IBSC System for Victims Management in Emergency Scenarios

Alexandra Rivero-Garc

ıa, Iv

an Santos-Gonz

alez, Candelaria Hern

andez-Goya and Pino Caballero-Gil

Departamento de Ingenier

ıa Inform

atica y de Sistemas, Universidad de La Laguna, Tenerife, Spain

Keywords:

Identity-based Signcryption, Keyed-Hash Message Authentication Code, Security, Triage, Emergency.

Abstract:

This work describes an optimized system designed to help the greatest number of injured people in emergency

situations, using the shortest possible time and cost. It is composed of a mobile application (assigned to

medical staff and helpers), a web service and Near Field Communication wristbands assigned to victims.

The mobile application is devoted to providing medical staff with the geolocation of victims as well as with

an assistant indicating the best route to follow in order to take care of them based on the severity of their

conditions and based on a triage method. Resolution of the routes is solved based on a classical problem, a

Travelling Salesman Problem, using a k-parition algorithm to divide the huge number of victims in different

clusters. Thus, each doctor has a speciﬁc area to assist victims. Besides, doctors can use a functionality

of the application to contact their peers through a video call when additional help is needed. The proposal

combines an keyed-Hash Message Authentication Code scheme to protect Near Field Communication tags and

an IDentity-Based Cryptosystem to the wireless communication. Speciﬁcally an IDentity-Based Signcryption

is used for communication conﬁdentiality, authenticity and integrity, both among peers, and between server

and medical staff.

1 INTRODUCTION

The communication technologies used in smartpho-

nes and the power of these devices can help in many

complex scenarios. Smartphones are used to support

different daily tasks, their small size and high perfor-

mance is a huge advantage. This paper presents a plat-

form for improving logistics of medical staff in emer-

gency situations in a distributed way. In particular, it

is based on data obtained from a triage application de-

veloped in (Rivero-Garcıa et al., 2014), where a mo-

bile system for victim classiﬁcation in emergency si-

tuations was implemented.

The deﬁnition of triage can be described as fol-

lows. A simple, complete, objective and fast pro-

cess to obtain an initial clinical assessment of people

with the objective of evaluating their immediate sur-

vival capacities and prioritizing them according their

severity is a triage. In order to achieve the classiﬁ-

cation, all triage systems distinguish two steps. The

ﬁrst triage or simple triage is used for the generation

of a classiﬁcation based on the severity of injuries of

the victims evaluating their survival skills in some se-

conds. The second triage is where medical staff ana-

lyses each patient’s state: bruises, wounds and inju-

ries. Speciﬁcally, in this work, Simple Triage and Ra-

pid Treatment Algorithm (START) method is used as

ﬁrst triage. Its output is the victim’s classiﬁcation ba-

sed on coloured tags, where each colour deﬁnes the

priority of the victim: black, dead or irrecoverable

victims; red, victims requiring immediate care; yel-

low, victims requiring urgent care but who can wait

for treatment from half an hour to one hour; green,

victims who are not seriously injured. They can wait

for treatment more than an hour. Here the use of Near

Field Communication (NFC) is proposed to deal with

the triage result. NFC stickers are used to save triage

results based on the generation of a keyed-Hash Mes-

sage Authentication Code (HMAC) scheme. Further-

more, the route to attend victims for each doctor is

shown through a map in their smartphones based on

the priorities of victims and they can share informa-

tion peer-to-peer with their colleagues in the affected

area. All these communications are protected through

an IDentity-based (ID-based) cryptography, speciﬁ-

cally a ID-Based Signcryption scheme (IBSC).

This work is organized as follows. Section 2 pro-

vides some preliminaries while Section 3 gives a glo-

bal view of the proposal. Then, Section 4 sketches

the system that is used to make decisions. The topic

of victim identiﬁcation through NFC tags and HMAC

schemes is dealt in Section 5. The protection of se-

curity related to the medical staff through an IBSC

scheme is proposed in Section 6. A brief security ana-

276

Rivero-García, A., Santos-González, I., Hernández-Goya, C. and Caballero-Gil, P.

IBSC System for Victims Management in Emergency Scenarios.

DOI: 10.5220/0006298702760283

In Proceedings of the 2nd International Conference on Internet of Things, Big Data and Security (IoTBDS 2017), pages 276-283

ISBN: 978-989-758-245-5

lysis is provided in Section 7. Finally, a few conclusi-

ons and future works close the paper.

2 PRELIMINARIES

There are still some weaknesses in emergencies ma-

nagement. The integration of new technologies into

emergency situations management and medical care

has allowed the development of tools that help to the

coordination between medical staff in emergency sce-

narios. There are different proposals designed to help

to ﬁnd missing persons after a large-scale disaster.

Such as People Locator and ReUnite (of Medicine at

NIH, 2017), Google person ﬁnder (Google, 2017) and

Safety Check of Facebook (Facebook, 2017). All

these systems try to verify and share the status of pe-

ople after some disaster, speciﬁcally the proposal of

Facebook share all the information with the victim’s

friends in this social network.

Some organizations are working to provide diffe-

rent solutions related to emergency situations. One of

then is Sahana foundation (Foundation, 2017) project

aims to provide a set of modular, web-based disas-

ter management applications. This project includes

tools for synchronization between multiple instances:

a Missing Person Registry, Request and Pledge Ma-

nagement System and Volunteer coordination. Since

this proposal is a web-based framework, it has the

problem of relying on communication to the centra-

lized web-server, and thus cannot take advantage of

mobile nodes. There are no solutions to the identi-

ﬁcation of victims. The unique identiﬁcation of af-

fected people is a requirement for any emergency

triage. Barcodes are a possibility because they fa-

cilitate mechanical reading (Neuenschwander et al.,

2003). Barcodes are cheap and easy to create, they

can be generated just using a standard printer. But in

an emergency situation having a printer in the affected

zone is not realistic. Radio Frequency IDentiﬁcation

(RFID) is a very useful technology for victim identi-

ﬁcation as it is explained in (Inoue et al., 2006) and

in (Baracoda, 2017). Two types of tags exists, pas-

sive tags, that use the energy received from the reader

to send the identiﬁer, and active tags, that include a

battery to increase its distance range. The problem of

this kind of communication is that a RFID reader is

needed and no security tools were provided. In (Gao

et al., 2007) a speciﬁc triage tag technology is pro-

posed. These electronic triage tags use noninvasive

biomedical sensors to continuously monitor the vital

signs of a patient and deliver pertinent information to

ﬁrst responders. These are not triage tag for emer-

gency situations.

The use of NFC (Near Field Communica-

tion)(Want, 2011a) is one of the bases of the propo-

sed system, speciﬁcally NFC stickers, for automatic

patient identiﬁcation. Unlike other technologies as

RFID (Zou et al., 2014), Bluetooth or Wi-Fi (Lee

et al., 2007), NFC is not oriented to the continuous

data transmission. It is necessary a temporally con-

tact between the devices that interact to allow the ex-

change of information in a quick and timely way. Alt-

hough, at ﬁrst, the distance factor for transmitting in-

formation may seem a limitation it is actually the key

in this technology. The need for proximity between

devices limits the types of attacks to develop. Besi-

des, not requiring pairing between devices facilitates

its use by health staff. NFC devices may operate in

two different modes. On the one hand, in the active

mode each device generates its own electromagnetic

ﬁeld (emulating the communication paradigm peer-

to-peer). On the other hand, in the passive mode

one device generates the electromagnetic ﬁeld with its

own power supply. In this way, it enables that other

device starts the connection taking energy from the

ﬁeld generated to power its circuit. Then the passive

device generates the response signal and transfers the

data. This mode of operation matches with the RFID

communication model and it is the one used in this

proposal.

3 GLOBAL VIEW

The main objective of the proposal is to generate a

tool to save as many time as possible in emergency

situations. Therefore, doctors have a map in their mo-

bile phones that helps them in every moment to de-

cide the route to patients. This route is based on the

severity of the injuries. Thus, collisions of doctors to

assist the same patient are avoided and decisions are

taken based on priority.

Two stages in the route generation are made. The

ﬁrst one consists in the evaluation of the affected area

applying START triage method to obtain a victims’

classiﬁcation based on coloured tags. This is genera-

ted by the ﬁrst aid team, where there are medical staff,

ﬁreﬁghters or even rescue services.

As we mentioned previously, each colour deﬁnes

the priority of the victim: black, dead or irrecoverable

victims; red, victims requiring immediate care; yel-

low, victims requiring urgent care but who can wait

for treatment from half an hour to one hour; green,

victims who are not seriously injured. They can wait

for treatment more than an hour. This colour result is

stored on tags. In this case NFC (Want, 2011b) tags,

speciﬁcally NFC stickers are used to save the triage

IBSC System for Victims Management in Emergency Scenarios

277

result. Note the proposed work uses NFC stickers but

multiple kind of NFC tags can be used, depending on

the emergency and the victims state. Each triage has

a location in the central server. At the end of this step

the system has a map with the location of each victim

and their triage like in ﬁgure1.

Figure 1: Victims’ location.

The second stage is based on the victim’s attention

taking into account the results of the triage priorities.

In this step the victims’ locations given by the ﬁrst

triage is essential being the starting point. A graph of

each colour is generated based on the victim’s loca-

tion in a map. Victims represent nodes and the rou-

tes to reach them are de edges. Each edge has a cost.

This cost is the distance between two nodes calculated

through the Haversine Formula (Knox, 2015), where

cos

AB∆λ

= cos(γ

) · cos(γ

) · hvsin(∆λ), then:

hvsin(

) = hvsin(γ

− γ

) + cos

AB∆λ

(1)

Where hvsin is the haversine function:

hvsin(θ) = sen

(

) =

1 − cos(θ)

(2)

d is the distance between two points (over the big-

ger circle of the sphere), R is the sphere radio, in our

case the Earths radio, γ

is the latitude of the point A,

is the latitude of the point B and ∆λ is the diffe-

rence of the longitudes.

Finally, if sen

= sen(γ

) · sen(γ

) and cos

cos(γ

) · cos(γ

), the distance (d) is:

d(A, B) = R ∗ arccos(sen

+ cos

∗ cos(∆λ)) (3)

All information related to the patients who must

be attended by a doctor is done through a mobile ap-

plication. It indicates to the medical staff through a

map his/her current location and the next patient to

assist.

The application has enabled a feature called

“emergency support”. With this function when a doc-

tor or nurse requires additional help from peers he/she

can activate this mode. When they activate this fea-

ture all health personnel in the affected area receives

the notiﬁcation and simply by clicking on it, they can

start a video call or a chat to help his/her colleague.

This functionality was designed to support healthcare

workers and improve the use of time in transfers be-

tween patients. Note that this feature opens a com-

munication channel between two partners through a

video streaming. Due to the high amount of infor-

mation exchanged the connection will take place by

Long Term Evolution (LTE)(Sesia et al., 2009), spe-

ciﬁcally LTE-Direct to ensure adequate and secure

communication between nodes that connect.

In the moment in which a doctor has just treat

a victim, he/she can take his/her mobile and read

the tag, mark the point as completed and check next

victim status. When the doctor arrives to the location

of the new victim, the node is automatically marked

on the map as being in the care process but he/she can

read the sticker to be sure of the authenticity of the

node. The period devoted to reach a new node is cal-

led “travelling time”. Doctors can receive notiﬁcati-

ons called “emergency support” when they are in this

“travelling time” to avoid constant notiﬁcations that

may mislead the staff in the middle of an assistance.

4 DECISION-MAKING SYSTEM

First of all, in the generation of doctors’ routes, an un-

directed graph is created from the points deﬁned du-

ring the triage. There are as many points as patients

on the map, these are the vertices of our graph. The

edges will be deﬁned undirected between the vertices.

This distance between points will be the cost of the

edge. The system generates one graph for each triage

colour, the main objective is treating patients based on

their injuries. First of all, the patients with red triage

are care, then patients with yellow triage and ﬁnally

the ones with green triage. Once each graph is genera-

ted, the amount of resources and the place where they

are needed. In this case resources are doctors (number

of doctors #d) that will assist patients. Their position

at all time is known. Speciﬁcally a graph based on

the Delaunay Triangulation (de Berg et al., 2008) is

created (as in ﬁgure 2) .

Initially, the system divides into clusters the red

graph. At the end, there are as many subgraphs as

doctors in the emergency area. Speciﬁcally our sy-

stem generates a k − partition based on (Hespanha,

2004), where k is the number of doctors (#d). The

system assigns to each doctor, depending on the loca-

tion, the node that is the highest priority and closest to

the coloration performed. That is, the nearest doctor

IoTBDS 2017 - 2nd International Conference on Internet of Things, Big Data and Security

278

Figure 2: Graph based on Delaunay Triangulation.

is distributed for each partition, excluding the doctors

already assigned. This is a quick solution for distri-

buting to doctors in different areas. If a new node is

generated, the system automatically add it to the nea-

rest cluster, and a new doctor’s route is recalculated.

Deﬁnition 1. k-partition. Consider G = (V, E) as an

undirected graph with the set V as vertex and the set

E as edges and where the edge cost function is c :

E → [0, ∞). A k − partition of V is a collection P =

, V

, ..., V

} of k disjoint subsets of V , whose union

equals V . The cost associated with P is deﬁned by:

C(P) =

∑

i6= j

∑

(v,v)∈Ev∈V

,v∈ j

c(v, v)

The l −bounded Graph Partitioning (l −GP) pro-

blem is based on ﬁnding a k − partition P that mi-

nimizes C(P), with no more than l vertices in each

partition. The problem is based on the MAXk −

CUT problem (de Sousa et al., 2016) that ﬁnd a par-

tition for F that maximizes the reward for a edge-

reward given as r : V × V → [0, ∞), where r(v, v) =

r(v, v), ∀v, v ∈ V . We considered a variation of this

problem called Hypergraph Max k-CUT (HMkC)

problem (Ageev and Sviridenko, 2000) with the sizes

of parts given and for a set of k integers s

, s

, ..., s

adds the constraint |V

| = s

, ∀i.

Note if there are red nodes (number of red no-

des #r) the other colours are not considered. When

these victims are attended the yellow nodes(y) are ta-

ken into account and ﬁnally green nodes(g).

When the graph is divided as in ﬁgure (ﬁgure 3)

the system assign one doctor for each zone. Then the

system analyses the path of each subgraph. This is

the problem known as the Travelling Salesman Pro-

blem (TSP) (Hoffman et al., 2013) and we solve this

through a Genetic Algorithm (Mudaliar and Modi,

2013).

These methods are adaptive and may be used to

solve optimisation and search problems. They are in-

spired by the behaviour of the species to evolve and

belong to the group of genetic algorithms.

Figure 3: k-Partition graph.

Populations are made up of different individuals.

In the problem posed here when talking about indivi-

duals we refer to victims / possible routes that can be

obtained.

A simulation has been carried out in order to vali-

date the use of this approach to build the routes. For

each subgraph a population of 100 individuals is rand-

omly generated. A selection of the best four individu-

als (the lowest cost route) is made. From them the

parts that routes have in common are selected as pa-

rents for generating the new population and children

are generated by permuting the order of the part that

does not match.

Once the new population is generated, random

mutations based on three different operations as

shown in the ﬁgure 4 are made. This iteration is repe-

ated 1000 times before getting the ﬁnal result.

(a) Flip operation

(b) Swap operation

Figure 4: Operations to generate mutations.

Finally, once the genetic algorithm is applied to

each subgraph obtained after the partition, the diffe-

rent routes for each doctor are obtained, such as it is

illustrated in ﬁgure 5.

The number of iterations and the population size

was chosen based on the results of time and costs we

obtained in different simulations. These values are

an approximation that can be adjusted at any time.

Thus, a graph of routes is generated for each doctor,

IBSC System for Victims Management in Emergency Scenarios

279

(a) Doctor #1 solution (b) Doctor #2 solution

(e) Doctor #5 solution (f) Doctor #6 solution

Figure 5: Routes of Doctors.

based on the combination of different subgraphs that

are produced with patients of the same priority level.

At the time of generating the graph of the following

categories (colours), the last vertex added to the previ-

ous graph is the starting point of the new graph. This

generation of separated subgraphs is based on the re-

gulations when applying triage schemes because pa-

tients may be attended in order depending of the seve-

rity of injuries. If a patient walks, and a medical staff

re-triages her/him, the system updates the information

and the mobile application updates the NFC tag if it

is necessary.

The incorporation of new medical staff or new ca-

sualties does not cause any problems or additional

cost. If more nodes are added to the graph the doctors’

routes are updated paying attention to the new charac-

teristics of the affected area. The routes will be rese-

ted and each doctor can continue his/her work wit-

hout worrying about such distractions. Given a con-

straint, doctors who are in the “travelling time” will

not receive the route update until he/she has attend

next victim, this will the starting point of the route

this is never stored in the NFC tag.

5 VICTIMS IDENTIFICATION:

NFC TAGS AND HMAC

SCHEME

A member of the medical staff is who assign NFC tags

to victims in the system proposed. All of these tags

contain the result of the triage, that is to say the colour

of the triage classiﬁcation, jointly with the location

and the result of a HMAC generated by the server, the

physical identiﬁer of the NFC tag (idTag) and some

server data explained later in the paper. The stored

information will serve as patient identiﬁcation both in

for triage as well as in the medical records generated

later on at the hospital. If some data is gathered the

system sends it to the server

The use of smartphones helps in the identiﬁcation

of patients through NFC stickers by using phones as

NFC readers. Apart of this, devices send the physical

tag identiﬁer to the server. In the server, two 64 by-

tes arrays are generated (Smart, 2016). They are ipad

and opad arrays, and they have default values deﬁned

at the initialization stage. The new arrays are genera-

ted through a XOR operation combining the previous

values and the Master Secret Key (msk). The results

are ipadkey y opadkey arrays (ﬁgure 6). After that,

the HMAC value is generated with the physical tag

identiﬁer and the triage colour result T

result

(it is a let-

ter for each colour: B, black; R, red; Y , yellow and

G, green), so the system applies a hash function to the

concatenation of these ﬁelds and the ipadkey. The

output of this hash concatenated with the opadkey is

the input to another hash function.

Figure 6: HMAC Keys Generation.

The global function may be described as:

H1 = HASH(ipadkey||idTag||T

result

)

HMAC(Tid, msk) = HASH((opadkey)||H1)

The hash function chosen for the implementation

is a SHA3

512

. The ﬁnal output will be the identiﬁer

that the will be saved in the NFC sticker tag, as you

can see in ﬁgure 7.

When a doctor or a member of the medical staff

want to access to the triage result of a patient, in

the affected zone, he/she has to read the NFC sticker

through the mobile application which sends the data

of the physical tag identiﬁer and HMAC to the server.

The server is who veriﬁes the authenticity of the tag

IoTBDS 2017 - 2nd International Conference on Internet of Things, Big Data and Security

280

Figure 7: HMAC Hash Operation.

and generates a new node. The doctors can see all the

nodes in the mobile phone, speciﬁcally the nodes on

his/her routes.

6 MEDICAL STAFF SECURITY:

IBSC SCHEME

Different communications modes are supported rela-

ted with medical staff. On the one hand, the commu-

nication with the server (to check NFC tags authenti-

city and synchronizing routes) and, on the other hand

the communication between them (video-calls and

chats). Authentication against the server and peers

and integrity of shared data is included. In both com-

munication modes an ID-Based Signcryption scheme

(IBSC) is used in order to achieve secure communica-

tions. This complex cryptosystem is a combination of

ID-Based Encryption (IBE) and ID-Based Signature

(IBS) that provides private and authenticated delivery

of information between two parties in an efﬁcient way

with a composition of an encryption scheme with a

signature scheme (Boyen, 2010). This approach of-

fers the advantage of simplifying management by not

having to deﬁne a public key infrastructure. This type

of scheme was chosen due to its low computational

complexity, efﬁciency in terms of memory and its usa-

bility.

A crucial part of the proposal is a Private Key Ge-

nerator (PKG), a server in charge of generating health

staff private keys. The identiﬁer of medical staff is

the number of registered medical practitioners and for

nurses the same (ID). Next, we describe the mathe-

matical basic tools used as well as the notation inclu-

ded in their description.

Deﬁnition 2. Considering two cycling groups (G, +)

and (V, ·) of the same prime order q. Pis a generator

of G and there is a bilinear map paring ˆe : G ×G → V

satisfying the following conditions:

• Bilinear: ∀P, Q ∈ G and ∀a, b ∈ Z, ˆe(aP, bQ) =

ˆe(P, Q)

• Non-degenerate: ∃P

, P

∈ G that ˆe(P

, P

) 6= 1.

This means if P is generator of G, then ˆe(P, P) is

a generator of Q.

• Computability: there exists an algorithm to com-

pute ˆe(P, Q), ∀P, Q ∈ G

Some hash functions denoted as follows are also

needed: H

: {0, 1}

∗

→ G

∗

, H

: {0, 1}

∗

→ Z

∗

, H

∗

→ {0, 1}

, where the size of the message is deﬁned

by n. The signcryption scheme used is the ID-Based

Signcryption Scheme (IDSC) proposed in (Malone-

Lee, 2002). Next we describe some basic notation

used: x

←− S stands for an element x randomly selected

from a set S, x ← y denotes the assignation of the va-

lue y to x and || is used for concatenation.

The steps needed for the signcryption scheme are

the following:

• SETUP: The initial parameters are established

and the server generates the master public key

(mpk) and the master secret key (msk). For that

a prime q based on some private data k ∈ Z, two

groups G and V of order q and a bilinear pairing

map ˆe : G × G → V are selected. P ∈ G is selected

randomly and the hash functions H

, H

and H

are also chosen.

msk

← Z

∗

mpk ← msk · P

• EXTRACT (ID): In this step, the secret key for

each member of the medical staff based on their

ID is generated. The public key Q

∈ G and

the secret key S

∈ G are calculated taking into

account the msk. It should be pointed out that

this key exchange between server and the doctor

is performed using the stream cipher SNOW3G

(Santos-Gonz

alez et al., 2014) under the session

key obtained through an Elliptic Curve Difﬁe-

Hellman (ECDH)(Bos et al., 2014). the safety of

following connections as you can see in ﬁgure ??.

← H

(ID)

← msk · Q

• SIGNCRYPTION (S

, ID

, m): All the mes-

sages m ∈ {0, 1}

will be encrypted and signed.

The receiver’s public key is generated taking into

account ID

and then the message is signed with

and encrypted with Q

giving as result σ (a

t-uple of three components: c, T, U).

← H

(ID

)

← Z

∗

T ← x · P

IBSC System for Victims Management in Emergency Scenarios

281

r ← H

(T ||m)

W ← x · mpk

U ← r · S

y ← ˆe(W, Q

)

k ← H

(y)

c ← k ⊕ m

σ ← (c, T, U)

• UNSIGNCRYPTION (ID

, S

, σ): If every-

thing is right, the message m ∈ {0, 1}

is returned.

Otherwise, if there are some problems in the sig-

nature or in the encryption of m, ⊥ is returned.

The sender’s public key is generated taking into

account ID

and then the message is unencrypted

with S

← H

(ID

)

split σ as (c, T, U)

y ← ˆe(S

, T )

k ← y

m ← k ⊕ c

r ← H

(T ||m)

Veriﬁcation:

ˆe(U, P) == ˆe(Q

, mpk)

· ˆe(T, mpk)

Note: if the veriﬁcation is successful m is retur-

ned, otherwise ⊥ is returned.

7 SECURITY ANALYSIS

The proposed scheme provides protection against dif-

ferent attacks. In this sections some of them are pre-

sented. On the one hand, a spooﬁng attack and/or

cloning of the card will be hardly successful since it

would involve the generation of the HMAC described

taking into account the master key of the server and

the ID card. Even if an outsider obtains this informa-

tion, it should be noted that the physical identiﬁer of

a NFC tag is unique to each element. On the other

hand, if someone emulate a NFC card from an An-

droid device, in this operating system, the emulated

device goes from being passive to being active. So

the attack would be detected since the application has

the restriction that only read NFC tags that are pas-

sive. At the time of its implementation in Android are

different and completely distinguishable communica-

tions.

Attacks related to make multiple requests to the

server, called Denial of Service (DoS) attack, are re-

stricted because only requests associated with a num-

ber of legitimate members of the medical staff will

take effect. Once the corresponding private key is as-

signed, more requests of this kind will be not atten-

ded.

Finally, the typically “Man in the Middle” attack

which conveys a successful authentication to the ser-

ver with an identiﬁer of legitimate members of the

medical staff is improbable. This false identiﬁca-

tion would be easily detectable because the number

of members who can make requests to the server is

limited to those who are working at the time of the

request. This authentication is one of the most impor-

tant points on every cloud computing system based on

mobile phones (Alizadeh et al., 2016).

8 CONCLUSIONS AND FUTURE

WORK

In this work, a system has been presented to may to

improve logistics and attention of casualties in ex-

treme situations. The priority is to serve the greatest

number of injuries using the shortest possible time

and cost. The tool consists on a mobile application,

NFC tags and a web service. The mobile application

helps health staff to know in every moment the posi-

tion of the victim and where they must go. Speciﬁ-

cally the system create a graph based on the Delau-

nay Triangulation and uses a k − partition to divide

it in clusters. Different subgraphs are obtained, as

many ones as doctors in the emergency area. When

the graph is divided the system assign one doctor for

each zone. Then the system analyses the path of each

subgraph through a Genetic Algorithm to solve it like

a TSP. The system has an “emergency support” tool

to contact peers through a video call when doctors re-

quire additional support. Data security is a key ob-

jective, so for this reason a HMAC scheme is used

to protect NFC tags and an ID-Based Signcryption is

used for the communications. A ﬁrst approach has

been implemented in Android and Nodejs with NFC

tags. More functionalities can be added to the server,

such as statistics, a real-time map with events, etc.

Thus, this task is part of a work in progress.

ACKNOWLEDGEMENTS

Research supported by TESIS2015010102, TE-

SIS2015010106, RTC-2014-1648-8, TEC2014-

IoTBDS 2017 - 2nd International Conference on Internet of Things, Big Data and Security

282

54110-R, MTM-2015-69138-REDT and DIG02-

INSITU.

REFERENCES

Ageev, A. A. and Sviridenko, M. I. (2000). An approxima-

tion algorithm for hypergraph max k-cut with given

sizes of parts. In Algorithms-ESA 2000, pages 32–41.

Springer.

Alizadeh, M., Abolfazli, S., Zamani, M., Baharun, S., and

Sakurai, K. (2016). Authentication in mobile cloud

computing: A survey. Journal of Network and Com-

puter Applications, 61:59–80.

Baracoda (2017). Idbluean efﬁcient way to add rﬁd re-

ader/encoder to bluetooth pda and mobile phones.

Available online: http://www.baracoda.com (accessed

on 15 February 2017).

Bos, J. W., Halderman, J. A., Heninger, N., Moore, J., Na-

ehrig, M., and Wustrow, E. (2014). Elliptic curve

cryptography in practice. In International Conference

on Financial Cryptography and Data Security, pages

157–175. Springer.

Boyen, X. (2010). Identity-based signcryption. Springer.

de Berg, M., Cheong, O., van Kreveld, M., and Overmars,

M. (2008). Delaunay triangulations. Computational

Geometry: Algorithms and Applications, pages 191–

218.

de Sousa, V. J. R., Anjos, M. F., and Le Digabel, S. (2016).

Computational study of valid inequalities for the max-

imum k-cut problem.

Facebook (2017). Facebook safety check. Available on-

line: https://www.facebook.com/about/safetycheck/

(accessed on 15 February 2017).

Foundation, S. (2017). Open source disaster

management software. Available online:

https://sahanafoundation.org/ (accessed on 15

February 2017).

Gao, T., Massey, T., Selavo, L., Crawford, D., Chen, B.-

r., Lorincz, K., Shnayder, V., Hauenstein, L., Dabiri,

F., Jeng, J., et al. (2007). The advanced health and

disaster aid network: A light-weight wireless medical

system for triage. Biomedical Circuits and Systems,

IEEE Transactions on, 1(3):203–216.

Google (2017). Google person ﬁnder web page. Available

online: https://google.org/ (accessed on 15 February

2017).

Hespanha, J. P. (2004). An efﬁcient matlab algorithm for

graph partitioning. Santa Barbara, CA, USA: Univer-

sity of California.

Hoffman, K. L., Padberg, M., and Rinaldi, G. (2013). Tra-

veling salesman problem. In Encyclopedia of opera-

tions research and management science, pages 1573–

1578. Springer.

Inoue, S., Sonoda, A., Oka, K., and Fujisaki, S. (2006).

Emergency healthcare support: Rﬁd-based massive

injured people management. In Proceedings of the

fourth International Workshop on Ubiquitous Com-

puting for Pervasive Healthcare Applications, Irvine,

CA.

Knox, R. W. (2015). Marcq saint-hilaire without tears. The

International Hydrographic Review, 52(2).

Lee, J.-S., Su, Y.-W., and Shen, C.-C. (2007). A compara-

tive study of wireless protocols: Bluetooth, uwb, zig-

bee, and wi-ﬁ. In Industrial Electronics Society, 2007.

IECON 2007. 33rd Annual Conference of the IEEE,

pages 46–51. IEEE.

Malone-Lee, J. (2002). Identity-based signcryption. IACR

Cryptology ePrint Archive, 2002:98.

Mudaliar, D. N. and Modi, N. K. (2013). Unraveling travel-

ling salesman problem by genetic algorithm using m-

crossover operator. In Signal Processing Image Pro-

cessing & Pattern Recognition (ICSIPR), 2013 Inter-

national Conference on, pages 127–130. IEEE.

Neuenschwander, M., Cohen, M. R., Vaida, A. J., Pat-

chett, J. A., Kelly, J., and Trohimovich, B. (2003).

Practical guide to bar coding for patient medication

safety. AMERICAN JOURNAL OF HEALTH SYSTEM

PHARMACY, 60(8):768–779.

of Medicine at NIH, N. L. (2017). People loca-

tor and reunite web page. Available online:

https://lpf.nlm.nih.gov/ (accessed on 15 February

2017).

Rivero-Garcıa, A., Hern

andez-Goya, C., Santos-Gonz

alez,

I., and Caballero-Gil, P. (2014). Fasttriaje: A mobile

system for victim classiﬁcation in emergency situati-

ons.

Santos-Gonz

alez, I., Rivero-Garc

ıa, A., Caballero-Gil, P.,

and Hern

andez-Goya, C. (2014). Alternative commu-

nication system for emergency situations. In WEBIST

(2), pages 397–402.

Sesia, S., Touﬁk, I., and Baker, M. (2009). LTE: the UMTS

long term evolution. Wiley Online Library.

Smart, N. P. (2016). Hash functions, message authentication

codes and key derivation functions. In Cryptography

Made Simple, pages 271–294. Springer.

Want, R. (2011a). Near ﬁeld communication. IEEE Perva-

sive Computing, (3):4–7.

Want, R. (2011b). Near ﬁeld communication. IEEE Perva-

sive Computing, (3):4–7.

Zou, Z., Chen, Q., Uysal, I., and Zheng, L. (2014). Ra-

dio frequency identiﬁcation enabled wireless sensing

for intelligent food logistics. Phil. Trans. R. Soc. A,

372(2017):20130313.

IBSC System for Victims Management in Emergency Scenarios

283