the smart grid,” in Proc. IEEE INFOCOM, Turin,
Italy, Apr. 14–19, 2013, pp. 980–988.
[91] D. Niyato, X. Lu, and P. Wang, “Adaptive power management for wire-
less base stations in a smart grid environment,” IEEE Wireless Commun.,
vol. 19, no. 6, pp. 44, 51, Dec. 2012.
[92] G. Miao, N. Himayat, Y. Li, and D. Bormann, “Energy efficient de-
sign in wireless OFDMA,” in Proc. IEEE Int. Conf. Commun., 2008,
pp. 3307–3312.
[93] S. Cui, A. Goldsmith, and A. Bahai, “Energy-efficiency of MIMO and
cooperative MIMO techniques in sensor networks,” IEEE J. Sel. Areas
Commun., vol. 22, no. 6, pp. 1089–1098, Aug. 2004.
[94] H. Kim, C.-B. Chae, G. de Veciana, and R. W. Heath, Jr., “A cross-layer
approach to energy efficiency for adaptive MIMO systems exploiting
spare capacity,” IEEE Trans. Wireless Commun., vol. 8, no. 8, pp. 4264–
4275, Aug. 2009.
[95] L. Li et al., “Energy-efficient transmission in cognitive radio networks,”
in Proc. IEEE Consum. Commun. Netw. Conf., Jan. 2010, pp. 1, 5.
[96] Q.-D. Ho and T. Le-Ngoc, “Smart grid communications networks: Wire-
less technologies, protocols, issues and standards,” in Handbook of Green
Information and Communication Systems, M. S. Obaidat, A. Anpalagan,
and I. Woungang, Eds. New York, NY, USA: Elsevier, 2012.
[97] G. Gur and F. Alagoz, “Green wireless communications via cogni-
tive dimension: An overview,” IEEE Netw., vol. 25, no. 2, pp. 50–56,
Mar./Apr. 2011.
[98] M. N. Alam, R. Jantti, J. Kneckt, and J. Nieminen, “Performance anal-
ysis of the IEEE 802.11s PSM,” J. Comput. Netw. Commun., vol. 2012,
pp. 438 654-1–438 654-14, 2012.
[99] L. Li, X. Hu, C. Ke, and K. He, “The applications of WiFi-based wireless
sensor network in Internet of things and smart grid,” in Proc. IEEE
ICIEA, Jun. 2011, pp. 789–793.
[100] V. Chandrasekhar, J. Andrews, and A. Gatherer, “Femtocell networks: A
survey,” IEEE Commun. Mag., vol. 46, no. 9, pp. 59–67, Sep. 2008.
[101] H. Claussen, L. T. W. Ho, and F. Pivit, “Effects of joint macrocell
and residential picocell deployment on the network energy efficiency,”
in Proc. IEEE 19th Int. Symp. Pers., Indoor, Mobile Radio Commun,
Sep. 15–18, 2008, pp. 1, 6.
[102] F. Richter, A. J. Febske, and G. P. Fettweis, “Energy efficiency aspects of
base station deployment strategies in cellular networks,” in Proc. IEEE
VTC–Fall, Anchorage, AK, USA, Sep. 2009, pp. 1–5.
[103] H. Sun, A. Nallanathan, and J. Jiang, “Improving the energy efficiency
of power line communications by spectrum sensing,” in Proc. Int. Conf.
Adv. Comput., Commun. Informat., 2012, pp. 758–762.
[104] H. Sun, A. Nallanathan, N. Zhao, and C. Wang, “Green data transmission
in power line communications,” in Proc. IEEE GLOBECOM, Dec. 3–7,
2012, pp. 3702, 3706.
[105] A. Hamini, J. Baudais, and J. Helard, “Green resource allocation for
powerline communications,” in Proc. IEEE ISPLC Appl., Apr. 3–6,
2011, pp. 393, 398.
[106] HomePlug Green PHY 1.1 Specification, “HomePlug Alliance,”
Portland, OR, USA, 2012. [Online]. Available: http://www.homeplug.
org/tech/whitepapers/HomePlug_Green_PHY_whitepaper_121003.pdf
[107] K. Christensen et al., “IEEE 802.3az: the road to energy efficient ether-
net,” IEEE Commun. Mag., vol. 48, no. 11, pp. 50–56, Nov. 2010.
[108] H. Hrasnica, A. Haidine, and R. Lehnert, Broadband Powerline Commu-
nications: Network Design. Hoboken, NJ, USA: Wiley, 2004.
[109] P. A. Haigh et al., “Development of a visible light communications
system for optical wireless local area networks,” in Proc. Comput.,
Commun. Appl. Conf., Jan. 2012, pp. 351–355.
[110] B. Kantarci and H. T. Mouftah, “Energy efficiency in the extended-reach
fiber-wireless access networks,” IEEE Netw., vol. 26, no. 2, pp. 28–35,
Mar./Apr. 2012.
[111] L. Chiaraviglio, M. Mellia, and F. Neri, “Reducing power consump-
tion in backbone networks,” in Proc. IEEE ICC, Dresden, Germany,
Jun. 2009, pp. 1–6.
[112] F. Idzikowski, S. Orlowski, C. Raack, H. Woesner, and A. Wolisz, “Sav-
ing energy in IP-over-WDM networks by switching off line cards in
low-demand scenarios,” in Proc. Conf. ONDM, Kyoto, Japan, Feb. 2010,
pp. 1–6.
[113] X. Dong, T. El-Gorashi, and J. Elmirghani, “On the energy efficiency
of physical topology design for IP over WDM networks,” J. Lightw.
Technol., vol. 27, no. 12, pp. 1931–1942, Jun. 2012.

[114] G. Shen and R. S. Tucker, “Energy-minimized design for IP over WDM
networks,” IEEE/OSA J. Opt. Commun. Netw., vol. 1, no. 1, pp. 176–186,
Jun. 2009.
[115] B. Kantarci and H. T. Mouftah, “Greening the availability design of
optical WDM networks,” in Proc. IEEE GLOBECOM—Workshop Green
Commun., Dec. 2010, pp. 1417–1421.
[116] B. G. Bathula and J. M. H. Elmirghani, “Green networks: Energy effi-
cient design for optical networks,” in Proc. IFIP Int. Conf. Wireless Opt.
Commun. Netw., Apr. 28–30, 2009, pp. 1–5.
[117] M. Xia et al., “Green provisioning for optical WDM networks,” IEEE
J. Sel. Topics Quantum Electron.—Special Issue Green Photon., vol. 17,
no. 2, pp. 437–445, Mar./Apr. 2011.
[118] N. Naas, B. Kantarci, and H. T. Mouftah, “Energy-efficient realistic
design and planning of optical backbone with multi-granular switching,”
in Proc. ICTON, Jul. 2012, pp. 1–4.
[119] J. Chabarek et al., “Power awareness in network design and routing,” in
Proc. IEEE INFOCOM, 2008, pp. 1130–1138.
[120] X. Dong, T. El-Gorashi, and J. Elmirghani, “IP over WDM networks
employing renewable energy sources,” J. Lightw. Technol., vol. 29, no. 1,
pp. 3–14, Jan. 2011.
[121] B. St. Arnaud, “CANARIE: Research networks to help reduce
global warming,” presented at the OFC/NFOEC Workshop Energy
Footprint ICT—Forecast Network Solutions, San Diego, CA, USA,
Mar., 2009.
[122] C. Cavdar, A. Yayimli, and L. Wosinska, “Cutting the electric bill by
routing and wavelength assignment with time-zones and time-of-use
prices,” in Proc. Commun. Photon. Conf. Exhib., Nov. 2011, pp. 1–3.
[123] J. Baliga, R. Ayre, K. Hinton, and R. Tucker, “Energy consumption in
wired and wireless access networks,” IEEE Commun. Mag., vol. 49,
no. 6, pp. 70–77, Jun. 2011.
[124] R. Kubo, J. Kani, Y. Fujimoto, N. Yoshimoto, and K. Kumozaki, “Sleep
and adaptive link rate control for power saving in 10G-EPON systems?”
in Proc. IEEE GLOBECOM, Dec. 2009, pp. 1–6.
[125] “ITU-T G-Series recommendations—Supplement 45 (G.sup45),”
International Telecommunication Union, Geneva, Switzerland,
May 2009.
[126] W. Shing-Wa et al., “Sleep mode for energy saving PONs: Advantages
and drawbacks,” in Proc. 2nd IEEE GLOBECOM Green Commun.
Workshop, Dec. 2009, pp. 1–6.
[127] M. Maier, “Fiber-wireless sensor networks (FI-WSNs) for smart grids,”
in Proc. 13th IEEE ICTON, Jun. 2011, pp. 1–4.
[128] N. Zaker, B. Kantarci, M. Erol-Kantarci, and H. T. Mouftah, “Quality-of-
service-aware fiber wireless sensor network gateway design for the smart
grid,” in Proc. IEEE ICC—Workshop Opt.-Wireless Integr. Technol. Syst.
Netw., Jun. 2013, pp. 863–867.
[129] N. Zaker, B. Kantarci, M. Erol-Kantarci, and H. T. Mouftah, “Smart grid
monitoring with service differentiation via EPON and wireless sensor
network convergence,” Opt. Switching Netw., 2014, to be published.
[130] L. Shi, S. S. Lee, and B. Mukherjee, “Energy-efficient long-reach
passive optical network: A dynamic wavelength allocation scheme,”
presented at the Photonics Switching, Monterey, CA, USA, Jul. 2010,
Paper PTuB1, to be published.
[131] B. Kantarci and H. T. Mouftah, “Ethernet passive optical network-long-
term evolution deployment for a green access network,” IET Optoelec-
tron., vol. 6, no. 4, pp. 183–191, Aug. 2012.
[132] Z. Fadlullah et al., “Smart FiWi networks: Challenges and solutions
for QoS and green communications,” IEEE Intell. Syst., vol. 28, no. 2,
pp. 86–91, Mar./Apr. 2013.
[133] R. Tucker, “Green optical communications—Part I: Energy limitations
in transport,” IEEE J. Sel. Topics Quantum Electron., vol. 17, no. 2,
pp. 245–260, Mar./Apr. 2011.
[134] R. Tucker, “Green optical communications—Part II: Energy limitations
in networks,” IEEE J. Sel. Topics Quantum Electron., vol. 17, no. 2,
pp. 261–274, Mar./Apr. 2011.
[135] J. Klaus, “Data center energy: Past, present and future (Part One),” Data
Center J., Sep. 2012. [Online]. Available: http://www.datacenterjournal.
com/it/data-center-energy-past-present-and-future-part-one/
[136] N. Xiong, W. Han, and A. VandenBerg, “Green cloud computing
schemes based on networks: A survey,” IET Commun., vol. 6, no. 18,
pp. 3294–3300, Dec. 2012.
[137] California Cap-and-Trade Program, Jan. 2013. [Online]. Available:
http://www.c2es.org/docUploads/calif-cap-trade-01-13.pdf
[138] C. Despins et al., “Leveraging green communications for carbon emis-
sion reductions: Techniques, testbeds, emerging carbon footprint stan-
dards,” IEEE Commun. Mag., vol. 49, no. 8, pp. 101–109, Aug. 2011.
[139] Google whitepaper. [Online]. Available: http://static.googleusercontent.
com/external_content/untrusted_dlcp/www.google.com/en/us/
corporate/datacenter/dc-best-practices-google.pdf
[140] J. Liu et al., “Project Genome: Wireless sensor network for data center
cooling,” Microsoft Architecture J., Dec. 2009, Tech. Rep. [Online].
Available: http://research.microsoft.com/pubs/78813/AJ18_EN.pdf
[141] H. Viswanathan, E. K. Lee, and D. Pompili, “Self-organizing sensing
infrastructure for autonomic management of green datacenters,” IEEE
Netw., vol. 25, no. 4, pp. 34–40, Jul./Aug. 2011.
[142] The Organisation for Economic Co-operation and Development
(OECD), Nomura Research Institute (NRI) Presentation. [Online].
Available: http://www.oecd.org/sti/ind/45009540.pdf
[143] D. Kliazovich, P. Bouvry, and S. U. Khan, “GreenCloud: A packet-level
simulator of energy-aware cloud computing data centers,” J. Supercom-
put., vol. 62, no. 3, pp. 1263–1283, Dec. 2012.
[144] IBM Tivoli Monitoring for Energy Management. [Online]. Avail-
able: http://publib.boulder.ibm.com/infocenter/tivihelp/v15r1/index.jsp?
topic=%2Fcom.ibm.itmem.doc%2Fdco_reports_baseline_power.html
[145] Digital Service Efficiency (DSE). [Online]. Available: http://dse.ebay.
com/sites/default/files/eBay-DSE-130305.pdf
[146] C. Pfeiffer, “Data centers and the smart grid,” Energy Manager Today,
Oct. 2012. [Online]. Available: http://www.energymanagertoday.com/
data-centers-and-the-smart-grid-086428/
[147] J. Pouchet, “Smart grid technology helps data centers conserve en-
ergy,” Environmental and Energy Management News, Dec. 2011.
[Online]. Available: http://www.environmentalleader.com/2011/12/14/
smart-grid-technology-helps-data-centers-conserve-energy/
[148] J. Liu, F. Zhao, X. Liu, and W. He, “Challenges towards elastic power
management in Internet data centers,” in Proc. IEEE Int. Conf. Distrib.
Comput. Syst. Workshops, Jun. 2009, pp. 65–72.
[149] A. Qureshi, R. Weber, H. Balakrishnan, J. Guttag, and B. Maggs,
“Cutting the electric bill for Internet-scale systems,” in Proc. ACM
SIGCOMM, 2009, pp. 124–134.
[150] L. Rao, X. Liu, L. Xie, and W. Liu, “Minimizing electricity cost: Opti-
mization of distributed Internet data centers in a multi-electricity-market
environment,” in Proc. IEEE INFOCOM, Mar. 2010, pp. 1–9.
[151] L. Rao, X. Liu, L. Xie, and W. Liu, “Coordinated energy cost manage-
ment of distributed Internet data centers in smart grid,” IEEE Trans.
Smart Grid, vol. 3, no. 1, pp. 50, 58, Mar. 2012.
[152] L. Rao, X. Liu, M. D. Ilic, and J. Liu, “Distributed coordination of In-
ternet data centers under multiregional electricity markets,” Proc. IEEE,
vol. 100, no. 1, pp. 269–282, Jan. 2012.
[153] J. Li, Z. Li, K. Ren, and X. Liu, “Towards optimal electric demand
management for Internet data centers,” IEEE Trans. Smart Grid, vol. 3,
no. 1, pp. 183–192, Mar. 2012.
[154] L. Rao, X. Liu, L. Xie, and Z. Pang, “Hedging against uncertainty: A tale
of Internet data center operations under smart grid environment,” IEEE
Trans. Smart Grid, vol. 2, no. 3, pp. 555–563, Sep. 2011.
[155] P. Wang, L. Rao, X. Liu, and Y. Qi, “Dynamic power management of
distributed Internet data centers in smart grid environment,” in Proc.
IEEE GLOBECOM, Dec. 2011, pp. 1–5.
[156] H. Wang, J. Huang, X. Lin, and H. Mohsenian-Rad, “Exploring smart
grid and data center interactions for electric power load balancing,” in
Proc. ACM Green Metrics, Pittsburgh, PA, USA, Jun. 2013, pp. 1–6.
[157] R. Wang, N. Kandasamy, C. Nwankpa, and D. R. Kaeli, “Datacenters as
controllable load resources in the electricity market,” in Proc. IEEE 33rd
ICDCS, 2013, pp. 176–185.
[158] B. Kantarci and H. T. Mouftah, “Designing an energy-efficient cloud
network,” IEEE/OSA J. Opt. Commun. Netw., vol. 4, no. 11, pp. B101–
B113, Nov. 2012.
[159] B. Kantarci and H. T. Mouftah, “The impact of Time of Use (ToU)-
awareness in energy and Opex performance of a cloud backbone,” in
Proc. IEEE GLOBECOM, Dec. 2012, pp. 3250–3255.
[160] B. Kantarci and H. T. Mouftah, “Time of Use (ToU)-awareness with
inter-data center workload sharing in the cloud backbone,” in Proc. IEEE
ICC, Jun. 2013, pp. 4207–4211.
[161] X. Dong, T. El-Gorashi, and J. Elmirghani, “Energy-efficient IP over
WDM networks with data centres,” in Proc IEEE 13th ICTON,
Jun. 2011, pp. 1–8.
[162] X. Dong, T. El-Gorashi, and J. Elmirghani, “Green IP over WDM net-
works: Solar and wind renewable sources and data centres,” in Proc.
IEEE GLOBECOM, Dec. 2011, pp. 1–6.
[163] X. Dong, T. El-Gorashi, and J. Elmirghani, “Green IP over WDM net-
works with data centers,” J. Lightw. Technol., vol. 29, no. 12, pp. 1861–
1880, Jun. 2011.
[164] K.-K. Nguyen, M. Cheriet, M. Lemay, M. Savoie, and B. Ho, “Powering
a data center network via renewable energy: A green testbed,” IEEE
Internet Comput., vol. 17, no. 1, pp. 40–49, Jan./Feb. 2013.
[165] M. Ghamkhari and H. Mohsenian-Rad, “Optimal integration of renew-
able energy resources in data centers with behind-the-meter renewable
generator,” in Proc. IEEE ICC, Jun. 2012, pp. 3340–3344.

[166] M. Ghamkhari and H. Mohsenian-Rad, “Energy and performance man-
agement of green data centers: A proﬁt maximization approach,” IEEE
Trans. Smart Grid, vol. 4, no. 2, pp. 1017–1025, Jun. 2013.
[167] M. Erol-Kantarci and H. T. Mouftah, “Overlay energy circle formation
for cloud data centers with renewable energy futures contracts,” in Proc.
IEEE Symp. Comput. Commun., Madeira, Portugal, Jun. 2014.
[168] K. Le, R. Bianchini, T. D. Nguyen, O. Bilgir, and M. Martonosi, “Cap-
ping the brown energy consumption of Internet services at low cost,” in
Proc. IEEE Green Comput. Conf., 2010, pp. 3, 14.
[169] C.Ren,D.Wang,B.Urgaonkar, andA.Sivasubramaniam, “Carbon-aware
energy capacity planning for datacenters,” in Proc. IEEE MASCOTS,
Aug. 2012, pp. 391, 400.
[170] H. Mohsenian-Rad and A. Leon-Garcia, “Coordination of cloud comput-
ing and smart power grids,” in Proc. IEEE Smart Grid Commun. Conf.,
Oct. 2010, pp. 368–372.
[171] E. Oh and B. Krishnamachari, “Energy savings through dynamic base
station switching in cellular wireless access networks,” in Proc. IEEE
GLOBECOM, 2010, pp. 1–5.
Melike Erol-Kantarci (M’08) received the B.Sc.
degree from the Department of Control and Com-
puter Engineering, Istanbul Technical University,
Istanbul, Turkey, in 2001. She received the M.Sc.
and Ph.D. degrees in computer engineering from
Istanbul Technical University in 2004 and 2009,
respectively. During her Ph.D. studies, she was a
Fulbright Visiting Researcher with the Department
of Computer Science, University of California at Los
Angeles. She is the Coordinator of the Smart Grid
Communications Lab and a Postdoctoral Fellow with
the School of Electrical Engineering and Computer Science, University of
Ottawa, Canada. Her main research interests are smart grid, cyber-physical sys-
tems, electriﬁcation of transportation, wireless sensor networks, and underwater
sensor networks. Dr. Erol-Kantarci was the recipient of the Fulbright Ph.D.
Research Scholarship in 2006 and the Siemens Excellence Award in 2004, and
she won two Outstanding/Best Paper Awards. She is an occasional Reviewer of
transactions and journals, and a TPC Member for various conferences. She is
the Vice Chair for Women in Engineering at the IEEE Ottawa Section.
Hussein T. Mouftah (F’90) is a Distinguished
University Professor and Senior Canada Research
Chair in Wireless Sensor Networks with the School
of Electrical Engineering and Computer Science,
University of Ottawa, Ottawa, ON, Canada. Dur-
ing 1979–2002, he was with the ECE Department,
Queen’s University, where he was, prior to his de-
parture, a Full Professor and the Department Asso-
ciate Head. He has six years of industrial experience
mainly at Bell Northern Research of Ottawa (then
known as Nortel Networks). He is the author or
coauthor of 9 books, 60 book chapters and more than 1300 technical papers,
12 patents, and 140 industrial reports. Dr. Mouftah served as Editor-in-Chief of
the IEEE COMMUNICATIONS MAGAZINE in 1995–1997 and IEEE COMSOC
DIRECTOR OF MAGAZINES in 1998–1999, Chair of the Awards Committee in
2002–2003, Director of Education in 2006–2007, and Member of the Board of
Governors in 1997–1999 and 2006–2007. He has been a Distinguished Speaker
of the IEEE COMMUNICATIONS SOCIETY in 2000–2008. He is the Joint
Holder of 19 Best Paper and/or Outstanding Paper Awards. He was the recipient
of numerous prestigious awards, such as the 2007 Royal Society of Canada
Thomas W. Eadie Medal, the 2007–2008 University of Ottawa Award for
Excellence in Research, the 2008 ORION Leadership Award of Merit, the 2006
IEEE Canada McNaughton Gold Medal, the 2006 EIC Julian Smith Medal, the
2004 IEEE ComSoc Edwin Howard Armstrong Achievement Award, the 2004
George S. Glinski Award for Excellence in Research of the U of O Faculty
of Engineering, the 1989 Engineering Medal for Research and Development
of the Association of Professional Engineers of Ontario (PEO), and the Ontario
Distinguished Researcher Award of the Ontario Innovation Trust. He is a Fellow
of the Canadian Academy of Engineering in 2003, the Engineering Institute of
Canada in 2005, and the Royal Society of Canada RSC Academy of Science
in 2008.

Energy efficient information and communication infrastructures in the smart grid a survey on interactions and open issues 1

More Related Content

What's hot (19)

Similar to Energy efficient information and communication infrastructures in the smart grid a survey on interactions and open issues 1 (20)

More from redpel dot com (20)

Recently uploaded (20)

Energy efficient information and communication infrastructures in the smart grid a survey on interactions and open issues 1