A Simulator for Drones and FANET Management Supporting

Multimedia Trafﬁc under Human Mobility

M. Tropea, F. De Rango and P. Fazio

Dimes Department, University of Calabria, via P. Bucci 39/c, 87036 Rende (CS), Italy

Keywords:

FANET, Coverage, Drones, Mobility Model, Footprint, Simulator.

Abstract:

In this paper a simulator for the management of a team of Unmanned Aerial Vehicles (UAVs) and drones has

been proposed. This new network is known as Fly Ad-Hoc Network (FANET), and it is a particular type of

Mobile Ad-Hoc Network (MANET) but with some speciﬁc aspects that allow to provide new services in future

generation networks. One of the possible applications is emergency situations or scenario where drones can

provide an additional or complementary access networks supporting web services and multimedia trafﬁc. In

this paper a simulator for FANET deploying has been proposed providing the possibility to simulate different

scenarios with different coverage areas. New coverage model has been included in the features and also

interesting human mobility model to support more realistic users mobility. Moreover, additional modules for

trafﬁc pattern generation have been implemented to create scenario where mobile users can activate multimedia

calls and trafﬁc on FANET. Some simulations have been led out to show how the simulator works.

1 INTRODUCTION

In this paper the capability of Unmanned Aerial Ve-

hicles (UAVs) of providing wireless connectivity to

users in a coverage area is analyzed. These devices

are able to form temporary communication network

in order to cover speciﬁc regions in many different

situations such as disaster and emergency events. In

particular, in this paper the task is that to provide a

study on a simulation environment created ad hoc for

the UAVs/drones coverage purpose. The proposed

simulator is realized in java and it is a platform able

to permit of simulating different scenarios and situa-

tions in which users need to communicate each other.

The simulator is able to consider a real map get by

Google on which insert a series of speciﬁc Points of

Interest (PoI) that represent the points in which users

move in order to reach the preﬁxed destination. The

simulator permits to set a series of parameters such

as users mobility, drone height and coverage radius,

different types and percentage of users in the map,

type of multimedia trafﬁc and its percentage in the

simulation. Moreover, the paper proposes two mod-

els, a human mobility model, that tries to provide a

way the users use for moving in the covered area, and

a footprint model, able to model the drone channel

and to provide the mathematical formulation for cal-

culating the correct height of the drone on the base

of the coverage radius. Moreover, in order to calcu-

late the available bandwidth for the drones in the path

followed by the users in their movements, a standard

Link State protocol has been considered in which the

topological changes are due to change in the parame-

ters status (bandwidth along the path) and the protocol

update are operated on periodical base. The results of

the simulative campaigns are shown in order to vali-

date the goodness of the proposed simulator software.

There are other important topics that are object of re-

search about Drones beyond the coverage providing

such as that regards the bandwidth management per-

forming mobility prediction of the users and the op-

portune admission call (Fazio et al., 2012; Fazio et al.,

2016; Fazio et al., 2017). The rest of this paper is

organized as follows: Section 2 presents the related

work on the considered research topic. In Section 3, a

description of the FANET simulator is given. In Sec-

tion 4, we describe the simulation environment cre-

ated in java. Section 5 shows the simulator implemen-

tation details. The numerical results are presented in

Section 6. Finally, Section 7 concludes the paper.

2 RELATED WORK

In the last years a lot of research has been focused

on the Unmanned Aerial Vehicles (UAVs) technology.

376

Tropea, M., De Rango, F. and Fazio, P.

A Simulator for Drones and FANET Management Supporting Multimedia Trafﬁc under Human Mobility.

DOI: 10.5220/0007979003760383

In Proceedings of the 9th International Conference on Simulation and Modeling Methodologies, Technologies and Applications (SIMULTECH 2019), pages 376-383

ISBN: 978-989-758-381-0

Many researchers have analyzed and studied many

different aspects of these new devices in order to show

their potentiality and capacity in many real contexts.

2.1 Simulator of Drones and UAVs

In this paragraph some of the literature works about

simulators proposed for UAVs/Drones devices under

different points of view and applicative domains will

be presented. The papers (De Rango et al., 2017a;

De Rango et al., 2017b) propose a simulator suit-

able for the agriculture domain in order to design

novel coordination and control techniques of a UAVs

team. The works presented many coordination tech-

niques both for monitoring the area and for coordinat-

ing the actions of the drones in the presence of par-

asites in order to analyze how the performance can

signiﬁcantly change if more constraints, such as en-

ergy, communication range, resource capacities, are

accounted. In (Bustamante et al., 2017) the authors

present the development of a virtual reality simula-

tor for the management of a UAV (Unmanned Aerial

Vehicle), focused on improving the quality of grown-

up people. This research has collected characteris-

tics of gestures and physical movements from users

made by other related research in order to study the

same interaction within a virtual world. In (Lom-

bardo et al., 2016) the authors adopt the popular ap-

proach of leveraging the X-Plane ﬂight simulator to

simulate the environment and aircraft dynamics. This

approach has found recent success in visualizing and

optimizing algorithms for basic UAV ﬂight control

as well as formation ﬂight. In (Zema et al., 2017)

the authors describe a novel simulation architecture

to implement distributed networked control systems.

Their proposal consists in the integration of already-

present and validated solutions into a compact pack-

age that features scalability and negligible architec-

tural delays.

2.2 Coverage Issues in Drones

In this paragraph some works about coverage issues

are shown. This typology of research is very studied

and analyzed because the FANET represents a tech-

nology able to improve and guarantee prompt help in

many real emergency situations. Then, many aspects

of the coverage and connectivity issues are object of

study. In (Al-Hourani et al., 2014) the authors pro-

vided a statistical generic Air-to-Ground RF propaga-

tion model for Low Altitude Platforms (LAPs), that

can substantially facilitate the planning efforts of air-

borne wireless services, since the RF planning can be

performed based on merely simple urban parameters,

rather than depending on site speciﬁc 3D-models that

are unlikely to be easily available and updated. In

(Park et al., 2015), an effective coverage decision al-

gorithm is proposed, which aims at providing seam-

less handover and thus establishing a fully connected

aerial network. They further calculate the seamless

handover success probability and the false handover

initiation probability to evaluate the proposed cover-

age decision algorithm. (Sae et al., 2016) studies the

coverage aspects of a Low Altitude Platform (LAP)

system that can form a temporary communication net-

work. The system consists of multiple autonomous

drones equipped with dual-band Wi-Fi access points

(APs) with ad hoc capabilities to form a mesh net-

work. The suitability of the LAP system is evaluated

from the coverage point of view with calculations and

simulations. In (Xie et al., 2017), the authors intend to

study the optimal deployment density of Drone Small

Cells (DSCs) to achieve maximum coverage consid-

ering the inter-cell interference considering both LoS

and NLoS links.

3 FANET SIMULATOR

The simulator to be implemented must allow to ana-

lyze the trend over the time of a network composed of

drones. The drones ﬂy over a city and communicate

with each other trying to create a network through an

appropriate messages exchange. During the ﬂight the

drones will have to be able to guarantee a connection

to the people who are within their coverage range and

who request the connection. The purpose of the sim-

ulations is therefore to analyze the network in terms

of bandwidth occupation, delays and lost packet. The

simulation area is a map that can be taken by google

maps platform in order to make the proposed environ-

ment extremely general and parametric. This permits

of considering whatever area in which we want an-

alyze the human movements on the basis of chosen

area characteristics while the users utilize their multi-

media applications such as Video and VoIP calls. The

chosen area has been represented by the class ”Grid”.

It contains all the information as: number of people in

the map; the number of clients classiﬁed by typology;

number of used drones; size of the ﬁeld; list of Points

of Interest where a person can stop.

While the main actors are represented by the

classes:

• Drone: could be deﬁned as the main class, as

guarantor of the functioning of each protocol that

is developed, such as the Link State, for the real-

ization of the network.

A Simulator for Drones and FANET Management Supporting Multimedia Trafﬁc under Human Mobility

377

Figure 1: Reference Scenario with drone coverage foot-

print.

• Client: a person must follow a speciﬁc model of

mobility, which we will see in the following, and

must be able to have the possibility of making au-

dio and video calls. He must also be able to send

information, such as his position, to the people

who request it.

As said so far, the paper focuses on the realiza-

tion of a simulator that allows to analyze a network

composed of drones, a FANET. For the realization of

this network we have relied on a classical graph struc-

ture, in which the nodes represent the Drones and the

Clients, and the links represent the connections be-

tween them. The graph is created considering a sparse

graph since the number of edge departing from a node

is independent of the number of nodes in the graph.

The network will be created by exchanging messages

between nodes that are at a distance allowing com-

munication. So the number of links varies according

to the number of nodes in the network and to their

distance. A key feature of our simulations is to keep

track of changes in link bandwidth, so, in addition to

information regarding the distance between nodes, an

edge stores the bandwidth parameter. Furthermore, an

edge is always identiﬁed by the couple of nodes that

it links. To manage the neighborhood information be-

tween nodes, a list is created for each vertex where

neighboring vertices will be stored.

3.1 FANET Creation and Management

The network is created dynamically by exchanging

messages between drones and also between drones

and clients. The main feature of the network is

that it must be able to create and manage itself au-

tonomously even when there are topological changes

or changes that affect the status of the links. For this

task, in this work, the FANET network uses one of the

most popular routing protocols, the Link State proto-

col. The method ”StartLinkState()” was implemented

in the Drone class to manage Link State operation.

This method is called up whenever a change of state

occurs in the network.

The following messages are sent for the creation

of the FANET:

• HELLO: it allows establishing the connection be-

tween drones and between drone and client;

• ACK: drone message of response to the received

Hello message;

• BYE: it allows the disconnection from the drone,

both for drones and for clients;

• LSA (Link State Advertisement): it contains the

Link State Table of the reference drone and will

be forwarded in ﬂooding to all the nodes of the

network.

• LSU (Link State Update): it is a Link Update (LU)

collector and allows network trafﬁc reduction.

• LU: it is not a proper packet of the standard Link

State operation. It is a message that the drone

sends itself whenever there is a band update on

one of its links. To reduce trafﬁc, these messages

are grouped in a single LSU message that will

be manage every 30 seconds. The drone keeps

track of the status changes of its links and at the

end it will update its LSA Database and generate

an LSA packet that will contain its own table of

neighbor nodes with updated bandwidth capabili-

ties.

All these packets inherit from the general class

”LinkStatePacket”, which contains the standard infor-

mation of each packet and it will be characterized by:

• the ID of the node that generates them;

• a sequence number that will be increased for each

generation;

• a numeric value to represent the Time To Live

(TTL) of the packet.

In the simulation process, each created drone will

have two tables that will continue to update during

the simulation: 1) LSA Table: it keeps track of neigh-

boring nodes. This table will be the one that is sent

from time to time for topological changes. In fact,

if the drone realizes that some neighboring node is

no longer connected, or notices a change in the sta-

tus of the links, then it will put the related changes

in this table and forward it to the neighboring nodes

via an LSA message. 2) LSA DB: it keeps track of

SIMULTECH 2019 - 9th International Conference on Simulation and Modeling Methodologies, Technologies and Applications

378

all the information on the network and will be cre-

ated and modiﬁed each time an LSA from the other

nodes on the network is received. When the simulator

starts, each drone must send a Hello message to the

nearby drones. The neighboring drones are all those

drones that are within a certain range of coverage, set

to 300 meters. The neighboring drones will respond

and, upon receiving the ACK, the drone that made the

Hello request will create a new line within own LSA

DB and own LSA Table.

Once all the ACK have been received, the drones

generate a LSA packet which will be forwarded to all

the neighboring nodes, which process the data in the

table contained in the packet and update their LSA

Database. They will then forward the received LSA

packet to all the neighbors, except to the node from

which they received it. Obviously, in this way, it will

be possible to obtain equal LSAs from several drones

and, in fact, a check on the sequence number of the re-

ceived LSA packet is always performed. Each drone

stores a list consisting of couple of values that will in-

dicate: the ID of the nodes from which an LSA was

received; the highest sequence number relative to the

LSAs received from that speciﬁc node. If an LSA

is received with a sequence number lower than that

stored in the list then it will be discarded and it will

not managed. The LSA packets will then be sent in

ﬂooding in the network whenever there is a topologi-

cal change, as can be the connection/disconnection of

a new node and in the simulator this happens when-

ever a Client connects or disconnects from the net-

work.

3.2 Human Mobility: Mobility Model in

Urban Space

There are a lot of studies on this subject and several

interesting models for our case, including a study by

the University of Milan (Jahromi, 2017). They ana-

lyzed data sets on smartphone call information in Mi-

lan for 67 days, detecting over 69 million phone calls

and over 20 million text messages. Later, they also

analyzed the WiFi datasets and GPS data with which

they found the movement of 178 people within 4 years

(between 2007 and 2011).

After the analysis of the data, they have obtained

a model of people mobility based on the classiﬁca-

tion in 3 categories of Points of Interest (PoI) that are

visited daily:

• Mostly Visited PoI (MVP): the places most visited

by a user that may be their workplace, or their

home.

• Occasionally Visited PoI (OVP): the places of in-

terest of a user, but occasionally visited. Often it

corresponds to favorite places or meeting points

visited during the week.

• Exceptionally Visited PoI (EVP): points of interest

rarely visited.

On this type of model we developed our idea of

mobility which includes 3 types of users:

• Worker: who represents the classic type of person

who spends most of the day at work. The students

can ﬁt into this category.

• Housewife: who represents the type of person

who spends most of the time, especially in the

morning, at home.

• Pensioner: who represents the type of person who

is free to move.

Each type of user will spend the time in the day in

a different way and will visit most likely the points of

major interest, also based on time slots, so we tried to

represent this mobility in the simulator by considering

the following models.

3.3 Footprint Model

For that concerns air to ground channel model in the

considered urban scenario, the ground receiver can

receive two main signals: one is the Line of Sight

(LoS) signal and the other one is the Non-Line of

Sight (NLos) signal (Al-Hourani et al., 2014). These

two typologies of signal can be considered as sepa-

rately conditions and, then, they are studied consider-

ing different probabilities of occurrence. These two

propagation conditions can be considered separately

assuming they have different probabilities of occur-

rence. Typical, as discussed in (Feng et al., 2006), it

can be assumed that the signal received by the desti-

nation can be considered only in one of the two pre-

vious typologies. Each typology has a speciﬁc prob-

ability of occurrence which is a function of environ-

ment, density and height of buildings and elevation

angle. The probability of receiving LoS and strong

NLoS components are signiﬁcantly higher than fad-

ing (Feng et al., 2006). Therefore, the impact of small

scale fading can be neglected. A common approach

to model air to ground propagation channel is to con-

sider LoS and NLoS components along with their oc-

currence probabilities separately. Note that for NLoS

connections due to the shadowing effect and reﬂection

of signals from obstacles, Path Loss (PL) is higher

than LoS. Hence, in addition to the free space prop-

agation loss, different excessive path loss values are

assigned to LoS and NLoS links. As it is possible

to view in ﬁgure 2 a coverage area is represented in

which h is the height of the drone and r is the radius of

A Simulator for Drones and FANET Management Supporting Multimedia Trafﬁc under Human Mobility

379

the coverage, d is the distance between an user on the

edge of the coverage area and the drones in the sky. d

is given by the following formula: d =

√

+ h

, in-

stead the angle θ, in radiant, between r and d is given

by: θ = tan

−1

(h/r).

Figure 2: UAV Footprint.

On the basis of this consideration, in the following

the formulas respectively of PL for the LoS and NLoS

conditions as in (Al-Hourani et al., 2014):

LoS

(dB) = 20log(4 f

d/c) + ξ

LoS

(1)

NLoS

(dB) = 20log(4 f

d/c) + ξ

NLoS

(2)

where PL

LoS

and PL

NLoS

are the average (PL) for

LoS and NLoS links, ξ

LoS

and ξ

NLoS

are the aver-

age additional loss to the free space propagation loss

which depend on the environment, c represents the

speed of light, f

the carrier frequency. The probabil-

ity P

of having LoS connections at an elevation angle

of θ is given by (Zheng et al., 2013)

(LoS) =

1 + α·exp(−β[(180/π)θ −α)]

(3)

where α and β are constant values which depend

on the environment (rural, urban, dense urban, etc.).

The NLoS probability is P

(NLoS) = 1 −P

(LoS).

The formula represented in equation 3 indicates that

the probability of having LoS connection between the

drone in the sky as Access Point (AP) and the users in

the coverage area is an increasing function of θ. This

means that by increasing the elevation angle θ, the

shadowing effect decreases and clear LoS path exists

with high probability. Finally, the average PL as a

function of the altitude h and coverage radius r be-

comes:

PL(r,h) = P

(LoS) ·PL

LoS

+ P

(NLoS) ·PL

NLoS

. (4)

On the base of the previous drone channel model

we want to provide a formula for computing the opti-

mal altitude h for the maximum ground coverage area

Table 1: Considered parameters in mathematical formula-

tion.

Parameter Value

2 GHz

LoS

1 dB

NLoS

20 dB

α 9.6

β 0.28

10dB

of radius r. Consider a drone in the sky able to trans-

mit a signal of power P

tran

, then the received power

rec

can be represented by the following formula:

rec

(dB) = P

tran

−PL(r, h). (5)

Once the received power P

rec

is calculated, an user

in the coverage area of the drone at an height of h is

able to receive the signal if its Signal to Noise Ra-

tio (SNR) is greater than a ﬁxed threshold (γ

). This

means that:

γ(r,h) =

rec

> γ

(6)

where N is the noise power. From the previous

formula, it is possible to assert that to ﬁnd the max-

imum achievable coverage radius we should have:

γ(r,h) = γ

. For a ﬁxed transmit power, the optimal

drone height which results in maximum coverage is

computed by solving the following equation as it is

possible to view in (Feng et al., 2006):

180(ξ

NLoS

−ξ

LoS

)βZ

π(Z + 1)

−

20µ

log(10)

= 0 (7)

where Z = α · exp([(180/π)tan

−1

(µ) − α]) and

µ = h/r. By solving equation 7, µ

opt

= h

opt

max

is computed and using equation 5, h

opt

and r

max

are

found.

4 SIMULATION ENVIRONMENT

In order to allow the movement of people, an urban

scenario was created taking as an example the google

map of the center of the city of Cosenza, in the south-

ern of Italy. Some Points of Interest (PoI) have been

chosen, such as shops, restaurants, hotels, churches,

etc. and then people’s homes: a grid of 144 houses

was chosen to cover most of the map. Subsequently,

the people movement has been realized. Each person

will start its day at 00:00 and will stay at home until 8

am. Later, based on the proposed model, people will

move throughout the day. In the simulator it is possi-

ble to set, in addition to the number of people, the

percentage of workers, housewives and pensioners.

The movements of the people are determined by the

SIMULTECH 2019 - 9th International Conference on Simulation and Modeling Methodologies, Technologies and Applications

380

”start()” method, which based on the time slots and

the various probabilities determines the next action

of the person and, in particular, the next destination.

Once the next destination is determined, each person

will try to reach it by following a path. The proposed

algorithm for ﬁnding the best route consists in search-

ing each time for a next PoI that can be reached within

a radius of 90 meters and that represents the PoI closer

to the destination. To better understand the reasoning,

it is possible to take the following ﬁgure, ﬁgure 3, as

an example: a person moves from point S to point D.

Within his range of action (where there are points 1,

2 and 3, as it is possible to view in ﬁgure) the point

2 satisﬁes the conditions of the algorithm, that is, it

represents the point reachable from S within a radius

of 90 meters and closer to the destination.

Figure 3: Example of search for the next PoI to reach the

destination.

A grid arrangement was then chosen, therefore, to

avoid the problem of not ﬁnding another PoI within

the radius of action that would allow the movement.

A person performs his movements towards the chosen

destination, moving at a speed of 3 km/h. There is a

procedure that is the heart of the whole movement as

it recursively, if after a movement the destination has

not been reached, recalculates the next PoI generating

again the events for the movement. Usually, a per-

son who is not working or not at home can stay in

a PoI with a time varying between 10 and 60 minutes

each time. All activities end at 22:00, the limit beyond

which the day of the people is considered concluded.

People will return home and this determines the end

of the simulation.

Figure 4: View of the PoI visited by a person during a day.

5 SIMULATOR

IMPLEMENTATION

We have implemented a simulator in order to analyze

various aspects of the FANET. In particular, the goal

is to control how the network behaves during a day us-

ing a Link State Standard approach in which network

drones will send an LSA to neighboring nodes when-

ever a topological change occurs. We have considered

as simulation area a portion of the city of Cosenza (in

the southern of Italy). In particular, we have consid-

ered the surrounding area of ”Corso Mazzini” (one of

the most important street of the city) where the peo-

ple who move in this area are able to make audio or

video calls. In our work we have considered three dif-

ferent video and audio calls distributions as shown in

the following:

• 30% of audio streams and 70% of video streams;

• 50% of audio streams and 50% of video streams;

• 70% of audio streams and 30% of video streams.

Simulations were carried out considering the num-

ber of users in the area of interest in increasing. In

addition to the variable parameters, different ﬁxed pa-

rameters have been set for the simulations: we have

set the drones without problems of autonomy; each

drone will have available 10 Mbps of bandwidth ded-

icated to the call admission of users within its range

of coverage; the range of coverage has been set with

a radius of 175m and the optimal height of the drone

has been obtained equal to about 120 meters from the

ground (Hayajneh et al., 2016; Mozaffari et al., 2015);

all drones are equipped with a buffer that stores video

and audio packets before forwarding them to the net-

work. The size of this buffer has been ﬁxed to a

maximum of 50 packets; the communication range

between the drones has been set to 300 meters; the

movement speed of people has been set at 3 Km/h;

a person can move from one PoI to another within

a radius of 90m; 166 PoIs were chosen: 144 houses

placed as checkerboard and 22 other PoIs including

shops, refreshment points and entertainment; each

person will be able to make a call each time it is in a

PoI. Considering the proposed mobility model, it can

therefore be noted that, with 70% of workers spend-

ing most of their time at work, the ﬂow of calls in the

network will be quite limited; to have a more realistic

case we have chosen to have users divided as follows:

70% of workers; 20% of housewives; 10% of pen-

sioners. For all the simulations, a network topology

consisting of 9 drones, distant 252 meter each other,

in rows of 3 has been considered, able to cover most

of the area of interest. The resultant network will be

so composed as shown in ﬁg 1.

A Simulator for Drones and FANET Management Supporting Multimedia Trafﬁc under Human Mobility

381

6 SIMULATION RESULTS

A series of simulations have been started with a grow-

ing number of people. The network was analyzed

from a minimum of 50 people to a maximum of 150.

For each simulation the considered parameters have

been:

• jitter of audio and video streams;

• delay of audio and video streams;

• number of packets sent, received and lost;

• bandwidth requests accepted and refused;

• number of calls made;

• percentage of bandwidth occupied for each drone

for that concerns the admission of calls.

The simulation results show the trend of some pa-

rameters such as sent and received packets, accepted

and refused requests of bandwidth, delay of the two

typologies of streaming. These parameters allow to

evaluate the goodness of the proposed simulator in

which we have varied some ingress parameters. In

particular, in ﬁgures 5 it is possible to view the num-

ber of sent and received packets while the number

of accepted bandwidth requests, calls done and re-

fused requests are shown in ﬁgure 6. The ﬁgure 5

shows how the number of packets increases varying

the number of people and how the number of lost

packets is very low. From the ﬁgure 6 it can be seen

how the number of bandwidth requests increases with

the number of people and how, therefore, the num-

ber of calls made and the number of bandwidth re-

quests refused will increase. In addition, in the ﬁgure

7, we see which drones have a greater demand load

and, therefore, a higher percentage of occupied band-

width. The drones 6 and 8 are those with the highest

occupancy rate, due to a greater concentration of PoIs

in the area covered by them. Furthermore, it can be

seen how the percentage of bandwidth increases for

an increasing number of people. In the ﬁgures 8 and

9 it is possible to view the delays in both multime-

dia considered trafﬁcs. Figure 8 shows the delay in

user communications that consider VoIP multimedia

trafﬁc. Finally, ﬁgure 9 shows the delays, minimum,

medium and maximum for video trafﬁc. The ﬁgures

show delays in line with the typical delay of these ty-

pologies of multimedia trafﬁcs.

7 CONCLUSIONS

In this paper we present the implementation of a

UAVs/drones simulator environment created in java

and useful for analyzing different aspects of the so

Figure 5: Sent and Received packets vs number of users.

Figure 6: Accepted bandwidth requestes, Calls done, re-

fused requestes vs number of users.

called FANET. The possibility of providing tempo-

rary communication network composed of these new

kinds of devices is a very important research topic of

these last years. This aspect is very important in many

cases, ﬁrst of all in emergency situations when it is

important to guarantee efﬁcient communications in

the considered area. In this paper we propose a simu-

lation environment for UAVs coverage. The proposed

simulator is created in order to provide a platform

able to permit of simulating different scenarios and

cases in which users need to communicate each other.

The performance evaluation of our proposed simula-

tor has been considered using multimedia trafﬁc such

as video and audio streaming as typology of users’ ap-

plications. We have simulated a team of drones able

to cover a specif area where the considered footprint

Figure 7: Number of call admission vs drone for different

number of users.

SIMULTECH 2019 - 9th International Conference on Simulation and Modeling Methodologies, Technologies and Applications

382

Figure 8: Audio stream delay (min, medium, max) vs num-

ber of users.

Figure 9: Video stream delay (min, medium, max) vs num-

ber of users.

and human mobility model have been taken in consid-

eration for carried out simulation results.

REFERENCES

Al-Hourani, A., Kandeepan, S., and Jamalipour, A. (2014).

Modeling air-to-ground path loss for low altitude plat-

forms in urban environments. In 2014 IEEE Global

Communications Conference, pages 2898–2904.

Bustamante, A., Guerrero, G., Rodrigues, N., and Pereira,

A. (2017). Uav simulator for grown-up people quality

of life enhancement. In 2017 12th Iberian Conference

on Information Systems and Technologies (CISTI),

pages 1–6. IEEE.

De Rango, F., Palmieri, N., Santamaria, A. F., and Potrino,

G. (2017a). A simulator for uavs management in agri-

culture domain. In 2017 International Symposium on

Performance Evaluation of Computer and Telecom-

munication Systems (SPECTS), pages 1–8. IEEE.

De Rango, F., Palmieri, N., Tropea, M., and Potrino, G.

(2017b). Uavs team and its application in agriculture:

A simulation environment. SIMULTECH, 2017:374–

379.

Fazio, P., De Rango, F., and Tropea, M. (2017). Pre-

diction and qos enhancement in new generation cel-

lular networks with mobile hosts: A survey on dif-

ferent protocols and conventional/unconventional ap-

proaches. IEEE Communications Surveys & Tutorials,

19(3):1822–1841.

Fazio, P., Tropea, M., De Rango, F., and Voznak, M. (2016).

Pattern prediction and passive bandwidth manage-

ment for hand-over optimization in qos cellular net-

works with vehicular mobility. IEEE Transactions on

Mobile Computing, 15(11):2809–2824.

Fazio, P., Tropea, M., Veltri, F., and Marano, S. (2012). A

novel rate adaptation scheme for dynamic bandwidth

management in wireless networks. In 2012 IEEE

75th Vehicular Technology Conference (VTC Spring),

pages 1–5.

Feng, Q., McGeehan, J., Tameh, E. K., and Nix, A. R.

(2006). Path loss models for air-to-ground radio chan-

nels in urban environments. In 2006 IEEE 63rd Vehic-

ular Technology Conference, pages 2901–2905.

Hayajneh, A. M., Zaidi, S. A. R., McLernon, D. C., and

Ghogho, M. (2016). Optimal dimensioning and per-

formance analysis of drone-based wireless communi-

cations. In 2016 IEEE Globecom Workshops (GC Wk-

shps), pages 1–6. IEEE.

Jahromi, K. K. (2016/2017). Human mobility in urban

space. Doctoral Dissertation, Universitas Studiorum

Mediolanensis.

Lombardo, C., Miller, I., and Wallace, J. (2016). Studying

the interaction of uas and human pilots using the x-

plane ﬂight simulator. In 2016 International Confer-

ence on Unmanned Aircraft Systems (ICUAS), pages

557–561. IEEE.

Mozaffari, M., Saad, W., Bennis, M., and Debbah, M.

(2015). Drone small cells in the clouds: Design, de-

ployment and performance analysis. In 2015 IEEE

Global Communications Conference (GLOBECOM),

pages 1–6. IEEE.

Park, K., Cho, B., Park, K., and Kim, H. (2015). Opti-

mal coverage control for net-drone handover. In 2015

Seventh International Conference on Ubiquitous and

Future Networks, pages 97–99.

Sae, J., Yunas, S. F., and Lempiainen, J. (2016). Cover-

age aspects of temporary lap network. In 2016 12th

Annual Conference on Wireless On-demand Network

Systems and Services (WONS), pages 1–4.

Xie, J., Dong, C., Li, A., Wang, H., and Wang, W. (2017).

Optimal deployment density for maximum coverage

of drone small cells. In 2017 IEEE 86th Vehicular

Technology Conference (VTC-Fall), pages 1–6.

Zema, N. R., Trotta, A., Sanahuja, G., Natalizio, E.,

Di Felice, M., and Bononi, L. (2017). Cuscus:

Communications-control distributed simulator. In

2017 14th IEEE Annual Consumer Communications

& Networking Conference (CCNC), pages 601–602.

Zheng, Y., Wang, Y., and Meng, F. (2013). Modeling

and simulation of pathloss and fading for air-ground

link of haps within a network simulator. In 2013 In-

ternational Conference on Cyber-Enabled Distributed

Computing and Knowledge Discovery, pages 421–

426.

A Simulator for Drones and FANET Management Supporting Multimedia Trafﬁc under Human Mobility

383