Content uploaded by Christos Tsoleridis
Author content
All content in this area was uploaded by Christos Tsoleridis on Dec 06, 2017
Content may be subject to copyright.
347
14 Vehicle-to-Grid
Networks
Issues and Challenges
Christos Tsoleridis, Periklis Chatzimisios,
and Panayotis Fouliras
14.1 INTRODUCTION
From an economist’s perspective, it would be rather difcult to nd the neces-
sary funds in order to build a spinning reserve of energy that could be utilized
as a source of electricity. This is where vehicle to grid (V2G) steps in. As electric
vehicles (EVs) nd their way to massive production, the cost to build such a tank
of spinning energy is compensated by the consumers who also gain by allowing
CONTENTS
14.1 Introduction .................................................................................................. 347
14.2 V2G Load Management Considerations .......................................................348
14.3 V2G Interconnection Specics ..................................................................... 351
14.4 The MAC Protocols ...................................................................................... 354
14.5 Challenges..................................................................................................... 358
14.5.1 Technical Aspect Challenges ............................................................ 358
14.5.1.1 PHY Layer ......................................................................... 358
14.5.1.2 MAC Metrics Relative to VANETs ...................................359
14.5.1.3 MAC Layer ........................................................................ 359
14.5.1.4 Requirements for V2G Communication ............................ 360
14.5.1.5 Security Threats and Authentication Protocol ...................360
14.5.1.6 Routing Protocols............................................................... 362
14.5.1.7 Wireless Charging..............................................................362
14.5.2 Macroscopic Integration-Related Challenges ................................... 362
14.5.2.1 V2G Communications Security and Reliability ................ 362
14.5.2.2 V2G Modeling Objectives .................................................363
14.5.2.3 GEVs Architecture ............................................................. 364
14.5.2.4 Integration Software ..........................................................364
14.5.3 Other Open-Research Issues .............................................................365
14.6 Conclusions ................................................................................................... 366
References .............................................................................................................. 366
© 2016 by Taylor & Francis Group, LLC
348 Smart Grid
the smart grid to utilize the vehicle’s battery when it is parked. This win-win situa-
tion has a handful of benets that should be exploited to the fullest. Therefore, the
marriage of several mature technologies such as smart meters/sensors and wireless
communication schemes, with the evolution of the power grid could be considered
as anticipated by all parties. Currently, governments worldwide attempt to decrease
their carbon emissions by increasing the utilization of renewable energy sources.
Photovoltaic and wind turbine-based power generators are intermittent energy
sources and there could be cases where generation surpasses demand. Storing this
surplus into a spinning reserve can later facilitate the reduction of carbon emissions
from conventional power generators during peak demand periods. Consumers will
also have the opportunity to lease the batteries of their vehicles when not on the
road and collect prots that will compensate—if not depreciate—their investment
in purchasing the EV.
V2G integration requires the establishment of common standards for smart grids.
The charging and discharging process cannot exist without reliable communication
between EVs, charging stations, and the smart grid. EVs are designed to move. As
a result, the grid will have to organize vehicles into groups managed by base sta-
tions, called aggregators, in order to enable effective applications. Consequently, an
aggregator will have to be able to communicate with the vehicles and the control
center and deliver critical information to the smart grid. There are numerous stud-
ies that target toward better medium access control (MAC) protocols to facilitate
vehicle-to-infrastructure (V2I), as well as vehicle-to-vehicle (V2V) communications.
Moreover, in this networked vehicular environment, additional applications can
be implemented, such as safety signaling between the vehicles. The feasibility of
communicating dynamically with neighboring vehicles has inspired approaches for
better and more resilient vehicular ad hoc networks (VANETs) that, among others,
guarantee critical safety information exchange through the novel MAC protocols.
The rest of this chapter is organized as follows. In Section 14.2, we briey inspect
the issue of load management in regard to the application of EV charge and discharge
requirements. In Section 14.3, we discuss the interconnection properties of EVs and
the smart grid in the wireless domain. Section 14.4 provides an enumeration of MAC
protocol considerations that points toward a better supporting layer for the V2G
operations. Finally, Section 14.5 concludes our survey by discussing open-research
issues and future challenges.
14.2 V2G LOAD MANAGEMENT CONSIDERATIONS
Apart from the recent implementations, there is also an interest in the analysis of
how a V2G system should be congured when EV connectivity is taken into account.
EVs are considered to be mobile energy-storage units, also called spinning reserves,
that are distributed and anticipated as the new major factor in energy storage and
manipulation. As discussed in Reference 1, worldwide events (e.g., Olympic Games
hosting, etc.) were also great chances to invest in preliminary V2G test implementa-
tions. The initial objectives of V2G were limited around peak power adjustments,
where the batteries of the vehicles store energy in low-load periods and offer that
power back to the grid when the demand is overwhelming. This also facilitates
© 2016 by Taylor & Francis Group, LLC
Downloaded by [Periklis Chatzimisios] at 05:56 02 May 2016
349Vehicle-to-Grid Networks
efciency in the utilization of distributed renewable energy sources as the intermit-
tency of such power generators can be effectively countered.
The main reason that car owners are expected to offer the battery to the grid is
that it is expected to be protable for the owner. Vehicles that are not tied to the
owner’s profession (e.g., taxi vehicles) are in most cases parked almost the whole
day. An average parked time is close to 23 hours per day (Figure 14.1). As a result,
there is plenty of time in which the vehicle’s battery can be available to the grid as
a storage unit. If the demand is high, the grid can draw power from the vehicles to
cover the extra demand. The ability to use the EVs as mobile storage units to shift
regional load, not only provides social and economic benets, but also seems to
be a better alternative to other ways of energy storage, for example, pump-storage
power stations. The charging and discharging times in EVs are in the order of
milliseconds as no mechanical components are involved whereas the efciency is
up to 80% according to test data—5% higher than the efciency of pump-storage
stations.
The participation of the vehicle in the V2G service that can provide demand peak
shifting would be a win-win schema for both the vehicle owners and the power grid
providers. Vehicle owners will be compensated for allowing the grid to make use of
the vehicle battery. In turn, the power grid will avoid the expenses of building xed
energy-storage facilities, utilize renewable resources more efciently, and improve
the performance of current power plants.
Moreover, there is a scenario where the EVs assist in frequency regulation by
charging according to grid frequency uctuations. This, appropriately managed
charging process is called grid to vehicle (G2V). In this case, the battery is strained
less than in the scenario where there is an actual deep charging and discharg-
ing cycle. The EV varies its charging power according to received signals and its
commitments in order to apply a secondary frequency regulation, known as G2V
regulation [2].
FIGURE 14.1 A parked EV linked to a charging station. Oslo, Norway, 2014.
© 2016 by Taylor & Francis Group, LLC
Downloaded by [Periklis Chatzimisios] at 05:56 02 May 2016
350 Smart Grid
The power reserve in the vehicle can also similarly act to the function of an unin-
terruptable power source (UPS). Especially in homes equipped with charging points,
the EV operates as a voltage source, capable of feeding their loads. This technology
begins to be denominated in the literature as vehicle to home (V2H) [3]. The EV,
while connected to the grid, can be used to temporarily replace the external grid
when there is an outage. In this way, cases like emergency evacuations could be
assisted and the reliability of the power supply could be enhanced as short-term
power outages can be made invisible to the end user.
In Reference 4, it is stated that despite the fact that a plug-in hybrid electric vehi-
cle (PHEV) can be charged from renewable resources, such as photovoltaic or wind
turbine establishments, the intermittency of power generation makes the charging
challenging. In PHEV-charging scenarios, the worst case would be the following:
the occurrences of critical peak periods (CPPs) to coincide with the time of charging
(TOC) of PHEVs. However, simple scheduling could not be effective enough as there
is always the need for communication and immediate signal exchange in order to
counter problems in real time. This is where the communication technologies t well
into the smart grid system and carry out the process of interactive synchronization
between utilities and consumers. Therefore, communications are an integral part for
scalable demand-response equilibrium.
Ιnformation about the status of assets in current power grid utilities is acquired
via the supervisor control and data acquisition (SCADA) system. In Reference 4, the
authors propose their communication-based PHEV load management (Co-PLaM)
scheme to control the load of the PHEVs. The authors assume that the control points
communicate with the utilities including the substation control center (SCC) using
a long-range wireless technology such as wireless interoperability for microwave
access (WiMAX). The SCC and the smart-charging station communicate by form-
ing a wireless mesh network (WMN) using the IEEE 802.11s standard. In this
schema, a simulation of the WMN distribution level was performed and data con-
sidering delivery ratio, delay, and jitter were collected. The mathematical analysis
of the blocking probability of Co-PLaM was provided and the required additional
capacity to supply the PHEVs was presented. The disadvantage of optimization-
based approaches is that load, grid capacity, and charging requests are assumed
to be known. Nevertheless, when communications are available, the decisions are
dynamically determined according to real-time data. This would apply well for the
integration of solar energy collectors and wind turbines where the output of gen-
erators uctuates signicantly during 24 hours of the day. This is why the utility
periodically updates the supplied power thresholds and noties the SCCs through
wireless communications. Since transmission and distribution system conditions
can vary due to unforeseen events, if there is information about the grid state in the
utilities using the SCADA system, it could be in-sync with the charging stations of
the PHEVs. In Co-PLaM, such information is communicated to the SCC that will
rst query for clearance to access the necessary power load given that it is grace-
fully available.
The simulation results for the Co-PLaM scheme showed that the energy-provi-
sioning threshold determines the number of maximum PHEVs accepted for charging
[4]: thresholds of 200 and 150 kWh correspond to 90 and 100 PHEVs charged during
© 2016 by Taylor & Francis Group, LLC
Downloaded by [Periklis Chatzimisios] at 05:56 02 May 2016
351Vehicle-to-Grid Networks
24 hours, respectively. Furthermore, the system could support prioritized charging in
the future for customers who pay more to get their vehicles charged as fast as possible.
Considering the consumer side, if the charging process of the PHEV takes place
at the owner’s home, the charging could be coordinated with other in-home activities
to avoid exceeding a certain level of overall consumption.
The selected avor of IEEE 802.11s uses a hybrid wireless mesh protocol
(HWMP) that combines on demand and proactive-routing algorithms. The MAC
layer is implemented based on the enhanced distributed channel access (EDCA)
standardized in IEEE 802.11e [5].
The peak of power demand for commercially available PHEVs is between 1.8 and
16.8 kW. Charging implementations include xed-demand cases or charging cycles
that draw more energy for the rst period of charging and then lower ones to be able to
charge more when there is little available time for charging. For example, the battery
of a Tesla Roadster can be charged within 4 hours at a peak power level of 16.8 kW.
It should be noted that currently, Tesla Motors is also investigating the possibility of
exchanging batteries rather than charging them in the charging stations. A prototype
changing the battery within 90 seconds has been already demonstrated. However, the
exchange process and how the replacement will be handled are still under testing [6].
14.3 V2G INTERCONNECTION SPECIFICS
The V2G applications can be placed within the map of communication requirements
of the smart grid. In Reference 7, they are classied as neighborhood area network
(NAN) applications that are the middle class between the home area network and
wide-area network application classes (Figure 14.2). Typical functions include the
delivery of pricing information from power utilities to EVs and EVs can provide
information about the status of the battery charge level back to the utilities. Typical
data sizes are expected to be 255 and 100 bytes, respectively, while latency should
be below 15 seconds and reliability over 98%.
According to Reference 8, two types of wireless communications are required for
a V2G system (Figure 14.3):
• The communication scheme between the aggregator and the control center
realized through IEEE 802.16.d and commercialized as WiMAX.
• The communication scheme between the aggregator and the EVs realized
through IEEE 802.11p wireless access for vehicular environment (WAVE).
WAN
100 km
10 km
100 mm
NAN
HAN
FIGURE 14.2 Network area hierarchy ranges in the smart grid.
© 2016 by Taylor & Francis Group, LLC
Downloaded by [Periklis Chatzimisios] at 05:56 02 May 2016
352 Smart Grid
Parking locations, whether at home or at underground parking lots, etc., would
have to provide bidirectional connectivity to the power grid, as well as two-way
communication to the aggregators [9]. The latter unies many vehicles and provides
a single interface for a large group of vehicles. Aggregators are required to commu-
nicate with the smart grid operator, called the control center. Concentrating infor-
mation to a small number of service providers who aggregate EV capacity, sets the
control center workload to feasible levels. Thus, the aggregators handle the customer
interfacing, metering, and billing, and leave higher-level processes for the control
center.
Considering the wireless protocols, the authors of Reference 8 selected the
WiMAX communications standard for communication between the control cen-
ter and the aggregators and the 802.11p for aggregator-to-EVs communication.
If the control center determines a decit in power coverage, a message can be
dispatched to the aggregators that can be forwarded to the EVs. The message can
be delivered to both parked and moving vehicles, so that their owners have the
option to connect to the grid. The security concerns of this information exchange
require source authentication, message integrity, replay attack resistance, and pri-
vacy protection.
The WiMAX was initially designed for cases with line of sight (LOS) within
the 10–66 GHz frequency band. The 802.16a amendment specied working bands
between 2 and 11 GHz that partially enable non-LOS transmissions. The WiMAX
standard denes the air interface that includes MAC and physical (PHY) layers.
Aggregator
Aggregator
Aggregator
WiMAX link
Wave link
Wired link while connected
Control center
FIGURE 14.3 V2G communication scheme.
© 2016 by Taylor & Francis Group, LLC
Downloaded by [Periklis Chatzimisios] at 05:56 02 May 2016
353Vehicle-to-Grid Networks
There are three different PHYs available that provide end-to-end implementation
along with the MAC layer.
• A single-carrier (SC)-modulated air interface.
• A 256-point fast Fourier transform (FFT) orthogonal frequency-division
multiplexing (OFDM), multiplexing scheme.
• A 2048-point FFT OFDM scheme.
The 802.11p WAVE denes modications to IEEE 802.11 for dedicated short-
range communication (DSRC) between the vehicles. There are enhancements that
could be derived from the standard for transportation safety such as collision avoid-
ance and emergency breaking.
In Reference 8, the authors developed two simulation models in MATLAB®
Simulink®, one for each of the communication protocols used. For the aggrega-
tor to EV, the LOS and non-LOS cases were inspected separately due to different
radio propagation characteristics. For the LOS case, the two-ray path loss model
was adopted to determine the received signal power level. Low spectral efciency
modulation schemes, such as binary phase-shift keying (BPSK) and quadrature
phase-shift keying (QPSK), which carry less information (bits) per symbol, require
lower energy per bit and can work in a higher noise oor environment since they are
less vulnerable to bit errors. Simulations show a higher packet error rate for higher-
modulation schemes.
Similarly to the vehicular case, the WiMAX requires high energy per bit over
noise power spectral density, which means that more energy is required for each bit
transfer. The distance between the two peers was set to 1000 m and the conclusion
was that BPSK modulation is the most robust as expected. Increasing the code rate
translates to a higher packet error rate. For the non-LOS path, it is evident that the
performance degrades proportionally to the distance and message signal increases.
In Reference 10, there is a discussion about the V2G integration and an overview
of the current working international joint ISO/IEC standardization of the vehicle-to-
grid communication interface (V2G CI). Efforts are made to take into account the
full potential of EVs and any possible use case to be exploited as it is expected that
EVs will be a commonplace for everyone. For instance, the rst use case dictates
charging of the EVs at home, which seems a trivial process, but if we look closely
at the nature of an EV’s power needs, there is no similar appliance currently within
a typical household. The capacity of a 30-kWh EV is able to power a four-person
household for a few days. Moreover, the EV is expected to recharge overnight.
Billing is also different as electric vehicle supply equipment (EVSE) is shared by
many consumers. Hence, there is no one-to-one relation that ties the consumer of the
power grid with the corresponding consumption.
In this context of information exchange between the EV and the grid, the authors
of Reference 10 provide an overview of the message structure and message patterns
as dened in V2G CI working draft of ISO/IEC 151180-2. Messages are exchanged
over an IPv6 link based on power-line communication (PLC) carrier medium. A sug-
gested encoding layout for the messages is the binary extensible markup language
© 2016 by Taylor & Francis Group, LLC
Downloaded by [Periklis Chatzimisios] at 05:56 02 May 2016
354 Smart Grid
(XML) that provides the loose restrictions of XML along with a binary serialization
to avoid an unnecessary overhead. Recent evaluations imply that the usage of W3C’s
efcient XML interchange (EXI) [11] ts the realization of the V2G-based interac-
tion signaling. The types of messages between EV and EVSE can be differentiated
as plugging-in, service discovery, authorization, power discovery, power request,
payment, charging cycles, and unplugging.
In Reference 12, the authors point out that EVs are a signicant capital invest-
ment that can facilitate in renewable and, in most cases, intermittent energy sources
through closely attended integration. It is also discussed that the IEEE 802.15.4
(Zigbee) protocol, by designing a low-power (<1 mW) connectivity implementa-
tion, can t well for metering and signaling communications for plug-in EVs (PEVs).
Communication-driven management of EV charging/discharging behavior is a pre-
requisite to scaled EV adoption, since the unattended and opportunistic charging of
EVs adds up to the inefcient overall load of power consumption even during peak
hours. Consequently, in order to meet the power requirements of EV transporta-
tion as a mainstream means of transportation, the load has to be shifted to off-peak
hours or additional power has to be generated. By simulating the interaction between
PHEVs and the power grid, the authors of Reference 12 conclude that utilities may
be able to reduce the extra capacity needed to serve PHEVs by implementing a low-
throughput communication system.
14.4 THE MAC PROTOCOLS
Mobile ad hoc networks (MANETs) where nodes self-congure themselves and
interact without using xed infrastructures or centralized administration are dis-
cussed in Reference 13. Such network topologies do not allow more than one trans-
mitting terminal at a given time for each channel. In order to effectively share the
medium, different existing MAC protocols suitable for VANETs were tested.
In the MANET domain, one of the rst MAC protocols to counter the shared-
medium problems was ALOHA with a random access-oriented approach and
S-ALOHA. The carrier sense multiple access (CSMA) protocol was also exam-
ined, concluding that the main weaknesses are the hidden- and exposed-terminals
issues. The hidden-terminal problem occurs when a terminal starts transmitting
while failing to detect another terminal that also transmits because it is out of
range. The exposed-terminal problem occurs when a transmission is falsely blo-
cked, because the transmitter senses a neighbor-transmitting node that will actu-
ally not interfere with the transmission. Multiple access with collision avoidance
(MACA) introduced the request-to-send (RTS) and clear-to-send (CTS) mecha-
nisms to counter the hidden-terminal problem by agreeing with the receiver on
the transmission.
Nevertheless, there are cases where the exposed-terminal problem does occur.
MACA wireless did counter the exposed-terminal issue by adding data-sending and
acknowledgment packets with regard to RTS and CTS packets. The busy tone mul-
tiple access (BTMA) MAC protocol proposed a new way to counter the hidden-
terminal problem by splitting the channel transmission into two channels: a data and
control channel (CCH). The latter is used to transmit the busy tone. When a node
© 2016 by Taylor & Francis Group, LLC
Downloaded by [Periklis Chatzimisios] at 05:56 02 May 2016
355Vehicle-to-Grid Networks
receives the busy tone, it retransmits the signal in order to notify its neighbors who
might be out of the transmission radius of the original signal [13].
Other ways to split the medium include division in terms of time. Time division
multiple access (TDMA)-based methods employ xed time frames where each frame
is further divided into several slots. The ve-phase reservation protocol (FPRP) was
the rst-proposed TDMA protocol in which the medium is divided into information
frames (IFs) to send data and reservation frames (RFs) for reservations.
In frequency division multiple access (FDMA), the medium is slotted in terms
of frequencies in order for multiple stations to transmit concurrently. Other MAC
proposals can be applied in each frequency channel such as memorized carrier sense
multiple access (MCSMA) where CSMA is used in each channel.
In code division multiple access (CDMA)-based protocols, several orthogonal
codes are available and each node uses a code to encrypt messages before transmis-
sion. For example, in multicode MAC (MC MAC), several codes are used with one
of the codes reserved for control packet transmissions.
VANETs are destined to adapt MANET-qualied protocols into use cases where
peers are vehicles that try to transmit and receive from other vehicles or infrastruc-
tures. Different approaches are considered to achieve reconciliation between perfor-
mance and reliability in VANETs (Figure 14.4):
1. In the WAVEs protocol that is referred as well as IEEE 802.11p, the PHY
and MAC layers are tuned for VANETs. By using OFDM, V2V, and V2I,
connections are possible over distances up to 1000 m. High speed between
Contention free
Contention based
CSMA/CA
ADHOC MAC
Directional
Wave
DMMAC,
modified R-ALOHA and
ACFM
VANETs
Structured
TDMA
FDMA
CDMA
SDMA
FIGURE 14.4 Overview of the discussed MAC properties.
© 2016 by Taylor & Francis Group, LLC
Downloaded by [Periklis Chatzimisios] at 05:56 02 May 2016
356 Smart Grid
peers is taken into account in which fast multipath-fading scenarios are
countered with the OFDM technology.
2. ADHOC MAC is an MAC protocol of the European project CarTALK2000
(FleetNET has been the follow-up) as a means of solving VANET com-
munication issues. The selected structure is a slotted MAC frame, indepen-
dent from the PHY layer, through the use of a dynamic TDMA mechanism
that could be adapted to the universal mobile telecommunications system
(UMTS) terrestrial radio access time division duplex (ULTRA-TDD). The
reliable R-Aloha (RR-ALOHA) protocol used in ADHOC MAC, employs
the dynamic TDMA mechanism by having each car select a basic channel
(BCH), which, in turn, is a time slot periodically repeated in consecutive
frames. In the implementation, each peer sends its frame information on
the BCH, containing a vector that indicates statuses sensed in the previous
frame.
3. Directional antenna transmission is also a way of bypassing MAC issues
in VANETs. The vehicles move within the allowed routes of the motorway
network and, therefore, directional antennas could certainly help in reduc-
ing collisions in cases of parallel neighboring vehicular trafc. Having mul-
tiple antennas allows the node to block the antenna that receives an RTS
transmission.
The authors of Reference 13 note as the main weakness of the 802.11 MAC, the
drawbacks in throughput caused by the CSMA/CA mechanism that cannot guaran-
tee a deterministic upper bound on the channel access delay. On the other hand, the
ADHOC MAC does not use the medium efciently and the number of vehicles in the
same broadcast domain cannot be greater than the number of slots in the time frame.
Finally, the directional-antenna-based MAC does improve network throughput by
ghting collisions but at a cost: several antennas are required in practice, making the
solution more expensive than single-antenna implementations.
In Reference 14, the authors propose a dedicated multichannel MAC protocol,
called DMMAC, which uses an adaptive broadcasting mechanism in order to pro-
vide collision-free and delay-bounded transmissions for safety applications under
various trafc conditions. In this approach for VANET environments, a hybrid chan-
nel access mechanism is exploited in order to deliver both the advantages of TDMA
and CSMA/CA. All vehicles are equipped with a single half-duplex radio to avoid
cross-channel interference from multiple radios for each node.
The MAC protocol for VANETs is required to be reliable and efcient as all MAC
protocols, but with the specialty of the highly dynamic network topology of moving
vehicles with regard to different kinds of quality of service (QoS). In 1999, the U.S.
Federal Communications Commission (FCC) allocated seven 10-MHz channels in
the 5.9-GHz band, including the six service channels (SCHs) and one CCH. This
layout, along with the nature of the vehicular network environment, which cannot
follow typical channel reuse techniques, has led to studies for multichannel MAC
protocols for higher throughput and network latency. Dividing the band according to
the regional data would not make any sense since the key factor in this case is prox-
imity among vehicles. The authors of Reference 13 also argue about the insufcient
© 2016 by Taylor & Francis Group, LLC
Downloaded by [Periklis Chatzimisios] at 05:56 02 May 2016
357Vehicle-to-Grid Networks
scheme of the current WAVE MAC with the contention-based channel access imple-
mentation, which cannot guarantee the QoS of safety, and other real-time applica-
tions in high-density scenarios.
In the DMMAC architecture, the channel coordination is similar to WAVE
MAC. Access time is equally divided into sync intervals with each one consist-
ing of a CCH interval (CCHI) and an SCH interval (SCHI) of the same length. In
DMMAC, there is further division of the CCHI into an adaptive broadcast frame
(ABF) and a contention-based reservation period (CRP). ABFs are, by design, suit-
able to deliver safety messages as they are set to inform the sender that they were
delivered, while also informing about the outcome of a transmission. The number
of time slots within each ABF is called the ABF length (ABFL). This length is not
statically specied for the whole network. However, there is a set of maximum and
minimum values predened in the system and each vehicle can adjust its ABFL in
every CCHI, accordingly. These adjustments, along with other details of adaptive
broadcasting, are presented in the adaptive broadcasting implementation protocol
(ABIP), which is a set of rules for regulating the access behaviors of the vehicles
in the ABF, how to reserve a slot as BCH, adapt the ABFL, and determine whether
to add virtual slots after the end of the ABF. The ABIP can provide every vehicle a
contention-free opportunity to transmit. However, vehicles still need to contend in
order to reserve a slot when there are many vehicles that want to reserve the BCH
si multa n e o u sly.
By using CSMA/CA, the CRP provides a means for vehicles to make reserva-
tions for non-safety-related applications. The CRP also depends on the ABFL of
the vehicle. In order to prevent potential collisions, some vehicles have additional
slots named virtual slots. Generally, a pair of vehicles need to exchange three types
of packets: CRP-REQ, CRP-RES, and CRP-ACK. SCHI also divides the channel
access time into equal-duration slots and all slots on the same channel are grouped
into one nonsafety application frame (NSAF). All NSAFs in one SCHI can be
reserved during the CRP for collision-free transmissions of non-safety-related data.
The comparison of results of DMMAC with WAVE MAC in terms of safety
packet delivery performance showed that DMMAC decreases slightly, whereas
WAVE MAC grows steadily worse as the competition between nodes to occupy the
medium increases.
As discussed in Reference 15, the specic area of VANET still faces signicant
challenges in the design of reliable and robust MAC protocols for V2V communica-
tions. VANETs are designed to provide coverage within 1000-m radius with roadside
units (RSUs) and other vehicles, while traveling at relative speeds up to 200 km/h,
regardless of the surrounding environment. Apart from the information considering
V2G integration, there is safety-related information that will be incorporated into
predened basic signaling schemes, such as lane change assistance, cooperative for-
ward incident warning, intersection collision avoidance, and emergency or incident
warning.
It is evident that an MAC protocol designed for infotainment has to take differ-
ent things into account compared to an MAC protocol designed for safety signaling.
However, both these cases are a requirement in VANETs since one complements the
other in order to be presented as a product with successful embodiment to the car
© 2016 by Taylor & Francis Group, LLC
Downloaded by [Periklis Chatzimisios] at 05:56 02 May 2016
358 Smart Grid
industry, be it electric or conventional. Safety messages are short and are required to
be delivered as soon and as reliably as possible, while infotainment services overtake
wider data load with less regard to low-latency requirements. Furthermore, these two
services are also opposite in the sense that one tries to increase the driver’s awareness
for potential threats, whereas the other increases potential sources of distraction.
Within a broadcast domain, the peer-granted access to occupy the PHY medium
is determined by the MAC layer. A categorization of the MAC mechanisms in
terms of access approach could be as either contention based or contention free.
The former is based on carrier sensing and backing off until the next attempt to
transmit; the latter divides access into time slots and uses synchronization schemes.
It is, however, possible to have a mixture of the two methodologies in the same
implementation.
A basic distinction between MAC mechanisms would also be the point of control
of the medium access. Medium access can vary in terms of methodology. It can
be completely random and the nodes would try to access the medium with little or
no coordination. On the other hand, there are more structured approaches where
there are certain time slots, or certain frequency channels allocated according to
prearranged layouts. More specically, for the structured approaches, there are four
fundamental techniques that can be tweaked or combined in various ways. These
are the TDMA, the FDMA, the CDMA, and the space division multiple access
(SDMA).
Contention-based methods tend to better utilize the medium and consume less
energy with less coordination required. There is also more resilience to network
changes. On the other hand, in scenarios with high trafc load and many peers con-
testing for a chance to transmit, the performance of contention-based implemen-
tations deteriorates signicantly due to increased collisions. Contention-free MAC
methods can restrict access delays to certain bounds, QoS can be guaranteed, and
the overall performance is better under increased network trafc load. Such methods
are considered more reliable and are expected to utilize the channel better. There
is, however, more coordination needed—especially in cases where the network is
rapidly changing and portions need to be reallocated frequently.
14.5 CHALLENGES
14.5.1 technIcal asPect challenGes
In this section, we present a selection of technical challenges, outlined in Table 14.1.
14.5.1.1 PHY L ayer
Challenges that need to be taken into account in the PHY layer include the Doppler
effect, multipath fading, adjacent channel interference, and interference from the
other RF sources. In addition, the mobility between nodes in V2V or V2I makes
things even more challenging as the surroundings constantly change. Hence,
assumptions for the effective guard interval length in OFDM transmissions are more
complicated. Link PHY properties vary continuously.
© 2016 by Taylor & Francis Group, LLC
Downloaded by [Periklis Chatzimisios] at 05:56 02 May 2016
359Vehicle-to-Grid Networks
14.5.1.2 MAC Metrics Relative to VANETs
Since VANETs are distinguished from other ad hoc networks due to high node
mobility [15], suitable metrics of evaluation for the MAC layer would be the maxi-
mum medium access delay, payload delivery delay, throughput, overhead, access
fairness, probability of successful delivery, and network stabilization time. The lat-
ter is very important for VANETs. One of the main objectives is a cost-effective and
scalable technology that minimizes the time of establishing connections and access
delays to the underlying wireless medium for V2V or V2G scenarios. There is no
resemblance to the cellular Internet connectivity that is inherently infrastructure
based.
14.5.1.3 MAC Layer
At the MAC layer, the challenges for VANETs include the hidden terminal, the
dynamic nature of VANETs, the scalability requirements, and the great diver-
gence in the requirements of applications designed for the vehicular environment.
Single-radio implementations are unable to transmit and receive simultaneously,
leading to indirect collision detection. Nodes in a VANET are inherently mobile.
Therefore, the MAC layer should be optimized for continuous disconnects and
roaming between RSUs and on-board units (OBUs) of other vehicles. Moreover,
in V2I schemes, the RSU can act as a coordinator for centralized MAC methods,
whereas in pure V2V schemes, there is no such option for access management and
coordination among allocated channels. Similarly, the QoS requirements are dif-
cult to meet, especially for safety messages, where the objective is to guarantee
message delivery within the time frame that the information will be valid and
useful.
TAB LE 14.1
Technical Challenges
Technical Aspect Challenges
A. PHY layer: Doppler effect, multipath fading, adjacent interference, interference from other
sources, and mobility between peers
B. MAC metrics relative to VANETs with mobile peers: Access delay, payload delivery delay,
throughput, overhead, access fairness, probability of successful delivery, and network
stabilization time
C. MAC layer: Hidden terminal, dynamic nature of VANETs, and QoS issues such as time-of-
delivery restrictions for safety signaling
D. Specic requirements for V2G communication: Latency, bandwidth, and effective radius.
Required information from EV to the aggregator and information available to the EV
E. Security threats and requirements to be met in a smart EV-charging service. Authentication
protocol: Safe integration with the current power grid information systems
F. Routing protocol challenges related to key factors of VANETs
G. Wireless charging: On-the-y charging, wireless charging efciency, effective distance ne-tuning,
and ease of use
© 2016 by Taylor & Francis Group, LLC
Downloaded by [Periklis Chatzimisios] at 05:56 02 May 2016
360 Smart Grid
14.5.1.4 Requirements for V2G Communication
As stated in Reference 16, the entities that are part of the V2G architecture require
very specic communication platforms in terms of latency, bandwidth, and effective
radius. Furthermore, the security of the information to be exchanged is also vital as
attacks can produce severe problems to power distribution. According to Reference
16, the typical information that the power aggregator needs from EVs includes the
ID of the EV, battery voltage, battery chemistry type, temperature, charging prole
(how much available), driving habit, etc. On the other hand, the OBU would be able
to obtain information about secure user identication, current grid frequency, charg-
ing station(s) location (GPS), metering data for actual power ow (demand/supply),
corresponding billing rates, etc.
14.5.1.5 Security Threats and Authentication Protocol
The opportunities of attackers targeting smart EVs are enumerated in References 17
and 18. Possible threats include impersonation, tampering with communication mes-
sages, eavesdropping, denial of service (DoS), privacy breaches, and disputes. The
authors propose actions to counter these threats by establishing
• Stronger entity authentication
• Enhanced message authenticity checks
• Centralized and role-based access control authorization
• Symmetric or/and public key encryption for condentiality assurance
• Nonrepudiation to increase the level of trust
• Measures to ensure maximum possible availability for key services in order
to withstand DoS attacks
• Anonymity and nonlinkability (privacy preservation) by incorporating a
trusted third party
A different distinction of the security concerns of an EV is enumerated in
Reference 19, after suggesting that such a vehicle can—from now on—be considered
as a fully connected network device:
1. Data: The information exchanged between the vehicle and the grid
needs to be protected from packet snifng and resilient to replay attacks.
Furthermore, the already-stored data must be immune to attacks, such as
structured query language (SQL) injection, etc.
2. Communication network: In the context of using the ZigBee protocol
for connections between the EV and the utility, all weaknesses already
addressed need to be amended before deployment. This applies to any wire-
less protocol potentially involved.
3. Infrastructure: Since EVs will be utilized as energy-storage assets for dis-
tributed energy resources (DERs), every device that acts as a mediator
along with the EV itself will have to be free of malicious software, viruses,
and vulnerable or exploitable network services. By verifying the sanity of
every component of the involved infrastructure, a considerable part of the
possible threats for the grid can stay under control.
© 2016 by Taylor & Francis Group, LLC
Downloaded by [Periklis Chatzimisios] at 05:56 02 May 2016
361Vehicle-to-Grid Networks
4. Firmware and software: A contemporary vehicle already has several elec-
tronic control units (ECUs) controlling the functionality of various in-
vehicle systems. This rmware must be updated assuring that the received
update comes from a trusted and authorized source. The maintenance and
repair technicians or even the users themselves must also be restricted from
tampering with these processes.
As a result, the authentication protocol is obliged to meet special challenges
related to EVs, such as large overhead and latency that are crucial for secure wireless
communications between fast-moving nodes. In this context, Chia etal. [20] focused
their research and deployment to cyber security, EV charging and telematics for EVs,
and the smart grid in a dense urban city environment such as Singapore. The deliv-
erables of the developed smart grid cyber security architecture for the EV-charging
infrastructure were assurance of the correct information for EV-charging coordi-
nation, secure payment and transaction integrity, safe integration with the power
grid information system in the midst of possible new attacks, and minimization of
exposure to potential risks for intelligent electronic devices within the smart grid.
Furthermore, in Reference 21, the authors identify unique security challenges in
an EV’s different battery states. Privacy preservation aims at decoupling identities
from their sensitive information. Their proposed battery status-aware authentication
scheme hides each EV’s identity from disclosing location-related information and
introduces challenge–response to achieve dynamic response without revealing the
user’s related privacy. Another privacy-preserving communication protocol for V2G
networks is proposed by Tseng in Reference 22. In this attempt, a restrictive partially
blind signature is utilized to protect the identities of the EV owners. Blind signatures
involve signing without revealing the content of the message to the signer. It is also
noted that the proposal is designed in a way to simplify the certicate management
infrastructure that as noted in Reference 23, can reach a considerable amount of
workload required in a smart grid.
Khurana etal. [23] note that the smart grid is poised to transform a centralized
and producer-controlled network to a decentralized and consumer-interactive net-
work. This dictates very specic requirements in terms of trust; for example, each
user is accessing accurate data created by the right device, at the expected location
and proper time, by an expected protocol, and that the data were not tampered with.
Another interesting conclusion was that the requirements for effective cyber secu-
rity solutions contain the parameter that power availability is more important to
most users than power ows information condentiality. Moreover, the transmission
substations authentication and encryption requirements involve cases with multicast
messages that must be delivered in less than 4 milliseconds. This implies that ef-
cient authentication algorithms will have to minimize the computational cost and
that packet buffering should be avoided so that presented requests are processed
immediately.
Toward this objective, Guo etal. [24] proposed a batch authentication protocol
called UBAPV2G that tried to deliver reduction of authentication delay, less compu-
tational cost, and less communication trafc versus the standard one-by-one authen-
tication scheme. However, in Reference 25, it was shown that this approach created
© 2016 by Taylor & Francis Group, LLC
Downloaded by [Periklis Chatzimisios] at 05:56 02 May 2016
362 Smart Grid
vulnerable use cases in which either the vehicle or the aggregator can generate a col-
lection of bogus signatures that satisfy the batch verication criterion, that is, forgery
attacks. Furthermore, in Reference 26, the authors propose a multidomain network
architecture for V2G infrastructure that includes hybrid public key infrastructure
(PKI), using hierarchical and peer-to-peer implicit cross-certications. Their simu-
lation results showed signicant reduction in the validation duration when compared
to the hybrid PKI scheme using explicit certicates.
14.5.1.6 Routing Protocols
Despite the fact that the routing requirements in VANETs are well dened and
exploited by the research community, there are several challenges to be ventured,
especially if low-cost and low-power consumption networks are to be used [27].
These challenges are related to key factors of VANETs’ mobility already addressed
in this chapter and enumerated in Reference 28. In References 29 and 30, there are
detailed classications and discussions of the current routing protocols for VANETs
whereas in Reference 31, future research directions for routing protocols in a smart
grid, in general, are proposed. The addressed topics include QoS architecture, secure
routing, secure and QoS-aware routing, hybrid routing using PLC and wireless com-
munication, cross-layer routing via multichannels and multiple-input–multiple-out-
put (MIMO) antennas, scalable routing, simulation tools and test beds for routing,
standardization and interoperability in routing, and multicast routing.
14.5.1.7 Wireless Charging
Chia etal. [20] also addressed the challenge of a successful wireless EV-charging
scheme. They concluded that ideas such as on-the-y charging while the vehicle is
traveling along charging lanes or while waiting at trafc lights can become a reality
through wireless charging. A solution to achieve high efciency (>90%) of wireless
power transfer over distances of several centimeters to meters makes use of a phe-
nomenon called magnetic resonance coupling. This phenomenon is a special case of
inductive coupling, taking place when the transmitting and receiving coils, together
with their matching circuits, are made to resonate at a specic power transmission
frequency and at a specic distance. The challenge involves successful integration in
the actual charging point because small deviations in the distance between the coils
results in severe deterioration in efciency. Current implementations try to ne-tune
to the optimum frequency after the vehicle is parked. Otherwise, the driver of the
vehicle would be required to place it in a very specic position that is a difcult task.
This process would naturally degrade the user’s convenience.
14.5.2 macRoscoPIc InteGRatIon-Related challenGes
In the second part of this section, there is an overview of macroscopic and integra-
tion-related challenges as shown in Table 14.2.
14.5.2.1 V2G Communications Security and Reliability
In every EV-charging planning context, even in battery swapping, the efciency of the
communication schema that delivers critical information about energy availability or
© 2016 by Taylor & Francis Group, LLC
Downloaded by [Periklis Chatzimisios] at 05:56 02 May 2016
363Vehicle-to-Grid Networks
battery throughout, is a fundamental factor that will greatly improve or worsen user
experience. Challenges to a V2G transition include battery technology evolution and
the high initial costs compared to conventional vehicles. Limitations to using the
PEV for V2G will likely be related to implementing assured and secure communica-
tions, particularly between an aggregator and a large number of PEVs.
Security issues are important in the communication network at home as well as
while visiting public-charging facilities [32]. An additional issue is that the distribu-
tion grid has not been designed for bidirectional energy ow; this tends to limit the
service capabilities of V2G devices. Conversely, the implementation of fully bidi-
rectional communications in the V2G infrastructure (and the smart grid, in gen-
eral) introduces new possible vulnerabilities. As discussed in Reference 33, attacks
such as DoS and price manipulation can prevent the owner of an EV to determine a
real electricity price that can result in suboptimal decisions on charging/discharging
planning. In their work, a policy of charging is proposed with respect to resiliency
under price information attacks.
Battery degradation issues [34,35] as well as investment costs and energy losses
[36] are also important research areas. According to Reference 37, if PEVs are to
become the preferred vehicles within the United Kingdom, a signicant investment
in electrical networks will be required. Moreover, each V2G entity can have multiple
roles within the system according to the current function performed: energy demand,
energy storage, or energy supply. This further complicates the security consider-
ations to be met as shown in Reference 38 where the authors also propose a role-
dependent scheme to preserve each entity’s privacy.
14.5.2.2 V2G Modeling Objectives
In the literature, the issue of successfully proling EV energy that needs incorpora-
tion into the contemporary household is well addressed. When EVs are connected to
the power grid for charging and/or discharging, they become griddable EVs (GEVs)
[39]. GEVs are considered to be primarily connected to the home (V2H) and then
considered for V2V and V2G. This indicates that the big picture includes all these
models. The modeling of V2H, V2V, and V2G systems should be based on the objec-
tives and their constraints. General objectives are load variance minimization, cost
minimization, cost-efciency optimization, cost-emission minimization, power loss
TAB LE 14.2
Macroscopic Challenges
Macroscopic Integration-Related Challenges
A. Importance of secure communications in V2G even in battery-swapping cases, battery degradation,
and investment costs
B. Modeling objectives of EVs and the household: Load variance, cost-efciency optimization, and
cost-emission minimization
C. GEVs: Grouping architecture and optimal sizing of GEVs
D. Integration software to couple EVs and other active V2G entities along with DERs and the rest of
the smart grid
© 2016 by Taylor & Francis Group, LLC
Downloaded by [Periklis Chatzimisios] at 05:56 02 May 2016
364 Smart Grid
minimization, load shift and peak load reduction, and reactive power compensation.
The demand-response management problem is dened in a scaled view of those
issues. Case studies and research projects indicate that the embodiment of innova-
tive technologies is required. Such enabling technologies include smart meters with
advanced-metering infrastructure (AMI), home energy controllers, energy manage-
ment systems (EMSs), and wired and wireless communication systems [40].
Moreover, in Reference 41, there is work toward the adjusting of the load uncer-
tainty in the presence of PEVs. Given that V2G technology is potentially a new
renewable energy resource, it can be utilized in order to decrease operational cost.
14.5.2.3 GEVs Architecture
Apart from PHY connections with the power grid, the GEV has other interactions
with the grid for V2H, V2V, and V2G operations: information and communication.
The V2G operation requires a reliable and secure two-way communication net-
work, enabling message exchanges between the GEVs and the power grid. There
are numerous suggestions for V2G communication networks such as References 24,
39, and 42. The diversity and exibility of V2G communication networks also pose
challenges to the architecture. A direct V2G control system is the simplest archi-
tecture, where the GEVs are directly supervised by the grid operator; but the large
number of GEVs penetrated in the grid increases the computation load of the grid
operator tremendously; this led to the adoption of indirect V2G architectures. Here,
as already stated, the third entity (aggregator) is involved in reducing the workload of
the grid operator. Consequently, the issue of optimal GEV aggregation sizing arises,
in which the parameters to be determined involve communication platform limita-
tions, as well as coordination computation load limitations.
14.5.2.4 Integration Software
A nonnegligible aspect of V2G challenges includes the high-level mechanics that
will enable its full potential. Toward that goal, the VOLTTRON platform [43]
provides an agent execution environment to fulll the strict requirements of V2G
applications such as coordinating EV charging with home energy usage. Another
interesting approach is considered in Reference 44, where the authors show that a
virtual power plant (VPP) that integrates V2G-enabled EVs has many similarities
with instant messaging (IM) and voice over IP (VoIP) in terms of communication
patterns. The move to propose the use of session initiation protocol (SIP), a well-
established standard, in order to transmit status, trip information, and charging
process control signals between EVs and the VPP. Finally, there is a discussion
of a web services-oriented approach [45] as a means to interconnect every V2G-
integrated device. Devices prole for web services (DPWSs) provides a generic
middleware and prole for embedded devices based on web service technologies.
It is closely related to universal plug and play (UPnP) [46]. Both offer nearly
the same functionality to the application layer: addressing, discovery, descrip-
tion, control, eventing, and presentation of devices and their encapsulated ser-
vices. The major advantage of DPWS over UPnP is its strict adoption of standard
WS-* specications. This makes DPWS very attractive in industrial automation
because the complexity and costs for integrating device-level processes into the
© 2016 by Taylor & Francis Group, LLC
Downloaded by [Periklis Chatzimisios] at 05:56 02 May 2016
365Vehicle-to-Grid Networks
existing information technology (IT) are minimized. IEC 61850 denes a set of
abstract objects and services that allows the description of functions and applica-
tions independently of a particular protocol or PHY device. The following list
summarizes the overall key requirements for application-layer protocol mappings
to IEC 61850 [45]:
• Standards based
• Support for utility enterprise IT and networking
• Multivendor interoperability
• Support for autoconguration
• Support for self-description
• Support for security
• Support for le transfers
14.5.3 otheR oPen-ReseaRch Issues
In the midst of a struggle to be adopted and implemented, the V2G research direc-
tion has many works that try to introduce, evaluate, conrm, and encourage its estab-
lishment. As suggested in Reference 47, the discharge of EVs affects the power grid
in four different aspects: economy, battery life, providing ancillary services, and
compensating intermittency of renewable energy generation. Furthermore, it is noted
that charging and discharging management strategies in different case studies repre-
sent a signicant point of future research directions.
In a similar context, Reference 48 enriches the research toward the optimal opera-
tion of charging stations considering the real-time electricity prices and V2G capac-
ity. Their simulations show considerable economic and reliability benets that need
further investigation. On the same issue, the authors of Reference 49 conclude that
PHEV penetration will have a great impact on the residential electricity distribu-
tion network and, as a result, the management of PHEV charge/discharge schedule
is a key issue in the research of PHEVs. On the other hand, in Reference 50, the
authors highlight the importance of the inefciencies of V2G connections and sug-
gest research directions.
Finally, on the front of software agent programming for PHEVs, the authors of
Reference 51 describe their ndings after simulating as well as implementing in real
life an agent who considers individual driving behavior and battery-discharging costs.
In a greater scale, development of the design, integration, simulation, and operation
of a whole-system V2G model are provided in Reference 52. The authors explore
four key areas of research: power system integration, V2G communications, system
management, and power network simulation. Their V2G model aims toward the pro-
vision of a test bed capable of challenging the full range of technological difculties
that have yet to be overcome in the eld of V2G technology.
Given that V2G technology has yet to receive a mass adoption, any research that
adds value or offers positive insight toward that goal could signicantly enable it.
As a result, works toward better charging and discharging management strategies,
optimal operation of charging stations, smarter V2G models, and more thorough
simulation tools are important future research areas.
© 2016 by Taylor & Francis Group, LLC
Downloaded by [Periklis Chatzimisios] at 05:56 02 May 2016
366 Smart Grid
14.6 CONCLUSIONS
In this chapter, we presented a perspective of the requirements that need to be
covered in the wireless communication scheme that can facilitate V2G integration
completion. The survey focused on current work on utilizing an EV to the fullest,
while keeping it interconnected to the power grid and other vehicles. An attempt was
made to assess which out-of-the-box wireless technologies are compatible with V2G
and what challenges and opportunities arise in newly introduced use cases. In this
context, the challenges were divided into two separate groups, the technical and the
macroscopic ones. Both groups pay special attention to the security issues that repre-
sent a crucial challenge in order to avoid, either economic damages to users or V2G
application operators, or even worse side effects due to uncontrolled and wide-area
power outages. Finally, this chapter provides additional open challenges and issues
that are related to V2G and could be explored by researchers.
REFERENCES
1. W. Xiaojun, T. Wenqi, H. JingHan, H. Mei, J. Jiuchun, and H. Haiying, The application
of electric vehicles as mobile distributed energy storage units in smart grid, in Power
and Energy Engineering Conference (APPEEC), Asia-Pacic, 2011.
2. J. Donadee and M. Ilic, Stochastic optimization of grid to vehicle frequency regulation
capacity bids, IEEE Transactions on Smart Grid, 5(2), 1061–1069, 2014.
3. J. Pinto, V. Monteiro, H. Goncalves, B. Exposto, D. Pedrosa, C. Couto, and J. Afonso,
Bidirectional battery charger with grid-to-vehicle, vehicle-to-grid and vehicle-to-home
technologies, IECON Industrial Electronics Society 39th Annual Conference of the
IEEE, Vienna, Austria, pp. 5934–5939, November 2013.
4. M. Erol-Kantarci, J. Sarker, and H. Mouftah, Communication-based plug-in hybrid
electrical vehicle load management in the smart grid, 2011 IEEE Symposium on
Computers and Communications (ISCC), Corfu, Greece, pp. 404–409, June 28–July 1,
2011.
5. J. Camp and E. Knightly, The IEEE 802.11s extended service set mesh networking
standard, IEEE Communications Magazine, 46(8), 120 –126, 2008.
6. M. Rogowsky, Tesla 90-second battery swap tech coming this year, Forbes, Retrieved
06-22-2013.
7. K. Murat, P. Manisa, and R. Saifur, Communication network requirements for major
smart grid applications in HAN, NAN and WAN, Elsevier Computer Networks, 67,
74–88, 2014.
8. E. Zountour idou, G. Kiokes, N. Hatzia rgyriou, and N. Uzunoglu, An evaluation study of
wireless access technologies for V2G communications, Intelligent System Application
to Power Systems (ISAP), vol. 2011, 16th International Conference, Creta, Greece,
pp.1–7, 25–28, September 2011.
9. M. Yilmaz and P. Krein, Review of the impact of vehicle-to-grid technologies on distri-
bution systems and utility interfaces, IEEE Transactions on Power Electronics, 28(12),
5673–5689, 2013.
10. S. Käbisch, A. Schmitt, M. Winter, and J. Heuer, Interconnections and communica-
tions of electric vehicles and smart grids, 2010 First IEEE International Conference
on Smart Grid Communications (SmartGridComm), Maryland, USA, pp. 161–166,
October 4 –6, 2010.
11. W. W. W. C. (W3C), Efcient XML Interchange Working Group, [Online]. Available:
http://www.w3.org/XML/EXI/. Accessed on March 1, 2014.
© 2016 by Taylor & Francis Group, LLC
Downloaded by [Periklis Chatzimisios] at 05:56 02 May 2016
367Vehicle-to-Grid Networks
12. T. Markel, M. Kuss, and P. Denholm, Communication and control of electric drive
vehicles supporting renewables, Vehicle Power and Propulsion Conference, 2009.
VPPC’09. IEEE, Dearborn, Michigan, pp. 27–34, September 7–10, 2009.
13. H. Menouar, F. Filali, and M. Lenardi, A survey and qualitative analysis of MAC pro-
tocols for vehicular ad hoc networks, Wireless Communications, IEEE, 13(5), 30–35,
2006.
14. L. Ning, J. Yusheng, L. Fuqiang, and W. Xinhong, A dedicated multi-channel MAC
protocol design for VANET with adaptive broadcasting, Wireless Communications and
Networking Conference (WCNC), 2010 IEEE, Syndey, Australia, pp. 1–6, April 18–21,
2010.
15. M. Booysen, S. Zeadally, and G.-J. van Rooyen, Survey of media access control proto-
cols for vehicular ad hoc networks, IET Communications, 5(11), 1619 –1631, 2011.
16. H. Guo, F. Yu, W.-C. Wong, V. Suhendra, and Y. D. Wu, Secure wireless communica-
tion platform for EV-to-grid research, Proceedings of the 6th International Wireless
Communications and Mobile Computing Conference, IWCMC 2010, Caen, France,
June 28–July 2, 2010. pp. 21–25, ACM, 2010.
17. M. Mustafa, N. Zhang, G. Kalogridis, and Z. Fan, Smart electric vehicle charging:
Security analysis, in Innovative Smart Grid Technologies (ISGT), Washington, DC,
USA, February 2013.
18. H. Liu, H. Ning, Y. Zhang, and L. Yang, Aggregated-proofs based privacy-preserving
authentication for V2G networks in the smart grid, IEEE Transactions on Smart Grid,
3(4), 1722–1733, 2012.
19. H. Chaudhry and T. Bohn, Security concerns of a plug-in vehicle, 2012 IEEE PES
Innovative Smart Grid Technologies (ISGT), Washington, DC, USA, pp. 1–6, January
16–20, 2012.
20. M. Y.-W. Chia, S. Krishnan, and J. Zhou, Challenges and opportunities in infrastruc-
ture support for electric vehicles and smart grid in a dense urban environment—
Singapore, 2012 IEEE International Electric Vehicle Conference (IEVC), Greenville,
USA, pp.1–6, March 4–8, 2012.
21. H. Liu, H. Ning, Y. Zhang, and M. Guizani, Battery status-aware authentication scheme
for V2G networks in smart grid, IEEE Transactions on Smart Grid, 4(1), 99–110, 2013.
22. H.-R. Tseng, A secure and privacy-preserving communication protocol for V2G net-
works, Wireless Communications and Networking Conference (WCNC), Paris, France,
pp. 2706–2711, April 2012.
23. H. Khurana, M. Hadley, N. Lu, and D. Frincke, Smart-grid security issues, IEEE
Security and Privacy, 8(1), 81–85, 2010.
24. H. Guo, Y. Wu, F. Bao, H. Chen, and M. Ma, UBAPV2G: A unique batch authentication
protocol for vehicle-to-grid communications, IEEE Transactions on Smart Grid, 2(4),
707–714, 2011.
25. H.-R. Tseng, On the security of a unique batch authentication protocol for vehicle-to-
grid communications, 2012 12th International Conference on ITS Telecommunications
(ITST), Taipei, Taiwan, pp. 280–283, November 2012.
26. B. Vaidya, D. Makrakis, and H. Mouftah, Security mechanism for multi-domain
vehicle-to-grid infrastructure, Global Telecommunications Conference (GLOBECOM
2011), Houston, Texas, pp. 1–5, December 2011.
27. V. Aravinthan, B. Karimi, V. Namboodiri, and W. Jewell, Wireless communication for
smart grid applications at distribution level—Feasibility and requirements, Power and
Energy Society General Meeting, 2011 IEEE, Detroit, Michigan, pp. 1–8, July 24–29,
2011.
28. S. Madi and H. Al-Qamzi, A survey on realistic mobility models for vehicular ad
hoc networks (VANETs), 2013 10th IEEE International Conference on Networking,
Sensing and Control (ICNSC), Evry, France, pp. 333–339, April 10–12, 2013.
© 2016 by Taylor & Francis Group, LLC
Downloaded by [Periklis Chatzimisios] at 05:56 02 May 2016
368 Smart Grid
29. S. Singh and S. Agrawal, VANET routing protocols: Issues and challenges, 2014
Recent Advances in Engineering and Computational Sciences (RAECS), Chandigarh,
India, pp. 1–5, March 6–8, 2014.
30. H. Sharma, P. Agrawal, and R. Kshirsagar, Multipath reliable range node selection
distance vector routing for VANET: Design approach, Electronic Systems, Signal
Processing and Computing Technologies (ICESC), Nagpur, India, 2014.
31. S. Uludag and N. S. K. Akkaya, A survey of routing protocols for smart grid commu-
nications, Elsevier Computer Networks, 56(11), 2742–2771, 2012.
32. Y. Zhang, S. Gjessing, H. Liu, H. Ning, L. Yang, and M. Guizani, Securing vehicle-to-
grid communications in the smart grid, IEEE Wireless Communications, 20(6), 66 –73,
2013.
33. Y. Li, R. Wang, P. Wang, D. Niyato, W. Saad, and Z. Han, Resilient PHEV charging
policies under price information attacks, 2012 IEEE Third International Conference
on Smart Grid Communications (SmartGridComm), Tainan, Taiwan, pp. 389–394,
November 2012.
34. TeslaMotors, About Tesla Motors, December 2013. [Online]. Available: http://www.
teslamotors.com/about/press/releases/tesla-dramatically-expands-supercharger-
network-delivering-convenient-free-long. Accessed on June 11, 2014.
35. M. Yilmaz and P. Krein, Review of benets and challenges of vehicle-to-grid technol-
ogy, 2012 IEEE Energy Conversion Congress and Exposition (ECCE), Raleigh, North
Carolina, pp. 3082–3089, September 15–20, 2012.
36. J. Driesen, K. Clement-Nyns, and E. Haesen, The impact of charging PHEVs on a resi-
dential distribution grid, IEEE Transactions on Power Systems, 25(1), 371–380, 2010.
37. K. J. Dyke, N. Schoeld, and M. Barnes, The impact of transport electrication on
electrical networks, IEEE Transactions on Industrial Electronics, 57(12), 3917–3926,
2010.
38. H. Liu, H. Ning, Y. Zhang, Q. Xiong, and L. Yang, Role-dependent privacy preserva-
tion for secure V2G networks in the smart grid, IEEE Transactions on Information
Forensics and Security, 9, 208–220, 2014.
39. C. Liu, K. Chau, D. Wu, and S. Gao, Opportunities and challenges of vehicle-to-home,
vehicle-to-vehicle, and vehicle-to-grid technologies, Proceedings of the IEEE, vol. 101,
no. 11, pp. 2409–2427, November 2013.
40. P. Siano, Demand response and smart grids—A survey, Elsevier Renewable and
Sustainable Energy Reviews, 30, 461–478, 2014.
41. S. Hossein Imani, S. Asghari, and M. Ameli, Considering the load uncertainty for
solving security constrained unit commitment problem in presence of plug-in electric
vehicle, Electrical Engineering (ICEE), Tehran, Iran, pp. 725–732, May 2014.
42. D. Tuttle and R. Baldick, The evolution of plug-in electric vehicle–grid interactions,
IEEE Transactions on Smart Grid, 3(1), 500–505, 2012.
43. J. Haack, B. Akyol, N. Tenney, B. Carpenter, R. Pratt, and T. Carroll, VOLTTRON™:
An agent platform for integrating electric vehicles and smart grid, 2013 International
Conference on Connected Vehicles and Expo (ICCVE), Las Vegas, Nevada, pp. 81–86,
December 2–6, 2013.
44. B. Jansen, C. Binding, O. Sundstrom, and D. Gantenbein, Architecture and com-
munication of an electric vehicle virtual power plant, Smart Grid Communications
(SmartGridComm), Gaithersburg, Maryland, pp. 149–154, October 2010.
45. J. Schmutz ler, S. Groning, and C. Wietfeld, Management of dist ributed energy resources
in IEC 61850 using web services on devices, 2011 IEEE International Conference on
Smart Grid Communications (SmartGridComm), Brussels, Belgium, pp. 315–320,
October 17–20, 2011.
46. U. F. S. UPnP Device Architecture 1.1. [Online]. Available: http://www.upnp.org/specs/
arch/UPnP-arch-DeviceArchitecturev1.1.pdf. Accessed on February 18, 2014.
© 2016 by Taylor & Francis Group, LLC
Downloaded by [Periklis Chatzimisios] at 05:56 02 May 2016
369Vehicle-to-Grid Networks
47. H. Xiao, Y. Huimei, W. Chen, and L. Hongjun, A survey of inuence of electrics
vehicle charging on power grid, IEEE 9th Conference on Industrial Electronics and
Applications (ICIEA), Hangzhou, China, pp. 121–126, June 2014.
48. W. Tian, Y. Jiang, M. Shahidehpour, and M. Krishnamurthy, Vehicle charging sta-
tions with solar canopy: A realistic case study within a smart grid environment, IEEE
Transportation Electrication Conference and Expo (ITEC), Dearborn, Michigan,
pp.1–6, June 15–18, 2014.
49. R. Yu, J. Ding, W. Zhong, Y. Liu, and S. Xie, PHEV charging and discharging coopera-
tion in V2G networks: A coalition game approach, IEEE Internet of Things Journal,
1(6), 578–589, 2014.
50. E. Dehaghani and S. Williamson, On the inefciency of vehicle-to-grid (V2G)
power ow: Potential barriers and possible research directions, IEEE Transportation
Electrication Conference and Expo (ITEC), Dearborn, Michigan, pp. 1–5, June
18–20, 2012.
51. D. Dallinger, J. Link, and M. BÜttner, Smart grid agent: Plug-in electric vehicle, IEEE
Transactions on Sustainable Energy, 5(3), 710–717, 2014.
52. J. Donoghue and A. Cruden, Whole system modelling of V2G power network con-
trol, communications and management, Electric Vehicle Symposium and Exhibition
(EVS27), Barcelona, Spain, pp. 1–9, November 17–20, 2013.
© 2016 by Taylor & Francis Group, LLC
Downloaded by [Periklis Chatzimisios] at 05:56 02 May 2016
