See discussions, stats, and author profiles for this publication at: [Link]

net/publication/261116600

Smart city for VANETs using warning messages, traffic statistics and intelligent
traffic lights

Conference Paper · June 2012

DOI: 10.1109/IVS.2012.6232229

CITATIONS READS
178 1,133

5 authors, including:

Carolina Tripp-Barba Ahmad Mohamad-Mezher

Universidad Autónoma de Sinaloa University of New Brunswick
62 PUBLICATIONS 554 CITATIONS 47 PUBLICATIONS 460 CITATIONS

SEE PROFILE SEE PROFILE

Mónica Aguilar-Igartua
Universitat Politècnica de Catalunya
118 PUBLICATIONS 806 CITATIONS

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

MobilitApp View project

CIPSEC (Enhancing Critical Infrastructure Protection with innovative Security framework ) View project

All content following this page was uploaded by Carolina Tripp-Barba on 26 June 2014.

The user has requested enhancement of the downloaded file.

 2012 Intelligent Vehicles Symposium
Alcalá de Henares, Spain, June 3-7, 2012

Smart city for VANETs using warning messages,

traffic statistics and intelligent traffic lights
Carolina Tripp Barba, Miguel Ángel Mateos, Pablo Regañas Soto, Ahmad Mohamad Mezher,
Mónica Aguilar Igartua
Departament d’Enginyeria Telemàtica.
Universitat Politècnica de Catalunya (UPC). Barcelona, Spain
{ctripp, [Link], [Link]}@[Link], {[Link], pauregas}@[Link]

Abstract—Road safety has become a main issue for gov- emergency services. VANETs may soon allow vehicles to
ernments and car manufacturers in the last twenty years. easily communicate among themselves and also with fixed
The development of new vehicular technologies has favoured infrastructure. This will not only improve road safety, but also
companies, researchers and institutions to focus their efforts on
improving road safety. During the last decades, the evolution of raise new commercial opportunities such as infotainment for
wireless technologies has allowed researchers to design commu- passengers.
nication systems where vehicles participate in the communication Car accident prevention, safer roads, pollution and con-
networks. Thus, new types of networks, such as Vehicular gestion reduction are some of the goals of VANETs. The
Ad Hoc Networks (VANETs), have been created to facilitate deployment of an efficient system to manage warning mes-
communication between vehicles themselves and between vehicles
and infrastructure. New concepts where vehicular networks play sages in VANETs has important benefits, from the perspective
an important role have appeared the last years, such as smart of both road operators and drivers. Efficient traffic alerts
cities and living labs [1]. Smart cities include intelligent traffic and updated information about traffic incidents will reduce
management in which data from the TIC (Traffic Information traffic jams, increase road safety and improve the driving
Centre) infrastructures could be reachable at any point. To test in the city. Furthermore, from the sustainable and economic
the possibilities of these future cities, living labs (cities in which
new designed systems can be tested in real conditions) have perspective, real-time traffic alerting will reduce trip time and
been created all over Europe. The goal of our framework is fuel consumption and therefore decrease the amount of CO2
to transmit information about the traffic conditions to help the emissions [2].
driver (or the vehicle itself) take adequate decisions. In this work, In this work, a smart city framework has been developed
the development of a warning system composed of Intelligent in where intelligent traffic lights (ITLs) set in the crossroads
Traffic Lights (ITLs) that provides information to drivers about
traffic density and weather conditions in the streets of a city is of a city are involved. These ITLs are in charge of gathering
proposed and evaluated through simulations. traffic information (e.g. traffic density) from passing vehicles,
Index Terms—Vehicular Ad Hoc Networks (VANETs), Traffic updating traffic statistics of the city and reporting those statis-
Information Centre (TIC), Smart Cities, Intelligent Transporta- tics to the vehicles. Also, ITLs will send warning messages to
tion System (ITS). vehicles in case of accidents to avoid further collisions.
The rest of the paper is organized as follows. Section II gives
I. I NTRODUCTION a state of the art of some relevant works of VANETs using ITS.
During the last few years, progress in wireless communi- Section III introduces our smart city framework. Section IV
cations has offered new research fields, providing network presents our proposal to calculate the vehicles’ density in the
connectivity in environments where wired solutions are im- city using ITLs. Section V summarizes our proposed scheme
possible. Among these, vehicular ad hoc networks (VANETs) of warning messages. Section VI shows simulation results.
are attracting a growing attention due to the promising im- Finally, section VII gives conclusions and future work.
portant applications, from road safety to traffic control and
entertainment for passengers. Smart cities would like to plan II. S TATE OF THE A RT
how to minimize their transportation problems due to the During the last decades, Intelligent Transportation Systems
increasing population that produces congested roads. VANETs (ITS) have emerged as an efficient way to improve the per-
aim at helping to alleviate this issue improving vehicles’ formance of the flow of vehicles in the roads. The goals of
mobility, increasing road safety and also seeking to have more ITLs are road safety, comfortable driving and distribution of
sustainable cities. updated information about the roads. Many proposals about
At the beginning of the development of vehicular tech- ITS have been presented in recent years. In this section some
nologies, the main goal was to have more efficient and works about ITS in smart cities are highlighted.
safer roads. Nowadays, thanks to the huge development of The work in [3] is a survey about multifunctional data-
wireless technologies and their application in vehicles, it is driven intelligent transportation system (D2 ITS), which are
possible to use Intelligent Transportation System (ITS) that supported by a large amount of data collected from various
will change our way to drive, improve road safety, and help resources: Vision-Driven ITS (input data collected from video

978-1-4673-2118-1/$31.00 ©2012 IEEE 902

sensors and used recognition including vehicle and pedestrian
detection); Multisource-Driven ITS (e.g. inductive-loop detec-
tors, laser radar and GPS); Learning-Driven ITS (effective
prediction of the occurrence of accidents to enhance the safety
of pedestrians by reducing the impact of vehicle collision);
and Visualization-Driven ITS (to help decision makers quickly
identify abnormal traffic patterns and accordingly take neces-
sary measures).
In [4] and [5] two adaptive traffic light systems based on
wireless between vehicles and fixed controller nodes deployed
at intersections are designed and developed. These systems
improve traffic fluency, reduce the waiting time of vehicles at
intersections and help to avoid collisions.
The e-NOTIFY [6] system was designed for automated
accident detection, reporting to the Emergencies Coordination
Center, and assistance of road accidents using the capabilities
Fig. 1. Intelligent Traffic Lights distribution
offered by vehicular communication technologies. e-NOTIFY
focus on improving post collision care with a fast and efficient
management of the available emergency resources, which
increases the chances of recovery and survival for those injured converge on this intersection. ITLs are placed as shown in
in traffic accidents. Fig. 1. To cover all this area the antenna pattern used is
In this work, we focus on the development of a smart city an omnidirectional propagation pattern. Therefore, each ITL
framework using intelligent infrastructure in the streets, in receives data from all passing vehicles on its cover range
our case intelligent traffic lights (ITLs). ITLs provide warning (the four streets and the intersection). Not having an ITL on
messages to the passing vehicles to inform drivers about traffic each intersection is more economic when implementing this
and weather conditions of the different streets of the city. This framework.
way, the smart city framework can help drivers to have a better
trip, reducing time to destination, preventing accidents and It is assumed that vehicles have a global positioning system
traffic jams and also saving petrol and reducing pollution. As (GPS) device, a driver assistant device, full map information
[3], our proposal manages traffic information seeking to avoid of the city including the position of the ITLs. Thus, vehicles
accidents, although the information here is gathered from the can easily select which is the nearest ITL.
vehicles themselves so no further infrastructure is needed. Also Every ad-hoc node (i.e., ITLs and vehicles) set on the
our proposal could easily be used by the traffic information scenario was configured with Ad hoc On-Demand Distance
centre to design an adaptive traffic light system similar to [4] Vector (AODV) [8] routing protocol. AODV was selected
and [5]. because of its simplicity. Although it is well known that
AODV is not suitable as routing protocol of general use in
III. S MART C ITY F RAMEWORK VANETs, there are some applications that might work well
The smart city framework we have designed includes ITLs with AODV. The advantage of AODV is its simplicity and
set in some of the crossroads. These ITLs collect real-time widespread use. The main drawback is that AODV needs end-
traffic data from the passing vehicles and calculate traffic to-end paths for data forwarding, which is difficult to handle
statistics such as traffic density in the adjacent streets (between because in VANETs end-to-end paths last not much due to
consecutive crossroads). At the same time, these ITLs can high speeds of vehicles. Other routing protocols that use other
communicate the traffic information to passing vehicles and strategies like greedy forwarding and geographical routing.
alert them with warning messages in case of accidents. These For instance, GPSR (Greedy Perimeter Stateless Routing) [9]
ITLs also form a sub-network that allows ITLs to share the and GOSR (Geographical Opportunistic Source Routing) [10]
collected information and calculate statistics of the whole city. have shown good performance in VANETs, but at the cost of
Thus, vehicles are well informed of the traffic situation in greater complexity and increased delay. Nonetheless, for some
the city. The following sections describe how this smart city applications that require a short delay AODV can perform
framework is designed and which use the ITL will have. well. In this paper we are considering smart city services
In the smart city projected, blocks have a regular square where vehicles send warning messages (weather conditions
design and buildings on its four sides. ITLs are responsible of and traffic density) to the closest ITL, so it is not necessary
managing the traffic of the vehicles, which form a VANET. to establish long paths that last long. Instead, vehicles need
These ITLs do not have to be located at each intersection. to establish very short paths (1-2 hops) to the nearest ITL.
Within all the traffic lights that are traditionally located in a Besides, the communication must be quickly since vehicles
city, only a few will be replaced by ITLs. This is because move fast and the period in coverage range of the ITL is short.
each ITL covers a whole intersection and the 4 streets that Thus, AODV is suitable for our purposes.

903
 TABLE I
F ORMAT INFORMATION OF MESSAGE STAT with the others ITLs in the city through the sub-network
Type Statistic Message (SM) STAT they form. The day has been divided into five periods due
stat type Traffic density (T Dst ) 0 to the usually variable traffic densities in a city throughout the
stat my id Car sending statistics Ci day. Thus, every ITL updates the traffic density per periods:
stat neighbours Number of neighbours (N oN ) NoNi
stat time Time of statistics report ti
T Dst6−9 , T Dst9−12 , T Dst12−15 , T Dst15−18 , T Dst18−21 .
stat dst ITL IP address ITLi For instance, T Dst6−9 gathers the average traffic density in
the city, during week days, from 06:00 AM to 09:00 PM. The
value T Dst6−9 will continuously be updated using Eq. (1),
where w is a small weight (e.g. w=0.25) to smooth out isolated
deviations, T Dst6−9,i is the updated average in iteration i and
T Dst6−9 is the last value received by that ITL. The same
computation will be done for the other periods of the day.

T Dst6−9,i = w· T Dst6−9,i−1 + (1 − w)· T Dst6−9 (1)

The ITLs of the city share that traffic information and

after that, each ITL will send back to each passing vehicle
a message with the updated traffic statistics of the city on that
period of time. With this information, the driver’s assistant
device can take proper trip decisions (e.g. avoiding congested
roads). Also, data routing protocols may use that information
Fig. 2. ITL obtaining traffic statistics in its intersection
to take suitable forwarding decisions (e.g. forward the packet
through denser streets where there are more possible forward-
ing nodes).
IV. M ANAGEMENT OF T RAFFIC D ENSITY
In this work we focus on the analysis of traffic density, V. M ANAGEMENT OF WARNING M ESSAGES
although similar analysis could easily be done for other traffic In the promising smart cities of near future, communications
statistics (e.g. number of passengers, trip time). between vehicles and the city will be constant, including
The messages sent by each vehicle to an ITL include the infrastructure-to-infrastructure, car-to-car and infrastructure-
type of message (a new message called Statistic Message, to-car communications, by means of city infrastructure and
SM), the identification of the vehicle (Ci ) sending the message, Traffic Information Centres (TICs). These packets will contain
the current value of the number of neighbours (N oNi ) in different type of information and should be prioritized accord-
its coverage range at that moment, the moment in which the ingly. For instance, packets containing information about an
message was sent (ti ) and the IP address of the ITL destination accident have to be prioritized over those containing other kind
(ITLi ). This message is sent by the vehicles each 2 sec. This of data such as entertainment data.
way, a car (v=40 km/h) sends 5 messages while it crosses a Upon the reception of a warning message, a vehicle should
100 m. street. The format of this message is shown in Table consider its current distance to the initial source of the warning
I. The ITL will update the traffic statistics upon the reception message and act consequently. For instance, a car being a long
of each new message, as it is explained below. distance away from an accident will not act the same way (i.e.,
Fig. 2 shows the procedure of getting and sending traffic will not brake) when receiving a warning about the accident
statistics from the vehicles to the ITLs. Each vehicle inter- since it does not affect the immediate security of that car.
changes hello messages (HM) with its neighbours and this Nonetheless, that warning message will inform the driver of
way it knows the amount of vehicles on its transmission that car (actually, the driver’s assistant device), who may vary
range. Then, the vehicle sends a Statistic Message (SM) with the trip plan consequently.
the number of neighbours to the nearest ITL. For example, We have implemented a simple warning service to prevent
C1 counts with three neighbours (C2 , C3 , and C4 ). Notice further collisions by alerting drivers about accidents and
that although C7 is inside its range they cannot establish dangerous road conditions. To achieve that goal, vehicles
any communication because of the buildings that represent send short warning messages once one of the situations
obstacles. The car C5 does not see any neighbour around so depicted in Table II has been detected. This information can
it sends a SM to the nearest ITL with a zero on it. be obtained from different sources. Regarding weather, data
ITL1 will receive the SMs and will update the traffic density can be collected by a Wireless Sensor Network (WSN) that
statistics by using an exponential weighted moving average periodically transmits the weather conditions to the nearest
(EWMA) to average current and historical values. Then, ITL1 ITL. Also, from small weather stations set in a few ITLs of
will store the results properly and will share its statistics the city. This information is spread through the city using

904
 TABLE II
WARNING MESSAGES: TRAFFIC AND WEATHER CONDITIONS

U : Initial driver speed

Fig. 3. Simulation scenario of a car accident

the sub-network formed by the ITLs. Complementarily, the
forecast proportioned by local public weather services could
be used as well. The sub-network of ITLs could share that sub-networks. One of the interfaces of the ITL will commu-
forecast information obtained from an Internet access point nicate with vehicles and the other interface will communicate
set in one of the ITLs. In case of accident, the vehicle itself with the other ITLs in the city.
(using sensors that detect that the car suffered an accident) To ensure node connectivity between ITLs, interface two
communicates this situation to the closest ITL and to the (used to connect with other ITLs) is determined by the distance
neighbouring vehicles. needed to cover 300 m between crossroads. Interface one (used
To know the traffic density, each ITL uses the statistics to connect with vehicles) communicates with the passing vehi-
collected by the network of ITLs in the city (as explained in the cles. As the streets have a length of 100 m, and the intersection
previous section) regarding the average number of neighbours of 40 m, this interface is configured to cover 130 m. Both
per vehicle in the streets along the day. Thus, depending on interfaces have an omnidirectional propagation diagram. As
the average number of neighbours, two adaptive thresholds each vehicle is only supposed to connect to interface one of the
determine the traffic density of the road: free, semi-congested ITL and also to other cars, the transmission range of vehicles
or very congested (see Table II). is configured to cover 130 m in an omnidirectional pattern.
We use a 4-bit field in the warning messages to code To ensure that it could be possible to differentiate data
traffic density (2 bits) and weather information (2 bits). The communication of the two wireless ad-hoc interfaces, they are
warning message also includes a field with the location of configured with two different sub-networks. The sub-network
the initial place of the warning message. Nearby vehicles that used by the ITLs to collect, send and calculate statistics from
receive such message will reduce their speed depending on the the vehicles is “[Link]/24”, whereas the sub-network used
warning message according to Table II. For instance, in a very to communicate ITLs among themselves is “[Link]/24”.
congested road with rain condition, warning messages inform There are only a few ITLs among the regular traffic lights.
nearby vehicles to reduce their speed to 40% of the initial In this case, an ITL receives data from any passing vehicle
driver speed (U in Table II). The driver’s assistant device in from any of the four streets covered by that ITL.
the vehicle will make the vehicle brake accordingly. B. Benefits of Using Warning Messages after Accidents
To evaluate the operation of warning messages, we use an
VI. S IMULATION R ESULTS urban scenario to show how the vehicles react under different
In this work, we evaluate the performance of the vehicles of traffic and weather conditions. Fig. 3 shows a neighbourhood
a VANET in a smart city using warning messages and traffic of the city where a car accident will happen. The simulation
statics managed by the ITLs set in some crossroads of the recreates a Manhattan 280x280 m2 scenario. The length of the
city. To achieve this evaluations we use the network simulator streets is 100 m, and the size of every cross is 40 m. These
NCTUns 6.0 [7] values were chosen to emulate the regular streets in the city
of Barcelona, Spain. In the scenario there are 5 vehicles and
5 ITLs. During the simulation, vehicle C18 has an accident
A. Configuration of Intelligent Traffic Lights (ITLs)
and remains broken close to the intelligent traffic light ITL11
ITLs are implemented using Multi-Interface Mobile nodes situated in the centre, see Fig. 3. Vehicles C15 , C16 , C17
with two wireless ad-hoc interfaces configured in two different and C19 are all travelling towards ITL11 . In this simulation

905
 TABLE III
TRAFFIC AND WEATHER CONDITIONS DURING THE
SIMULATION
Traffic Weather Period Average vehicle
conditions conditions of time velocity
Free segment Sun 30 s 40 km/h
Accident Sun 50 s 0 km/h

TABLE IV
SIMULATION SETTINGS
Parameter Value
Medium capacity 11 Mbps
Packet size 256 bytes
Transmission range 130 m
Carrier sense range 180 m
Simulation time 80 sec
MAC specification IEEE 802.11b
Area 320 x 320 m2
Maximum Average speed 40 km/h
Number of nodes 5 ITLs and 8 vehicles
Mobility model CarAgentMod (NCTUns)
Routing protocol AODV Fig. 4. Traffic density simulation scenario

city respecting streets, crossroads and traffic lights.

it is shown how the use of ITL helps to avoid collision As it was said in Section III, ad-hoc nodes (i.e. ITLs and
among the other vehicles and the broken vehicle, thanks to vehicles) use AODV. We simply modified AODV to be able
our warning scheme that makes them brake beforehand. Table to collect traffic statistics while establishing the routing paths.
III summarizes the traffic and weather conditions during the To do that, we use modified RREQ messages that carry SMs
simulation. (see Table I).
The accident will occur in the second 30. The traffic lights The objective of these simulations is to evaluate if the
number ITL10 , ITL12 , ITL13 and ITL14 will send messages of use of ITLs reduce the driver’s reaction time after accidents.
good weather conditions and free traffic segment during the According to the Dirección General de Tráfico (DGT) [11],
80 sec of this simulation. The traffic light number ITL11 will responsible of the transportation policy in Spain, the average
send during 30 sec messages of good weather conditions and reaction time of a driver is 1 sec, so a car (v = 40 km/h) before
free traffic segment. After that, ITL11 will send good weather start braking still travels 18,63 m. Using our framework the
conditions and accident during the next 50 sec. Each of the driver’s reaction time was 0,084 sec, which represents that the
ITLs sends these packets to the vehicles in the four streets that distance travelled will be reduced to 8,45 m. Table V shows
go from the crossroads where they are located to the next 4 the time and distances that a vehicle, in average, travels with
closest crossroads. and without the use of our smart city framework. In this case,
ITLs will broadcast 256 bytes messages every 0.2 seconds it can be appreciated that the safety distance from the car to the
(i.e. 5 messages each second) with information about traffic obstacle has been reduced around 55% from 18,63 m without
conditions, weather conditions and accident warnings. Vehicles the use of ITLs to 8,45 m using them, which increases road
move randomly through the streets at an average velocity safety notably.
of 40 km/h (it automatically decreases when approaching an
intersection and the vehicle turns). Simulation settings are C. Measure of the Traffic Density in the Smart City
summarized in Table IV.
To evaluate the operation of the traffic statistics system, we
The cars in the simulation were represented using smart
use a Manhattan scenario with streets that form 4x4 blocks
vehicles equipped with IEEE 802.11b interface on ad-hoc
(100m x 100m each) (Fig. 4). It has obstacles that represent
mode. These vehicles are controlled by a program called
buildings, and ITLs which are responsible to manage the traffic
agent ([Link]) that makes vehicles move through the
of the vehicles that form the VANET.
The simulation consists on a random number of smart vehi-
TABLE V cles moving around the city and establishing communications
DRIVER’S REACTION TIME AND DISTANCE TRAVELLED
with the nearest ITL to send information of the current number
Use of Non use of
ITLs ITLs of neighbours. Traffic statistics are updated as explained in
Driver’s reaction time 0,084 s 1s section IV and according to eq. (1). This data is collected
Distance travelled till reaction 0,93 m 11,11 m every 2 sec. Every time an ITL receives data from a passing
Braking period of time 1,355 s 1,355 s car it updates the statistics of traffic density on its surrounding
Distance travelled during braking 7,52 m 7,52 m
Total distance travelled 8,45 m 18,63 m area, stores it on an individual file and shares it with the rest
of the ITLs of the city.

906
 statistics. The goal is that the driver’s assistant device can take
proper trip decisions, for instance to avoid congested roads,
and therefore reducing the trip time and pollution as well.
Besides, our smart city framework includes warning messages
sent by possible broken vehicles to make approaching vehicles
brake beforehand and thus avoid more collisions. Simulation
results show the effectiveness of this scheme, reducing the
distance to brake and the driver’s reaction time.
As a near future work, ITLs could communicate to passing
vehicles indicating where are the free parking spots in the
city. With this information, the driver assistant device could
indicate the driver where free spots are located. This system
could use a WSN to get the data about free parking spots and
communicate it to the nearest ITLs. The ITLs could share that
Fig. 5. Average number of neighbours per vehicle measured by ITL44 , set information though the sub-network they form. This would
in downtown save trip time, petrol and CO2 as a consequence, which helps
to have sustainable smart cities.
Also, statistics collected by the ITLs can improve data
routing protocols selecting the path that offers a higher chance
to forward a packet to the destination successfully. We will de-
sign a VANET routing protocol that considers those statistics
in its operation.
ACKNOWLEDGEMENTS
This work has been partially funded by the Spanish Ministry
of Science and Education under the project CICYT CONSE-
QUENCE (TEC2010-20572-C02-02) and partially supported
by the Comissionat per a Universitats i Recerca del DIUE
from the Generalitat de Catalunya and the Social European
Budget (Fons Social Europeu) with the grant FI-AGAUR; and
by the Autonomous University of Sinaloa, México.
Fig. 6. Average number of neighbours per vehicle measured by ITL59 , set R EFERENCES
in the city entrance
[1] European Network of Living Labs (ENoLL),
[Link]
[2] Ferrari, G., Busanelli, S., Lotti, N., Kaplan, Y., “Cross-Network Informa-
Fig. 5 shows the behaviour of the statistics collected by tion Dissemination in VANETs”, 11th International Conference on ITS
Telecommunications, pp. 351-356, 2011.
ITL44 (set in downtown, see Fig. 5). Here, simulations show [3] Junping, Z., Fei-Yue, W., Kunfeng, W., Wei-Hua, L., Xin, X., Cheng, C.,
400 sec (i.e. 15 h from 6 AM to 9 PM), so that 27 sec in the “Data-Driven Intelligent Transportation Systems: Survey”, IEEE Trans-
simulations represents 1 h. The results show the density of actions on Intelligent Transportation Systems, Vol. 12, Issue 4, pp. 1624-
1639, 2011.
cars in downtown along the day. With this information drivers [4] Maslekar, N., Boussedjra, M., Mouzna, J., Labiod, H., “VANET based
can obtain which are the roads more congested in each part of Adaptive Traffic Signal Control”, IEEE 73rd Vehicular Technology Con-
the city. Fig. 6 shows the results obtained by ITL59 , which is ference (VTC Spring), pp. 1-5, 2011.
[5] Gradinescu, V., Gorgorin, C., Diaconescu, R., Cristea, V., Iftode, L.,
located in the entrance of the city. We can see the behaviour “Adaptive Traffic Light Using Car-to-Car communications”, IEEE 65th
in one day, where the more congested periods of time are Vehicular Technology Conference (VTC Spring), pp. 21-25, 2007.
between 12-15 PM. Streets are almost free between 6-9 AM. [6] Fogue, M., Garrido, P., Martinez, F. J., Cano, J. C., Calafate, C. T., Man-
zoni, P., Sanchez, M., “Prototyping an Automatic Notification Scheme
and 18-21 PM. for Traffic Accidents in Vehicular Networks”, Wireless Days (WD) IFIP,
pp. 1-5, 2011.
VII. CONCLUSION AND FUTURE WORK [7] The NCTUns 6.0 Network Simulator and Emulator. Available at
[Link]
In this work we have designed a smart city framework [8] Perkins, C.E., Belding-Royer, E. M., Das, S.R., “Ad hoc on-demand
for VANETs that include intelligent traffic lights (ITLs) that distance vector (AODV) routing”, IEEE Personal Communications, pp.
16-28, 2001.
transmit warning messages and traffic statistics. We have [9] Karp, B., Kung, H. T., “GPSR: Greedy Perimeter Stateless Routing for
implemented the framework in the NCTUns 6.0 [7] simulator. wireless Networks”, MobiCom 2000.
Simulation results show that the use of ITLs in smart cities [10] Zhongyi l., Tong, Z., Wei, Y., Xiaoming, L., “GOSR: geographical
opportunistic source routing for VANETs”, ACM SIGMOBILE Mobile
can not only improve road safety but also the driver’s quality Computing and Communications, Vol. 13, Issue 1, 2009.
of life. We have explained how the ITLs gather traffic and [11] Dirección General de Tráfico “Revista Tráfico 192”, 2011,
weather conditions of the roads and how they update those [Link]

907

View publication stats