A Systemic Approach for IoT Security

Arbia Riahi, Yacine Challal, Enrico Natalizio, Zied Chtourou, Abdelmadjid

Bouabdallah

To cite this version:

Arbia Riahi, Yacine Challal, Enrico Natalizio, Zied Chtourou, Abdelmadjid Bouabdallah. A
Systemic Approach for IoT Security. DCOSS, 2013, Boston, United States. pp.351-355,
�10.1109/DCOSS.2013.78�. �hal-00868362�

HAL Id: hal-00868362

https://hal.science/hal-00868362v1
Submitted on 1 Oct 2013

HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est

archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents
entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non,
lished or not. The documents may come from émanant des établissements d’enseignement et de
teaching and research institutions in France or recherche français ou étrangers, des laboratoires
abroad, or from public or private research centers. publics ou privés.
 A systemic approach for IoT security

Arbia Riahi∗ , Yacine Challal † , Enrico Natalizio† , Zied Chtourou∗ , Abdelmadjid Bouabdallah†
∗ VRIT Lab - Military Academy of Tunisia, Nabeul, Tunisia. e-mail: [email protected], [email protected]
† Heudiasyc - Université de Technologie de Compiègne, Compiègne, France. e-mail: <firstname.lastname>@hds.utc.fr

Abstract—In this paper we want to explore a new approach

for security mechanisms design and deployment in the context
of Internet of Things (IoT). We claim that the usual approach to
security issues, typical of more classical systems and networks,
does not grab all the aspects related to this new paradigm of
communication, sharing and actuation. In fact, the IoT paradigm
involves new features, mechanisms and dangers that cannot
be completely taken into consideration through the classical
formulation of security problems. The IoT calls for a new
paradigm of security, which will have to consider the security
problem from a holistic perspective including the new actors and
their interactions. In this paper, we propose a systemic approach
to security in IoT and explore the role of each actor and its
interactions with the other main actors of the proposed scheme.
Keywords—Internet of Things, Security, Systemic Approach

I. I NTRODUCTION
The Internet of Things (IoT) paradigm is one of the most
thrilling innovations of the recent years. The exploitation of Figure 1. A systemic approach for IoT security
the IPv6 addressing space, along with the miniaturization of
electronic and transceiver devices opened the way to provide
each object on Earth with an Internet address and the tech- II. N ODES
nological support to transform it in a communicating object.
Once each object possesses communication capabilities, the In this Section we will present the main actors of the
number of possible applications becomes potentially infinite. systemic approach to security in IoT, introduced in [1]. It is
This good news is counterbalanced by the consideration that worth to note that the real novelty of the scheme in Figure
also the number of possible attacks to persons’ and objects’ 1 is the introduction of the “Intelligent Object” at the center
security will grow exponentially. Therefore, a new paradigm of of the interactions among Person, Process and Technological
trust, security and privacy is required to face these future issues Ecosystem. In the following we will introduce each of the
in the IoT. In [1] authors describe a systemic and cognitive mentioned actors and their functions in the scheme.
approach for IoT security. In their work, they consider three
main axes: effective security for tiny embedded networks, A. Person
context-aware and user-centric privacy, and the systemic and The first node plays a fundamental role in the IoT security
cognitive approach for IoT security. In this paper, we will framework. The human resources are responsible for security
focus on the third axis. Actually, authors affirm that the rules management, which includes:
IoT is a complex system in which people interact with
the technological ecosystem based on smart objects through • Defining security practices and rules.
complex processes [1] as shown in figure 1. In this approach,
connections between different nodes have a specific character • Auditing practices and rules efficiency.
depending on complex environment of the IoT. By taking into • Applying practices and rules when into operational
consideration the dynamic and complex nature of this model, mode.
in this work, we will present our perspective in respect of the
main elements illustrated in Figure 1 and that we will call Due to the complex environment of the IoT, this node is a vital
“nodes” and “tensions”. component in security management and enhancement. To this
purpose, the human component should be able to analyse the
In order to explain this model, we will describe each node context of IoT, individuate its advantages and limitations, and
and its functions in Section II. Then, we believe that the exploit the technology evolution to bring adequate solutions.
tensions between the different nodes need a special study and
discussion, which will be our goal in Section III. In Section
B. Process
IV we will provide the reader with the logical connections
between some classical domain of applications for the IoT and The second node refers to a means to accomplishing
the presented scheme. Finally, we will provide our conclusions tasks in the IoT environment according to some security
in Section V. requirements. The process is required to be compliant with
the security policies in order to keep the environment secure III. T ENSIONS
at different levels. Furthermore, due to the complexity of the
model and the presence of different interactions originating In the systemic and cognitive approach for IoT security of
from this node, security processes are difficult to implement. Figure 1, the nodes are the originating and destination actors
of a tension that represents their interaction, and takes into
The Federal Financial Institutions Examination Council’s consideration the complexity of the environment. Specifically,
(FFIEC) presented a first classification of standard areas to the tensions that we are going to consider are: identifica-
deal with when considering security processes: tion/authentication, trust, reliability, auto-immunity, privacy,
responsibility and safety. To better explain our systemic ap-
• Information Security Risk Assessment. proach, these tensions need to be deeply analyzed, measured
and discussed.
• Information Security Strategy.
• Security Controls Implementation. A. Identification and authentication
• Security Monitoring. “Identification and authentication” is the tension that ties
the intelligent object with the person. In the IoT context,
• Security Process Monitoring and Updating [3]. objects are spread globally. An efficient resolution scheme
needs to be set to identify different entities. Privacy and other
In practices, security process need to meet requirements of security issues must be taken into consideration as well as the
standards, strategies, policies, procedures and other afferent specific function of the object, which can change over the time.
documents. Thus, an adequate compromise must be found Furthermore, an object can have one core identity and several
between complexity of security process practices and the temporary identities; an hospital can become a meeting place
needed security level. for a health conference or a shelter after a fire [4].
A lot of research has already been proposed on this axis.
C. Intelligent Object We will limit our analysis to the presentation of some impor-
tant projects of this domain. First, we cite SENSEI (Integrating
This node is the heart of the new approach. It refers the Physical with the Digital World of the Network of the
to an “object” augmented by the electronic features needed Future) which proposes an architecture that integrates Wireless
to let it communicate with other objects in the surrounding Sensors and Actuators networks to ensure their cooperation [5].
environment. These objects will become active participants in The second project is BRIDGE (Building Radio Frequency
business, information and social processes [2]. In fact, objects IDentification for the Global Environment) which aims to
in the IoT framework will be able to cooperate, share and find and invent tools permitting the deployment of RFID
exchange information about the environment, and respond and EPCglobal Network applications [6]. Other projects can
to events happened in the environment by accomplishing be listed here, such as SmartProduct [7], SWIFT (Secure
adequate operations. Due to their expected pervasivity, the Widespread Identities for Federated Telecommunications) [8],
correct design and development of security practices within the QR-Code [9] and Digital Object Identifier [10].
conception of intelligent objects is fundamental to ensure the
right level of security to the whole environment surrounding Open research issues: In [2], many research issued have
them. been illustrated. For example, global ID schemes need to
be considered when intelligent objects and humans interact.
Also, an efficient identity management approach should be
D. Technological ecosystem defined. Mobility, privacy, pseudonimity, anonymity aspects
This node refers to technological choices made to ensure need deeper analysis and research. For example, when we limit
IoT security. According to [26], information security technol- our scope to to the RFID framework, we can easily individuate
ogy falls into several broad categories: partially studied research topics related to the definition of
distributed logical readers or the study of a RFID networks
• Security Design and Configuration where both tags and readers are mobile.

• I&A: Identification and Authorization B. Trust

• Enclave internal “Trust” is the tension that ties the intelligent object with
• Enclave boundary the technological ecosystem. Basically, it represents the level
of confidence that the environment can grant to the intelligent
• Physical and environmental object. The IoT environment can be permeated with very het-
erogeneous objects, which might differ for both their functions
The choices related to each of these elements may include and their capabilities. In such a heterogeneous environment,
system architecture, communications protocols, implemented when defining trust management, we must take into account
algorithms, access control methods, performance, etc. It is also the severe resource constraints to which the objects are
evident that a trade-off among security requirements, feasi- subjected, and which will constraint the choices of the techno-
bility and technology evolution should be found in order to logical ecosystem. Thus, trust management operations such as
ensure the appropriate level of security without degrading the establishing, updating, and revoking keys and certificates are
performance of the system. very important research topics in the IoT framework.
 An important project that deals with the mentioned issues D. Responsibility
is uTRUSTit [11]. Its objective is to model and implement a
tool for building and testing trust. In another context, Gligor “Responsibility” is the tension that ties the intelligent
and Wing present “a theory of trust in networks of humans and object with the process. In order to share resources and other
computers that consists of elements of computational trust and added values, which are useful for different processes, privi-
behavioral trust” [12]. In [14], authors propose a subjective leges and access rights must be clearly defined according to
behavioral trust model for Social IoT, which exploits the privacy constraints. In addition, responsibilities and liabilities
“social” bound existing among objects (ownership, parental, rules of each entity must be considered in order to avoid
co-location, etc.). This model has been further detailed in [13]. dangers when the object regulates a process.

Open research issues: The main objectives of trust research In literature, two main access control models have been
in IoT framework are the following. First, the conception of developed: Role-based access control (RBAC) and Attribute-
new models for decentralized trust. Second, the implementa- based access control (ABAC). In practice, main implemen-
tion of trust mechanisms for the cloud computing. Third, the tations include XACML (Policy decision language based on
development of applications based on node trust (ex. routing, XML) and its extension known as Distributed Access Control
data aggregation, etc.). [22]. Recently a new model was proposed in [32]. The IACAC
(Identity Authentication and Capability based Access Control)
According to [12], an interesting issue is to develop a presents a new scheme for authentication and access control
theory for computational trust. In turn, this means to deal in IoT and aims at replacing the existing approaches.
with relationships between computational trust and behavioral
trust, in order to create new protocol areas, and to maintain Open research issues: The main objective of responsibility
stability trust properties. In practice, authors propose a network for IoT is to make access control rules easy to create, under-
infrastructure to manage trust concepts [12]. stand and manipulate. In this sense, a possible research issue is
the integration of the IACAC model in an identity management
On the another hand, when managing trust, aspects such as component of a RFID middleware [32]. Another direction can
topology of the objects, coverage deployment, target tracking, focus on access control rules propagation and revocation [32].
localization and IoT applications should be considered [27].

C. Privacy E. Autoimmunity
“Privacy” is the tension that ties the person with the “Autoimmunity” ties the intelligent object in self-loop. The
technological ecosystem. Privacy is an important tension in objective of this tension is to propose an artificial immune
the systemic model for IoT security because of the ubiquitous system solution for IoT. In this trend, two main research
character of the environment. Despite the existence of adequate activities can be evoked here. First, authors in [24] describe a
research activities in privacy management mechanisms in simulation and immunity test of a wireless sensor in order
general, there is still a list of objectives to be fulfilled. To to avoid electromagnetic disturbance in substation. Second,
make things clearer, in [16], authors divide privacy into three immunity-based schemes can be used to detect intrusions in
main axes: 1) Privacy in data collection, 2) Privacy in Data the IoT. For example, the authors of [23] simulate self and non-
Sharing and Management and 3) Data security issues. In the self antigen in IoT, as well as immature, mature and memory
following we will cite the most important research activities detector, to detect attacks in the IoT.
for each of these directions.
Open research issues: As a relevant open research issue
With reference to the first research axis, “privacy in data related to the autoimmunity, we propose the conception of
collection”, we can mention the cryptographic solutions and a new autoimmunity technique, where intelligent objects can
the blocking approaches detailed in [15] and [19]. For the distinguish if special access or privileges are permitted to the
second axis, we can enumerate aggregation of data collected by all or a part of the system content data, and react according to
sensors [16], the Platform for Privacy Preferences (P3P) [17], the context constraints.
semantic web [18] and other privacy-preservation mechanisms,
such as: k-anonymity, l-diversity, and t-closeness. Finally, in
the data security issues, we can name password protection F. Safety
[19] and [20], cryptographic solutions and web entities with a
semantic policy language [21]. “Safety” is the tension that ties the person with the process.
An environment permeated with intelligent objects is supposed
Open research issues: Even though a lot of research has to cope with many security challenges. One of these is ensuring
already been proposed for this tension, still many topics need safety when a sudden failure occurs for one or many system
to be further investigated. Here we can propose a list of components. Then, safety must be considered as a mean to
interesting topics. First of all, the automated key management reduce the possibility of damage.
scheme. This operation is very sensitive in the case of IoT.
It may include key provisioning, updating, revocation, trans- When considered as a research axis for the IoT, safety
porting and key agreement. Also non-cryptographic operations purposes have been widely investigated. We can list as ex-
like enrollment, backup and recovery should be addressed to amples: the CuteLoop project [28], EURIDICE (European
guarantee a high level of security. Another issue is to develop Inter-Disciplinary Research on Intelligent Cargo for Efficient,
a new scheme for asymmetric key management including safe and environment-friendly logistics)[29] and SToP (Stop
generation, validation and distribution. Tampering of Products) [30].
 Table I. I OT A PPLICATION D OMAINS -T ENSION EXAMPLES
Tension / Ap-
Identification Privacy Trust Safety Responsibility Reliability Auto-immunity
plication
At the
RFID-based subscription Objects that
identification moment, a perfectly Detection, Stop techniques
Transportation Vehicles and Traffic
management of customer can accomplished analysis and in case of
and logistics consumers monitoring and
consumers, provide private their previous avoidance of intrusion
domain safety services control devices
providers and data in order to tasks in a supply anomalies detection
products benefit from chain
some advantages
Patients should Medical
Data, including institutions must
trust medical Parameters
Identification of clinical ensure the Reliability of the An object that
Health-care institutions in setting of
staff and diagnosis and patient safety link in case of provides alert in
domain terms of health-care
patients treatment, must during remote diagnosis case of accident
reliability and objects
be kept private health-care
privacy
activities
A trainer that For the
Financial details continuity of
assign the
Identification of of a given services
Newly bought adequate Disaster
Smart the employees project Control devices assurance, an
objects exercise prediction and
environment in the same (domestic or for personal electric power
belonging to the according to the alerting
domain enterprise professional) environment supply reliability
same owner health
should be should be
parameters of
confidential guaranteed
the trainee

Open research issues: Several applications can be consid- V. C ONCLUSION

ered as drivers of technological and scientifical innovations
along this research axis. For example, we can focus on the In this paper we proposed a systemic approach for IoT se-
environment observation: pollution effects, forest fires studies, curity base on [1]. The model is made up of four nodes: person,
etc.; the physical security of building: VMC, leaks, intruders, technological ecosystem, process and intelligent object. The
etc.; and also the commercial field: protection of products last node is the newest and reflects the IoT dimension. These
against counterfeit. nodes interact through tensions, namely identification, trust,
privacy, safety, auto-immunity, reliability and responsibility.
As a first task, we aimed to define each node and its roles.
G. Reliability Then, we focused on the analysis of literature and open issue
related to the tensions. To this purpose, we described each
“Realibility” is the tension that ties the process with the tension’s meaning, effect, related work and possible research
technological ecosystem. The reliability deals with data and issues. Finally, we proposed real examples taken from classical
communications management. The reliability aims at guaran- application domains to substantiate the use of our systemic
teeing availability of information over time through efficient approach.
ways of managing data repositories. Reliability of communi-
cation links can be ensured through the redundancy provided
by multiple paths. R EFERENCES
In this direction, we can list two main projects: the Soft Re- [1] Yacine Challal, Internet of Things Security: towards a cognitive
liability Project [25] and PEARS (Feasibility Privacy-Ensuring and systemic approach, HDR Thesis, Université de Technologie
Affordable RFID System / Feasibility) [31]. de Compiègne, 2012.
[2] O. Vermesan, P. Friess, P. Guillemin, S. Gusmeroli, H. Sund-
Open research issues: An open research issue is related maeker, A. Bassi, I. S. Jubert, M. Mazura, M. Harrison, M.
with the development of clustering communication strategies Eisenhauer, P. Doody, Internet of Things Strategic Research
Roadmap, 2011.
to ensure links reliability. Another issue will focus on creating
an automated solution for IoT service management to enhance [3] http://www.ffiec.gov.
their reliability. [4] R. Roman, P. Najera, J. Lopez, Securing the Internet of Things,
IEEE Computer, vol. 44, no. 9, pp. 51–58, September 2011.
[5] http://www.ict-sensei.org.
IV. T HE SYSTEMIC APPROACH FOR NEW APPLICATIONS [6] www.bridge-project.eu.
OF THE I OT [7] http://www.smartproducts-project.eu.
[8] http://www.ist-swift.org/.
The integration of an intelligent object within the IoT im-
[9] http://www.qrcode.com.
plies the creation of new applications as well as the extension
[10] http://www.doi.org/.
of existing ones. In this Section, we want to detail the link
between some of the classical application domains and the ten- [11] http://www.utrustit.eu.
sion of our systemic approach, in order to highlight constraints [12] V. Gligor and J. M. Wing, Towards a Theory of Trust in Net-
works of Humans and Computers, 19th International Workshop
and requirements imposed by the security. Concretely, in Table on Security Protocols, Cambridge, UK, March 28-30, 2011.
I we choose transportation and logistics, health-care and smart
[13] M. Nitti, R. Girau, L. Atzori, A. Iera, and G. Morabito, A
environment as domains of application, and we characterize Subjective Model for Trustworthiness Evaluation in the Social
each tension of the proposed scheme by the means of an Internet of Things, 23rd Annual IEEE International Symposium
example in the context of that domain. on Personal, Indoor and Mobile Radio Communications, 2012.
[14] L. Atzori, A. Iera, and G. Morabito, SIoT: Giving a Social
Structure to the Internet of Things, IEEE communications
letters, 2011.
[15] D. Molnar, D. Wagner, Privacy and Security in Library RFID:
Issues, Practices and Architectures, CCS, 2004.
[16] C. C. Aggarwal, P. S. Yu, Privacy-preserving data mining :
models and algorithms, Springer, 2008.
[17] http://www.w3.org/TR/P3P11/.
[18] F. Gandon, N. Sadeh, Semantic Web Technologies to Reconcile
Privacy and Context Awareness, Web Semantics: Science,
Services and Agents on the World Wide Web, vol. 1, no. 3,
pp. 241–260, 2004.
[19] R. Kumar, E. Kohler, M. Srivastava, Harbor: software-based
memory protection for sensor nodes, IPSN Conference, 2007.
[20] R. Acharya, K. Asha, Data integrity and intrusion detection
in wireless sensor networks, Proceedings of the IEEE ICON,
2008.
[21] L. Kagal, T. Finin, A. Joshi, A Policy-based Approach to
Security for the Semantic Web, ISWC, 2003.
[22] http://xacmllight.sourceforge.net/.
[23] C. Liu, J. Yang, Y. Zhang, Research on Immunity-based Intru-
sion Detection Technology for the Internet of Things, Seventh
International Conference on Natural Computation, 2011.
[24] A. Bo, Z. Weidong, C. Xiang, L. Jikun, S. Yingbin, L. Shaoyu,
A Study on Immunity of Wireless Sensor Unit in Substation,
Electromagnetic Compatibility (EMC EUROPE), 2012 Inter-
national Symposium on , vol., no., pp.1,5, 17-21 Sept. 2012.
[25] http://www.softreliability.org.
[26] L. Kiely, T. Benzel, Systemic Security Management: A new
conceptual framework for understanding the issues, inviting
dialogue and debate, and identifying future research needs, In-
stitute for Critical Information Infrastructure Protection (ICIIP),
2008.
[27] H. C. Hsu, K. D. Chang, J. L. Chen, H. C. Chao, A Survey on
Trust Management Mechanisms for Wireless Sensor Networks
& Future Internet of Things, Journal of Electronic Science and
Technology, vol. 9, no. 4, pp. 364–367, 2011.
[28] www.cuteloop.eu
[29] www.euridice-project.eu
[30] www.stop-project.eu
[31] www.friendlytechnologies.com
[32] N. Mahalle, B. Anggorojati, N. R. Prasad and R. Prasad,
Identity, Authentication and Capability Based Access Control
(IACAC) for the Internet of Things, Journal of Cyber Security
and Mobility, Vol. 1, No. 4, p. 309-348., mar. 2013.