Introducing Cellular Network Layer into SUMO for Simulating

Vehicular Mobile Devices’ Interactions in Urban Environment

Siim-Toomas Marran, Artjom Lind and Amnir Hadachi

ITS Team, Institute of Compute Science, University of Tartu,

Ulikooli 17, 51014, Tartu, Estonia

Keywords:

Simulation, Mobility Big Data, Call Detail Records, Cellular Networking, Urban Road Trafﬁc, GPS Devices,

Mobile Devices, SUMO.

Abstract:

During the last decade researchers have been demonstrating the importance of mobile data or CDR data in

depicting the human mobility patterns. However, this type of data is not easy to get access to from mobile

operators. Besides, in order to make this type of data available and enable their usage for the scientiﬁc com-

munities the process can face many constraints that can constitute obstacle. From this perspective, this paper

introduces a way to produce realistic real-life mobility logs through the trafﬁc simulation tool SUMO, which

has been enhanced with a cellular network layer to mimic cellular networking behavior.

1 INTRODUCTION

The mobility data is a broad connotation that refers

to a various types of data, which describes people’s

movement and activities. Some of the common types

of the mobility data are Call Detail Records (CDR)

(Hadachi et al., 2014), Global Positioning System

(GPS) logs (Hadachi et al., 2013), Vehicular ad hoc

network (VANET) logs (Karnadi et al., 2007), Radio-

frequency identiﬁcation (RFID) system logs (Finken-

zeller and M

uller, 2010), WiFi access points rela-

ted data (Bulut and Szymanski, 2013), etc. Moreo-

ver, with all the advancement in information and tele-

communication technologies (ICT). Mobile data has

a great potential in sensing the urban mobility dy-

namics. From this perspective, it is clear that mo-

bile data or CDR data are valuable. In general, CDR

are data records, produced inside of the cellular net-

work between the user equipment and the base sta-

tion, documenting various telecommunication tran-

sactions. Besides, it is estimated according to (Sta-

tista, 2017) that 4.77 billions mobile users produce

daily unimaginable amount of data, with restricted

access. The mobility big data has been used for de-

veloping many different algorithms and applications

such as triangulation and trilateration, Kalman ﬁlter

for mobile positioning (Lind et al., 2017) , hand-

ling the crowd management (Pan et al., 2013) , de-

ploy real-time incident detection solutions (Zaldivar

et al., 2011) , manage and avoid congestions in the

trafﬁc ,develop Vehicle-To-Infrastructure (V2I) appli-

cations , bind the mobility with various Internet of

Vehicles (IoV) and Geographic Information Systems

(GIS) to achieve new systems upgraded with the hu-

man activity related functionality. As a consequence,

there occur moments when it is not feasibly easy to

acquire real data for speciﬁc purpose, we rely on si-

mulating and generating synthetic data based on mat-

hematical models. For example, we have open source

SUMO simulator (DLR, 2017) that provides the abi-

lity to simulate real-life urban trafﬁc at microscopic

level. This open source simulation package provi-

des realistic simulation due to the OpenStreetMap

(OSM) (OSM, 2017) map import functionality and

in-built conﬁgurable behavior models. The routable

entities can be equipped with various devices. In this

paper, we are integrating we are integrating SUMO

with mobile and GPS devices. Hence, we introduced

also a cellular network layer with all its functionalities

linking the routable devices and the network. This al-

lows us to generate from the microscopic trafﬁc simu-

lator the associated CDR and GPS logs, which can be

later used for the training and testing various machine

learning models in the real life context.

2 RELATED WORK

SUMO is a suite of applications to prepare and to si-

mulate various real-life trafﬁc scenarios. Therefore, it

has been used for many intelligent transport systems

582

Marran, S., Lind, A. and Hadachi, A.

Introducing Cellular Network Layer into SUMO for Simulating Vehicular Mobile Devices’ Interactions in Urban Environment.

DOI: 10.5220/0006808305820589

In Proceedings of the 4th International Conference on Vehicle Technology and Intelligent Transport Systems (VEHITS 2018), pages 582-589

ISBN: 978-989-758-293-6

related research topics. For example, the project Traf-

ﬁcOnline is a good illustration of the use of SUMO

for simulating the usage of mobile phone data, where

the real-world GSM data was used to develop a telep-

hony model (consisting of the following properties:

call start, call duration), to determine the travel times

(Krajzewicz et al., 2012). However, the dynamic pro-

perties, e.g. cell size variations, cell changing mecha-

nics, of the GSM network were not considered.

Another related research topic is VANET simulati-

ons. The researcher tried to create an integration

between the trafﬁc simulator SUMO and third party

network simulators, such as OMNet++ (OMNeT++,

2017), where they have created bidirectionally cou-

pled hybrid simulators and Viens (Sommer et al.,

2008). Besides, the list is long for example there

are also NS-2 (ns 2, 2017) and NS-3 (ns 3, 2017)

that have been used for VANET system’s for de-

tailed and realistic performance evaluation. Similar

to Veins, TraNS (Trafﬁc and Network Simulation En-

vironment) based on NS-2, was introduced to com-

bine two disjointly developed research simulators into

one in order to deal with VANET performance evalu-

ation (Piorkowski et al., 2008). The NS-3 was in-

troduced as replacement to NS-2 and it was used in

the VANET crossroad scenario to evaluate the per-

formance of HWMP, OLSR and DD routing proto-

cols, while using IEEE 802.11p standard and Two-

RayGround Propagation Loss Model, to send multiple

Constant Bit Rate (CBR) ﬂows over UDP between 20

source-destination pairs. (Kolici et al., 2015)

Furthermore, we have the emergence of the vehi-

cular networking concept, where the vehicle is con-

nect to everything. Therefore, the birth of many

type of communications related to VANET such

as Vehicle-To-Everything (V2X), vehicle-to-vehicle

(V2V), and vehicle-to-infrastructure (V2I). The main

purpose of introducing the connectivity into the vehi-

cle to solve issues related to trafﬁc control and ma-

nagement via the share or collection of reliable in-

formation from the vehicles. For this purpose, many

simulators have been created such as VSimRTI (We-

del et al., 2009). Obviously VANET solutions aren’t

only viable solutions to tackle the road trafﬁc esti-

mation problem. Due the widespread of the mobile

communications we can estimate the trafﬁc density

on the bigger and more important roads with the help

of cellular network, since many commuters are car-

rying mobile devices with them. Such approach has

proved to be accurate and very capable to detect and

quantify state changes (Bolla and Davoli, 2000).

3 SYSTEM DESIGN AND

ARCHITECTURE

The purpose of this simulation is to produce CDR-

like activity logs with the help of microscopic road

trafﬁc simulator SUMO, where we added an extra

layer – cellular networking. This integration of a new

layer introduces to SUMO’s core and support packa-

ges many changes and new features.

3.1 Models

The most important step in the process of integrating

the cellular network into SUMO is to deﬁne the mo-

dels of each new component. The new models can

be divided into two main categories: Cellular network

layer and devices. The cellular network layer contains

the following elements:

Cellular Tower has been added to the cellular net-

work layer with following properties: the Cartesian

coordinates; real life Geo-spatial information; list of

cellular antennas attached to the tower and keeps the

list of connected mobile devices’ references. Cellu-

lar Antenna is also known in the cellular network as

a transmitter. Antennas are entities attached to the

cellular towers. The model has the following proper-

ties: coverage as a polygon shape (e.g.hexagon, cus-

tom polygons are also supported); signal strength; fre-

quency, radius; reference to its tower and an id value,

which represents Common Gateway Identiﬁer (CGI)

value. Mobile Event Data is a data structure model to

describe a mobility event for a certain routable entity.

Concerning the devices added to the simulator can be

resumed to the GPS and mobile devices and they are

deﬁned as follows: GPS Device is, in the simulator

context, a device which can be attached to the routa-

ble entity like pedestrian, vehicle, bicycle etc. Mobile

Device is a device that generate the mobile event data,

attachable to the routable entity.

3.2 Core Mechanics of the Simulation

The core mechanics of the road trafﬁc simulation

SUMO has been gathered into one class which con-

tains the road network and orchestrates the simula-

tion. This class is initialized during the start of the

program and is given to the network builder as a re-

ference to create the infrastructure for the simula-

tion. During the road network building, the cellular

network is being loaded into the simulation network.

Then, when the simulation performer has all the vital

dependencies, then the simulation cycle starts. The

core of the simulation contains of multiple event con-

trollers, which are being ran after every tick. The si-

Introducing Cellular Network Layer into SUMO for Simulating Vehicular Mobile Devices’ Interactions in Urban Environment

583

mulation have been divided into four components ac-

cording to their event controller action types and in-

termediate smoothing and they are as follows:

Begin of time step events is the original event con-

troller in SUMO, which, as an example, takes care of

trafﬁc light events, trigger type events(lane speed, re-

routing), routing device, and pedestrian move events.

Intermediate actions, depends on the previous event

control cycle and smooths the network for the next

event control cycle. During this process, the system is

detecting regularly for collisions, checking the trafﬁc

lights, checking if the edges are active, plans the vehi-

cles movement, executing the vehicles movement, run

lane changes, load new routes and mobile events for

the cellular simulation, insert new vehicles to the net-

work, insert new events, etc. During the execution of

vehicles movement, there is a case when the vehicle

has arrived to the destination and, immediately after-

wards, the vehicle shall be removed from the simula-

tion. This function has been altered to have checks for

existing mobility devices (GPS, mobile) to run their

last events immediately and then invalidating those

events to stop them from recurring. This entire code

alternation prevents losing event data due the prema-

ture removal of the vehicle. End of time step events

is an event running cycle which runs the events depen-

dent on the updated location of the routable elements.

There are also step by step logger events, vehicular

devices speciﬁc events (Bluetooth, GPS, mobility ma-

nagement related state update) and vehicle ﬂow ca-

librator events. Mobility events is a newly added

event running cycle entirely for the purpose of mo-

bility event simulation. The events are created by the

mobile devices from the data, which has been atta-

ched to the vehicle after it has been loaded into the

simulation during the intermediate actions.

3.3 Data Import and Export

The simulator depends on the data which has been

prepared beforehand for the simulation and therefore

imported into the application through Extensible Mar-

kup Language (XML) ﬁles. SUMO has in its package

multiple programs which help to produce the network

and other simulation related data in XML document

format to run the simulation properly. For the cel-

lular networking layer, there were created multiple

XML data import handlers and their respective XML

Schema Deﬁnition (XSD) ﬁles. First, cellular net-

work related data like antennas and their respective

transmitters with coverage areas. Second, mobile

events data, which were described in the Section 3.1.

For the export there were designed output XML for-

mats for the GPS and mobile devices. When the spe-

ciﬁc routable entity ﬁnished the navigation in the net-

work then all its devices data were exported into the

resulting XML ﬁle.

4 SIMULATOR

IMPLEMENTATION

The implementation of cellular network log genera-

tor and it’s integration into SUMO base code was

done based on introducing two prerequisite applica-

tions: Mobility Event Simulation Generator (MES-

GEN) and Cell Coverage Area Generator (HEXA-

GEN).

4.1 MESGEN

Mobility Event Simulation Generator (MESGEN) is a

supplementary application required by the main simu-

lator application. MESGEN is responsible for genera-

ting the cellular events-time-line for each vehicle (cel-

lular device inside vehicle). SUMO simulator uses

the cellular events-time-line to trigger the events in

certain cell towers during the simulation when vehi-

cle traverses the corresponding coverage area. Se-

veral cellular activity proﬁles have been deﬁned and

are illustrated in Table 1. Each proﬁle is speciﬁed

by three attributes: frequencies of calls, SMS or and

data usage activities. The events generated by MES-

Table 1: User cellular activity proﬁles (counting amount of

events per hour). The numbers in the table correspond to

events frequency as follows: none (0), low (1), medium (2)

and high (3).

Proﬁle Calls SMS Data

Casual 1 1 1

Only Keep Alive 0 0 0

Business 3 1 1

Teenager 1 3 2

Student 1 1 2

Talkative 2 1 1

Media streamer 1 1 3

GEN are similar to those of cellular event data mo-

del (original model was simpliﬁed by removing not

used events). The MESGEN event types are illustra-

ted in Table 2. The simpliﬁed ﬂowchart of the cel-

lular mobility management entity is illustrated in Fi-

gure 2. For example MESGEN events SMS SEND and

SMS REC occur mimics the sequence of cellular net-

work equipment activities that result in delivering an

SMS message (send or receive). MESGEN genera-

tes a history of events for each vehicle participating

in simulation using the simulation life-time, vehicle

activity proﬁles, road network information and cell

VEHITS 2018 - 4th International Conference on Vehicle Technology and Intelligent Transport Systems

584

Table 2: MESGEN event type descriptions.

Message Type Description

SMS REC SMS Received

SMS SEND SMS Sent

CALL INIT Calling out

CALL REC Receiving a call

WEB COM Uses mobile data

Listing 1: The appearance of the MESGEN output ﬁle.

</mEvents>

</vehicle>

</vehicles>

coverage areas. The result events are associated to a

vehicle and have time speciﬁed as an offset of SUMO

simulation life-time. An example of MESGEN out-

put is illustrated in the Listing 1 having vehicles with

associated cellular events. The pseudo code in the Al-

gorithm 1 illustrates the algorithm used in MESGEN

generator.

Minimum length and maximum length of diffe-

rent events can be conﬁgured through constant varia-

bles. It is also possible to deﬁne time buffers between

occurring events. The values of low and medium fre-

quency events thresholds and random number seed

can also be set through the conﬁguration ﬁle.

4.2 HEXAGEN

HexagonGen is an accessory level script, written in

Python, to generate 120 °sectorized cells for the cel-

lular network behaviour simulation. It takes an in-

put of latitude and longitude coordinate values, size

(length of hexagons side), hexagon network width,

and height values. The given coordinate pair is the

center of the drawn network (tower position). The

result of the HexagonGen is an XML ﬁle which con-

tains the cell coverage elements and can be fed di-

rectly into SUMO.

5 MOBILITY BEHAVIOUR AND

MANAGEMENT

5.1 Microscopic Simulation Devices

SUMO provides basic interfaces to communicate with

the vehicle and abstract methods, which need to be

implemented to develop a working device.

GPS device (MSDevice

GPS class in the code) is a test

device, it was created with the objective to study

Algorithm 1: Mobile’s events generator.

Precondition: V ∈ N set of vehicles, P ∈ N

set of proﬁles, T ∈ N vehicle departure

times, l

call

maximal call duration, l

sms

maximal interval between sending SMS,

data

maximal web surﬁng duration, E set of event types as was speciﬁed in Table

1: function TRIGGEREVENT(e,t, l)

2:  Triggers event of type e ∈ E at t, lasting l sec.

3: end function

4: function RAND(n

)

5: n ∼ U([n

, n

])  random number from (n

, n

)

6: return n

7: end function

8: function MESGENVEHICLE(v,t

start

, t

end

)

9: n

calls

sms

data

← P

 mobile proﬁle of v

10: n

all

∑

11: t ← T

 departure time of v

12: while t

end

−t ≤ 0 do

13: n = RAND(0, n

all

)

14: if n ≤ n

calls

then

15: l = RAND(0, l

call

)  Call duration

16: if RAND(0, 1) > 0.5 then  Call direction

17: TRIGGEREVENT(CALL

INIT,t, l)

18: else

19: TRIGGEREVENT(CALL REC,t, l)

20: end if

21: t ← t +l

22: else if n ≤ n

calls

+ n

sms

then

23: if RAND(0, 1) > 0.5 then  SMS direction

24: TRIGGEREVENT(SMS SEND,t,0)

25: else

26: TRIGGEREVENT(SMS REC,t,0)

27: end if

28: t ← t + RAND(0, l

sms

)

29: else

30: l = RAND(0, l

data

)  Web Surﬁng Duration

31: TRIGGEREVENT(WEB COM,t, l)

32: t ← t +l

33: end if

34: end while

35: end function

36: function MESGENALL(t

start

, t

end

)

37: for all v ∈ V do MESGENVEHICLE(v,t

start

, t

end

)

38: end for

39: end function

SUMO simulation cycle for the upcoming mobility

device. MSDevice GPS has two subclasses called Rou-

teGPSInfo and GPSSignalUpdate. RouteGPSInfo is a

data structure that records geographical coordinates

and the timestamp. GPSSignalUpdate represents GPS

signal, it is an extension of SUMO’s Command class

(base SUMO microsim event class). During every si-

mulation step, if the vehicle has a device, the event is

triggered the vehicles movement.

Figure 1: The class diagram of MSDevice Mobile and its

sub-entities.

Introducing Cellular Network Layer into SUMO for Simulating Vehicular Mobile Devices’ Interactions in Urban Environment

585

Mobile device (MSDevice Mobile class in th code) is

emulating multiple functions, which in real GSM

stack are performed by several logical units. Next

section will explain how mobile device simulation is

performed. Figure 1 illustrates how mobile device

component is integrated into SUMO and how does it

interact with other modules.

5.2 Mobility Management

During the simulation the state of mobile device is

managed by MobilityManagement. The state is upda-

ted at the end of each simulation step event. Vehi-

cles are initialized based on the parameters ﬁle, which

speciﬁes what devices are equipped (mobile, GPS,

etc). Mobility management entity is only created in

case of vehicle being equipped with a mobile device.

MobilityManagementUpdate is an extension of SUMO

Command base abstract event class and is responsi-

ble for triggering the state update. The new state is

assigned based on previous one, the state transitions

are illustrated in the Figure 2. Mobility management

has a reference to the mobile device to call out mo-

bility related procedures from the lower layers. Then,

there are mobility management state related variables;

the T312 timer is for reminding a mobile device to

make periodic location updates after every user de-

ﬁned steps (by default 10 steps). Some state rela-

ted extra scenarios, deﬁne whether the mobile has

been turned on, the SIM card inserted, the IMSI at-

tached, and deﬁne whether the mobile has PLMN re-

lated information. In the Figure 2 a mobile mobi-

lity life cycle is described visually for better under-

standing. At the start of the mobility management

Figure 2: The simple work ﬂow of the mobility manage-

ment.

update, the creation of the current time step’s mo-

bility events for the simulation for this speciﬁc de-

vice takes place. During the mobility management

update event, there will be a call out to the mana-

gement cores update function which increments in-

stantly the periodic location update with respect to

the previous state that goes through state machine

to determine the next steps: a) If the mobility ma-

nagement state is classiﬁed as NULL the core up-

date will change mobile to the state as it was just

physically turned on; b) MM IDLE NORMAL SERVICE

starts mobile cell selection method on the de-

vice level.c) MM IDLE SEARCH FOR PLMN starts

mobile cell selection method, but since the de-

vice has never been connected to the network, it

also waits for the surrounding PLMN information;

d) MM IDLE NO CELL FOUND start mobile cell se-

lection, since there is currently no connectivity; e)

MM CONNECTION ACTIVE starts the cell selection

process for the handover purposes, if needed.After

cell selection, the work-ﬂow will move from the de-

vice back into the mobility management, where the

state of MM will be updated and location update ini-

tiated. If the state machine has done its work, then

there is a successive check to determine if the periodic

update is needed and, if it is, then it will be executed.

The rule with the periodic location update is that if a

mobile has not had any location updates during some

certain amount of simulation steps, then the mobility

management has to step in and remind the transmit-

ter that it still exists and has not been lost during the

commute.

5.3 Wave Propagation

During the cell selection, the mobile device has mul-

tiple challenges. It has to determine the nearest poly-

gons, determine in which one of them it resides, and

determine the signal strength from the base stations

transmitter. The signal strength has been envisioned

by the free space wave propagation model:

(d) =

(4π)

, (1)

Where P

(d) is the received power, P

is the transmit-

ter output power, G

is transmitter gain G

is antenna

gain, d is the distance between the MS and the to-

wer, and L is the system loss factor. (Nishith Tripathi,

2014). This latter indicates the power received by the

antenna under ideal conditions. in addition, the free

space model predicts the powers decay to be the ne-

gative square root of the distance. In our case, we

have not included the system loss factor. Besides, the

distance in the equation 1 is calculated between the

tower and the vehicle based on Haversine distance:

d = 2r Arcsin

sin



− φ



+ f (φ

, φ

)

f (φ

, φ

) = cos(φ

)cos(φ

)sin



− λ



(2)

VEHITS 2018 - 4th International Conference on Vehicle Technology and Intelligent Transport Systems

586

where φ is the latitude, λ the longitude and r is the

Earth’s radius in meters (6 371 000 meters).

During the signal strength calculation, we use

the conﬁguration supplied tower transmitter output

strength (dBm) and the transmitter’s frequency. From

the frequency, we calculate the wavelength. Wave-

length equals the speed of light divided by the trans-

mitter emitted frequency.

5.4 Location Update, Cell Selection and

Reselection, Handover

The procedures are key component of keeping track

of the mobility management level state, this can be

seen also in the Figure 2.

Cell selection is a mobile device level method for me-

asuring the current signal strength value of the serving

antenna and manage handover procedure. During the

(a) The example of R-

tree after all the shapes

have been inserted into

the tree.

(b) The R-tree visualized as a

diagram

Figure 3: Overview of R-tree process.

cell selection, the ﬁrst step is to ﬁnd surrounding cel-

lular network cells. Those cells (polygons) are struc-

tured into the R-tree entity during the creation of the

mobility tower control unit, where they were added

for a faster search, as illustrated in the Figures 3(a)

and 3(b), R-tree is a tree data structure to handle spa-

tial data efﬁciently by indexing shapes for the future

access. (Guttman, 1984) After searching the nearest

polygons to the device, we check through the vector

of cells and determine whether the vehicle is in the

cell or not using the winding number algorithm. The

vehicle might be in multiple cells; therefore, we are

calculating the best signal strength and accordingly,

we pick the transmitter and cell to camp on. In es-

sence, the winding number algorithm is an algorithm

which counts the number of times the polygon winds

around the point of interest. If the point is not inside

of the polygon, then the resulting winding number is

0. (Kai Hormann, 2001) During the cell selection, the

registration to the tower or the deregistration from the

tower will be determined. This is emulating the regis-

tering the mobile devices location area into the visitor

location register. The ﬁnal step is the status update of

the mobility management.

Location update is the mobility management level

procedure for updating the devices location in the net-

work. Location update resets the periodic update ti-

mer (sets the timer back to 0) and updates the mobi-

lity management state.

Handover procedure is triggered by the state machine

at the level of MM CONNECTION ACTIVE when there

is a better quality cell ﬂagged true.

5.5 Mobility Events and the Simulation

Cycle

We have covered the mobility management tied

event object called MobilityManagementUpdate. In this

section, we will discuss the mobile device related

event. MobilityEvent extends the SUMO Command ab-

stract event base class. The mobility events have

their own MSEventControl container-event queue and

the events are being executed at the very end of the si-

mulation step, after the mobility management update

events. Each mobility event has a type, which is de-

clared in the enum of MobilityEventType in the class of

MSMobilityEventData. The event has the mobile device

reference to let the mobile know about the radio link

failure and trigger the log creation about the occur-

red event. Other attributes of the mobility event are

the events start time and end time. Calls and the web

communication events depend on the timeframe vari-

ables.

5.6 Mobility CDR-like Logs

Call detail record (CDR) is the information about in-

coming and outgoing mobile activities, e.g., calls or

SMS messages. The data in the CDR is about the

event originated and the terminated parties (both sides

phone numbers), time of connection through the star-

ting time and the call duration, call event type, unique

generated id for the record, etc. (Horak, 2008) Mobi-

lityLog class is a data structure that, emulates the es-

sence of CDR. The attributes of the mobile logs are

illustrated in Table 3 (also visible in the Figure 1):

6 RESULTS

The results are illustrated through the outcome of the

logs generated by the simulator. The ﬁgure 4 is re-

ﬂecting the interactions between the mobile devices,

GPS devices and the cellular network.

6.1 GPS Logs

MSDevice GPS related logs can be created in two

ways: through main SUMO conﬁguration ﬁle or

Introducing Cellular Network Layer into SUMO for Simulating Vehicular Mobile Devices’ Interactions in Urban Environment

587

Table 3: Mobility Log Record Attributes.

Attribute Description

timestamp Amount of seconds since

simulation started

event type CALL INIT, CALL REC,

SMS SEND, SMS REC,

WEB COM, KEEP ALIVE

imsi subscriber ID

(vehicle ID)

cgi Common Gateway Identiﬁer (CGI)

the transmitter id

(or simply a cell id)

signal strength radio signal strength

during the event

latitude latitude coordinate

(vehicle actual location)

longitude longitude coordinate

(vehicle actual location)

speed vehicle speed

during the event (in m/s)

through SUMO route ﬁle. Listing 2 illustrates the

GPS signal log, it contains timestamp, longitude, and

latitude.

Listing 2: GPS coordinates (longitude and latitude) with a

timestamp.

<gps-output>

<gps-device id="gps_1">

</signal>

</signal>

</signal>

</vehicle>

</gps-output>

6.2 Cellular Network Simulation Logs

The MSDevice Mobile class generates logs in case the

corresponding mobile device has conﬁguration values

set to true, and the mobile data events have been ge-

nerated by MESGEN. In addition, the cellular network

coverage ﬁle (XML) should be speciﬁed. Otherwise

there would be no connectivity for the devices and the

events cannot occur.

The conﬁguration of the mobile devices is similar

to the GPS devices. One must deﬁne the mobile de-

vice into the route ﬁle or one can set all the vehicles

to carry a mobile device in the SUMO conﬁguration

ﬁle.

In the results, the cellular network behaviour si-

mulation written into XML ﬁle are illustrated in Lis-

ting 3. The logging entity is deﬁned in the Fi-

gure 1 (class MobilityLog) and is explained in the

Section 5.6.

Listing 3: CDR-like mobility logs.

<mobile-output>

<event timestamp="92.00" eventType="KEEP_ALIVE"

imsi="1" cgi="t-438-1" signalStregnth="-21.01"

latitude="58.29" longitude="26.46292491"

speed="13.68143750"/>

<event timestamp="94.00"

eventType="SMS_REC"

imsi="1" cgi="t-438-1" signalStregnth="-21.70533381"

latitude="58.29089506" longitude="26.46336419"

speed="12.64468285"/>

</vehicle>

<event timestamp="27.00" eventType="CALL_INIT"

imsi="27" cgi="t-852-1" signalStregnth="-21.11894073"

latitude="58.25022813" longitude="26.43376626"

speed="0.00000000"/>

<event timestamp="28.00" eventType="CALL_INIT"

imsi="27" cgi="t-852-1" signalStregnth="-21.21370266"

latitude="58.25021502" longitude="26.43379995"

speed="2.45816080"/>

</vehicle>

</mobile-output>

Figure 4: 5G standard proposes use of micro-cells, there-

fore an example of HexagonGen generated micro-cells in

SUMO.

7 CONCLUSION

In this article, we are introducing a new layer into

SUMO simulator. Our layer simulates the cellular

network behavior and generates the logs of the inte-

ractions between the mobile devices and the mobile

network. In addition, in order to make the integra-

tion with SUMO we created two main components:

Mobility Event Simulation generator and mobile cell

coverage generator. The results of our simulator is a

set of mobile logs similar to CDR data that reﬂects

the communication and the interactions between the

vehicular mobile devices in the SUMO’s vehicles and

the cellular network. In general, the outcome is very

encouraging and there is many enhancements that can

be added; especially with in regards to the mobile net-

work model for reﬂecting the mobile network proto-

cols and behavior.

VEHITS 2018 - 4th International Conference on Vehicle Technology and Intelligent Transport Systems

588

ACKNOWLEDGEMENT

This research work was supported by IUT34-4 ”Data

Science Methods and Applications” (DSMA) project.

REFERENCES

Bolla, R. and Davoli, F. (2000). Road trafﬁc estimation

from location tracking data in the mobile cellular net-

work. In Wireless Communications and Networking

Conference, 2000. WCNC. 2000 IEEE, volume 3, pa-

ges 1107–1112. IEEE.

Bulut, E. and Szymanski, B. K. (2013). Wiﬁ access point

deployment for efﬁcient mobile data ofﬂoading. ACM

SIGMOBILE Mobile Computing and Communicati-

ons Review, 17(1):71–78.

DLR (2017). DLR - Institute of Transportation Systems

- SUMO Simulation of Urban MObility. http://

www.dlr.de/ts/sumo. 2017-12-15.

Finkenzeller, K. and M

uller, D. (2010). Rﬁd handbook:

fundamentals and applications in contactless smart

cards, radio frequency identiﬁcation and near-ﬁeld

communication. UK: John Wiley & Sons, Ltd.

Guttman, A. (1984). R-trees: A dynamic index structure for

spatial searching, volume 14. ACM.

Hadachi, A., Batrashev, O., Lind, A., Singer, G., and Vai-

nikko, E. (2014). Cell phone subscribers mobility

prediction using enhanced markov chain algorithm.

In Intelligent Vehicles Symposium Proceedings, 2014

IEEE, pages 1049–1054. IEEE.

Hadachi, A., Mousset, S., and Bensrhair, A. (2013). Ap-

proach to estimate travel time using sparsely sam-

pled gps data in urban networks. Electronics Letters,

49(15):957–958.

Horak, R. (2008). Telecommunications and Data Commu-

nications Handbook, 2nd Edition. Wiley.

Kai Hormann, A. A. (2001). The point in polygon pro-

blem for arbitrary polygons. Computational Geome-

try, 20(3):131–144.

Karnadi, F. K., Mo, Z. H., and Lan, K.-c. (2007). Rapid

generation of realistic mobility models for vanet. In

Wireless communications and networking conference,

2007. WCNC 2007. IEEE, pages 2506–2511. IEEE.

Kolici, V., Oda, T., Spaho, E., Barolli, L., Ikeda, M., and

Uchida, K. (2015). Performance evaluation of a vanet

simulation system using ns-3 and sumo. In Advanced

Information Networking and Applications Workshops

(WAINA), 2015 IEEE 29th International Conference

on, pages 348–353. IEEE.

Krajzewicz, D., Erdmann, J., Behrisch, M., and Bieker,

L. (2012). Recent development and applications of

sumo-simulation of urban mobility. International

Journal On Advances in Systems and Measurements,

5(3&4):128–138.

Lind, A., Hadachi, A., and Batrashev, O. (2017). A new

approach for mobile positioning using the cdr data of

cellular networks. In Models and Technologies for In-

telligent Transportation Systems (MT-ITS), 2017 5th

IEEE International Conference on, pages 315–320.

IEEE.

Nishith Tripathi, J. H. R. (2014). Cellular Communications:

A Comprehensive and Practical Guide. Wiley. 1032

pages.

ns 2 (2017). The Network Simulator - ns-2. https://

www.isi.edu/nsnam/ns/. 2017-12-15.

ns 3 (2017). ns-3. https://www.nsnam.org/. 2017-12-15.

OMNeT++ (2017). OMNeT++ Discrete Event Simulator -

Home. https://www.omnetpp.org/. 2017-12-15.

OSM (2017). OpenStreetMap. https://

www.openstreetmap.org. 2017-12-15.

Pan, B., Zheng, Y., Wilkie, D., and Shahabi, C. (2013).

Crowd sensing of trafﬁc anomalies based on human

mobility and social media. In Proceedings of the

21st ACM SIGSPATIAL International Conference on

Advances in Geographic Information Systems, pages

344–353. ACM.

Piorkowski, M., Raya, M., Lugo, A. L., Papadimitratos, P.,

Grossglauser, M., and Hubaux, J.-P. (2008). Trans:

realistic joint trafﬁc and network simulator for vanets.

ACM SIGMOBILE mobile computing and communi-

cations review, 12(1):31–33.

Sommer, C., Yao, Z., German, R., and Dressler, F. (2008).

On the need for bidirectional coupling of road trafﬁc

microsimulation and network simulation. In Procee-

dings of the 1st ACM SIGMOBILE workshop on Mo-

bility models, pages 41–48. ACM.

Statista (2017). Number of mobile phone users worldwide

from 2013 to 2019. https:// www.statista.com/

statistics/274774/forecast-of-mobile-phone-users-

worldwide/. 2017-12-15.

Wedel, J. W., Sch

unemann, B., and Radusch, I. (2009).

V2x-based trafﬁc congestion recognition and avoi-

dance. In Pervasive Systems, Algorithms, and Net-

works (ISPAN), 2009 10th International Symposium

on, pages 637–641. IEEE.

Zaldivar, J., Calafate, C. T., Cano, J. C., and Manzoni,

P. (2011). Providing accident detection in vehi-

cular networks through obd-ii devices and android-

based smartphones. In Local Computer Networks

(LCN), 2011 IEEE 36th Conference on, pages 813–

819. IEEE.

Introducing Cellular Network Layer into SUMO for Simulating Vehicular Mobile Devices’ Interactions in Urban Environment

589