IoT Privacy in 5G Networks
Emanuele Catania and Aurelio La Corte
Department of Electric, Electronic and Computer Science Engineering,
Università degli Studi di Catania,
viale A. Doria 6, 95121 Catania, Italy
Keywords: Internet of Things, Ultra-Dense Networks, Privacy, Security, 5G.
Abstract: In the Internet of Things (IoT), objects, equipped with sensing, processing, storage and decision-making
capabilities, actively interact with one another and with humans. Even if they have been conceived and
programmed to make all their activities in security, several factors, such as weak implementations of
communication protocols, metadata information exchange, and architectural flaws, could jeopardize security
and privacy. Moreover, due to its complexity and attitude to change rapidly, and to the ultra-densification
trend of the current communication infrastructure, new threats to the privacy might arise. After a brief
introduction to IoT privacy issues, we describe how the evolution of the current wireless communication
infrastructure toward the 5G generation network might undermine the privacy in the IoT. Then we propose a
methodology of analysis, which looks at privacy threats from different perspectives and at various levels of
abstraction.
1 INTRODUCTION
By definition, the Internet of Things is a composition
of physical entities capable of sensing, computing and
acting in response to the information they can acquire
and manage (Sfar, et al., 2017). Thanks to this
paradigm, “people and things can be connected
anytime, anyplace, with anything and anyone, ideally
using any path/network and any service” (Kende,
2014).
Mobility, scalability, interoperability and
resource constraints characterize the million
interconnected both wireless and wired devices of
which the IoT is composed (Porambage, et al., 2016).
Ubiquity is one of the key features that the
communication infrastructure underlying the IoT
should have. Undoubtedly, cellular networks, due to
their diffusion, enable IoT implementation, also
ensuring stable transmissions and acceptable delays.
However, such networks have not been conceived to
support machine-to-machine (M2M) communication
(which are characterized by intermittent behavior and
small-sized data packets). Furthermore, it has been
foreseen that in the near future almost all data traffic
supported by communication networks will be
produced by smart devices (CISCO, 2016). Using
only cellular network might not be sufficient to
satisfy the M2M requirements, since during
transmission, machine-type communications might
easily exceed their uplink capacity.
In order to foster performance improvements of
the current communication infrastructure, given also
the prominent IoT diffusion trend, cellular networks
will be cooperating with other wireless network
technologies (e.g. WLAN, relay-assisted and device-
to-device communications, wireless personal area
networks, LTE-U). Ultra-dense networks (UDNs),
namely hierarchical networks in which the density of
access nodes is at least a magnitude greater than the
density of users, will meet future communication
requirements, providing a very high connectivity and
data rate.
1.1 Motivation
Although the benefits that it may produce, the IoT
may cause severe security implications. Inability or
unwillingness of devices owner to update and fix
devices’ security flaws, limited capability of devices,
and the lack of, or incompatibility among
communication standards makes hard addressing the
security challenges in the IoT (Mannilthodi &
Kannimoola, 2017). Leakage of sensitive information
is one of the most serious threats to the privacy. The
Catania, E. and La Corte, A.
IoT Privacy in 5G Networks.
DOI: 10.5220/0006710501230131
In Proceedings of the 3rd International Conference on Internet of Things, Big Data and Security (IoTBDS 2018), pages 123-131
ISBN: 978-989-758-296-7
Copyright
c
2019 by SCITEPRESS Science and Technology Publications, Lda. All rights reserved
123
most spread devices in the IoT cannot implement
strong security and cryptographic functions since
they are equipped with computationally and resource
limited micro-controllers (Malina, et al., 2016). For
this reason, a growing body of literature has evaluated
and proposed lightweight encryption algorithms and
privacy-by-design methodologies.
Since it is an evolving, heterogeneous, and wide
technological environment, it could be very hard
guaranteeing the privacy to the whole IoT. A method
for identifying privacy weaknesses fitting well to the
complex IoT it would be desirable. Furthermore, new
security challenges should be considered given the
paradigmatic revolution that will overwhelm current
communication networks soon.
1.2 Contribution
Figure 1: Privacy assessment and discovery methodology.
The present study has offered a framework for the
exploration of privacy issues for the IoT. It introduces
a methodology of analysis aiming at disclosing
privacy weaknesses that might affect the IoT eco-
system from different viewpoints. In particular, it is
inspired by the popular Zachman framework and the
LINDDUN framework. Among all the abstractions,
the “Network” and the “Time” (see Figure 1) enrich
the sets of observed information with respect to the
LINDDUN framework.
We show how the proposed privacy threat
analysis framework operates at different level of
abstraction, thus highlighting privacy weaknesses
from various perspectives. We describe an
application example of our methodology to the IoT,
starting from the most abstract level, namely the
“Contextual” perspective. As to avoid being long-
winded, we did not carry out a thorough analysis, thus
deepening also the “Physical” viewpoint, but we
stopped to the “Logical” is.
1.3 Paper Organization
Section 2 outlines the related works and describe the
effect on the IoT privacy of communication
infrastructure densification. Section 3 presents the
proposed privacy assessment methodology. An
example of application of the methodology on the IoT
in ultra-dense cellular networks is presented in
Section 4. Then, in the final section, the conclusion.
2 LITERATURE REVIEW
Wireless radio channels are widely used in M2M
communications. Because of their limited resources,
provide these devices with powerful, also
computationally expensive, security capabilities are
often infeasible. This makes M2M communications
extremely vulnerable. Furthermore, attacks against
information privacy become more effective as the
underlying communication infrastructure is an ultra-
dense network (Chen, et al., 2017). In this section, we
introduce some background on IoT privacy and
analyze the effect of the huge deployment of limited-
resource devices in UDNs.
2.1 Privacy in the IoT
Because of design tradeoffs in term of cost,
complexity, and energy required for fulfilling their
operation, many devices in the IoT are usually
resource-limited. To cope with unauthorized access,
data theft, and eavesdropping, devices should
be provided with authentication, authorization
mechanisms, and data preservation capabilities,
ensuring freshness, authenticity, confidentiality, and
integrity of information. Privacy (i.e. unlinkability,
data secrecy, and anonymity) has to be accurately
preserved since personal and sensitive information
could be stolen and abused by an adversary.
Encryption is fundamental to provide sensitive data
with a basic level of privacy. Indeed, it prevents that
transmitted data can be intercepted and read by
passive adversaries. Nevertheless, encrypting the
information might mean make use of computationally
expensive cryptographic primitives (e.g. pairing-
IoTBDS 2018 - 3rd International Conference on Internet of Things, Big Data and Security
124
based cryptography), which could not be executed by
every IoT device. In order to identify suitable
cryptographic approaches to the IoT, Malina et al.
(Malina, et al., 2016) measured the performance of
the most used primitives (such as RSA, secure
hashing algorithms and AES) on some of the most
common micro-controllers (ARM, MSP430f X)
equipping IoT devices. They found that while
hashing and symmetric ciphering operations take few
milliseconds and can also run on very limited
microcontrollers, stronger approaches, such as RSA
asymmetric signing (by a 2048-bit private key), can
cause delays into hundreds of milliseconds, which are
intolerable in real-time IoT applications. Processing
of complex operations could be left to the cloud or to
communication gateways, resulting in the reduction
of both devices energy consumption and computation
delays (Shariatmadari, et al., 2015). However, this
method requires trustful gateways and secure
communication among parties. A viable technique to
protect entities (both devices and users) from being
traced is hiding their real identity by means of
pseudonyms. Anyway, as also suggested in (Bailey,
2012) and in (Shaik, et al., 2015), when attackers
eavesdrop data packet within a sufficiently wide time
window of observation, they might disclose real
victims’ identifier. As described later in this paper,
when connected to an LTE network, IoT devices can
decode messages broadcasted by the network to
locate a specific subscriber. Such messages contain
only temporary identifiers, but a passive adversary
could be able to exploit decoded information to
retrieve associations among temporary unique
identifiers. Furthermore, colluding IoT users
positioned in proximity of the occasionally visited
locations by the IoT target (i.e. the victim), even
though protected by a pseudonym, might reveal to the
attacker the target’s real identity and its private
activities (Zhou, et al., 2017).
Here we argue that although the utilization of
protection approaches in design and implementation
stages, privacy objectives in the IoT could not be
achieved because of its complexity. Then, a
comprehensive understanding of motivations behind
privacy weaknesses and resulting identification of
appropriate mitigation actions in response to them
requires an organic methodology capable of
analyzing the wide and diverse IoT from distinct
perspectives.
2.2 UDN and IoT Privacy
UDNs can effectively cope with the future networks
data requirements, also provide energy and spectrum
efficiency. Composed of heterogeneous nodes with
different radio access technologies (e.g. LTE, Wi-
Max, IEEE 802.15.x), transmit powers, and coverage
area, UDNs are characterized by a multi-tier
architecture. In detail, high-power nodes and low-
power nodes, with large and small radio coverage, are
placed respectively in macro-cell tiers and in small-
cell tiers. Cellular communication infrastructure, if
from one hand make it possible offer ubiquitous
connectivity to the most devices, from the other hand
is inefficient for transmitting small, infrequent data as
required by M2M communications. Moreover,
cellular network communications could make it
possible to track entities involved in information
exchange processes (Bailey, 2012), thus affecting
their location privacy.
Spatial distribution of low-power nodes might
influence the whole network security, as asserted in
(Chen, et al., 2017). Specifically the probability of
positive secrecy rate, that is the capacity deviation of
the operating channel from the eavesdropper channel,
increases as the density of low-power nodes growths
(until a critical point, after which is not observed any
enhancement in term if secrecy performance).
Moreover, the higher the density of entities involved
in communication processes, the higher is the risk of
information eavesdropping (Yang, et al., 2015).
Undeniably, while moving within a UDN, entities are
likely to be subject to more handover processes than
in the existing networks, making it possible for
untrusted subjects to take part in the just mentioned
processes. Albeit finding trusted security
organizations responsible for credential distribution
could solve the abovementioned problem (Swetina, et
al., 2014), undesired network delays due to a large
number of involved devices, in addition to high costs,
make their adoption infeasible in practice. In
consideration of this, physical layer security seems to
be best suited for the 5G network with respect to
cryptographic security. Indeed, the former approach,
in addition to having high scalability, does not require
complex operations to be fulfilled (differently than
the cryptographic). Even computationally powerful
adversaries, in fact, cannot compromise the network
security (Yang, et al., 2015).
Despite scientific community has raised many
concerns about UDN security and IoT security, no
one to the best of our knowledge has studied the effect
of network densification on the privacy of the IoT.
IoT Privacy in 5G Networks
125
Figure 2: In this figure, we describe the proposed privacy threat modelling. We identified four main steps, namely “describe
the system”, “map privacy threats to system elements”, “identify system-specific weaknesses”, and “prioritize threats”.
Ellipses and rectangles represent respectively the entities involved and the actions they perform when interacting with each
other in the privacy threats identification process.
3 METHODOLOGY
To tackle the problem of privacy in IoT, we propose
an assessment methodology which combines the
popular Zachman framework (ZF) (Zachman, 1987)
with the LINDDUN framework (Wuyts, 2015). The
proposed approach aims at providing a tool for
acquiring awareness about, and then react to, privacy
weaknesses that might affect the system from both
microscopic and macroscopic perspectives.
LINDDUN is mainly a methodological approach,
which uses data flow diagrams to list entities,
processes, data flows, and data stores. Then, by mean
of further successive steps, it maps, elicits and
prioritizes threats, guiding towards the identification
of mitigation strategies and privacy enhancing
technologies. The ZF allows logically organizing and
classifying artifacts involved in the design and
development of information systems. Different
perspectives match with different aspects of the
system, allowing decomposing the verification of
privacy properties in small, though sometimes
interdependent, modules. Privacy assessment on IoT
applications is a large complex task that requires a
systematic verification approach on both software
and hardware. We remark that the LINDUNN
framework is not aimed at the IoT domain. In addition
to entities, data (flows) and data processes,
knowledge on physical location in which event
happens (e.g. authentications, data exchange)
together with time information, might help to better
understand motivations behind privacy issues and
identify more suited privacy enhancing solutions.
Here we give some explanation about the
“Perspective” dimensions of our proposal.
Contextual (i.e., what the system should
do): refers to the description of
information, processes, locations,
involved entities, events, and
motivations. It gives an overall, also
non-detailed, view of purposes, extents,
and relationships among elements of the
IoT eco-system or its subsystems.
Conceptual (i.e., how the system should
operate): less abstract and more
descriptive with respect to the former
IoTBDS 2018 - 3rd International Conference on Internet of Things, Big Data and Security
126
perspective, gives an overview of
models, semantic relationships, and
processes
Logical: indicates processing structure,
how applications are architected, rules
and information models.
Physical: the most concrete as compared
to the aforementioned perspectives, aims
at analyzing the IoT from technical
points of view (technology constrained
models) providing information on
physical quantities and parameters.
Each perspective aims at identifying privacy
threats by analyzing the system from different
viewpoints and may be ground for investigations into
the threats causes. Disclosed threats on one
perspective, can be reported and investigated from
other perspectives. This implies that also threats can
be described according to more viewpoints.
Moreover, information, processes, locations, entities,
and event (see Table 1) can be related to each other.
Table 1: In this table contextual and conceptual information are retrieved using a taxonomy on the IoT (Yaqoob, et al., 2017)
and a smartphone data taxonomy (Mylonas, et al., 2012).
# Data Processes Network/Location Entities Time/Events
Contextual
(What)
Measured and
elaborated data
Communication
and acquisition
of information
Requiring for
and feeding a
service
Buildings
Public areas
Private areas
Wired network
Wireless
networks
Things
Synchronous
and
asynchronous
events
Conceptual
(How)
Messages
(content,
metadata)
Connection
(re)configuration
Electronic
addresses (of
entities)
Services
Searching
(entities,
services, etc.)
Notification
Decentralized
data processing,
auditing and
information
sharing
Realtime
messaging
Connection
management and
control
Wireless
communication
systems and
infrastructures
(e.g., LTE –
LTE/A –
LTE/U, IEEE
802.11 x, IEEE
802.15 x,
WiMax, ZigBee,
etc.)
Ethernet (real-
time Ethernet,
EtherCAT),
PLC, MoCA
Smartphones
Vehicles
Laptops
Sensors
Access Points
Users
Smart home
systems
Smart healthcare
systems
Intelligent
building systems
Smart meters
Etc.
Decentralized
communication
Event
notification
Real-time-
analysis
Peer-to-peer
communication
Decentralized
auditing
Decentralized
file sharing
Logical
(LTE only)
GUTI
MSISDN
RRC Message
Body
RRC keys
Resource
configuration
Report
(measurement
and link failure)
S-TMSI
IMSI
RRC connection
establishment
RRC connection
release
Broadcast of
system
information
RRC connection
reestablishment
(NB-IoT only)
Radio link
establishment
RRC key sharing
Paging
TA
eNB Location
Device Location
MME
HHS
eNB
Relay
Device’s LTE
network
Interface
Mobility
Management
Entity
Home Subscriber
Server
Paging
Triggering
Initial security
activation
Establishment of
signal radio
bearer
Establishment of
data radio bearer
Handover
Configuration of
lower protocol
layers
IoT Privacy in 5G Networks
127
For example, connection (re)configuration
information may be related to the connection
management and control processes. Privacy threats
can be grouped into seven families, that is linkability,
identifiability, non-repudiation, detectability,
distinguishability, unawareness (of information
content), and non-compliance to policy.
For the sake of completeness, we report the
definition of threat categories, as indicated in (Wuyts,
2015). Linkability occurs when two entities can be
related to each other. Identifiability refers to a
capability of an adversary to infer the identity of an
entity. Non-repudiation stands for the inability of a
subject to demonstrate that he/she could not carry out
a specific action. Detectability implies that it possible
detect whether an entity exists or not. Disclosure of
information happens when protected individuals
information can be accessed by unauthorized entities.
Unawareness is related to unconsciousness about
supplied information to the system. To conclude, non-
compliance refers to the inability of the system to be
compliant with regulations, policies, and agreements
with users.
4 PRIVACY THREAT ANALYSIS
Most of the IoT applications require both data and
communications security, in addition to ubiquitous
connectivity. In order to be compliant with the
proposed methodology, abstractions (see Figure 1)
should be listed and analyzed from every perspective.
As to provide an example of a use of the proposed
method, in this paper privacy analyses covered only
three of the four perspectives (omitting the Physical
one).
4.1 Contextual and Conceptual
Perspectives
Contextual perspective allows observing and tackling
the privacy problem from a very non-concrete point
of view. A high-level architectural representation of
systems, in addition of delimiting the boundaries of
analysis, might allow identifying the critical elements
involved in communication processes. In Figure 3, we
provide an example of system representation from
this perspective. Vehicles, smartphones, laptops,
smart homes and their appliances, access points, and
users are some example of interconnected entities
within the network. Access points to communication
networks may be deployed in public or in private
areas. Privacy specifications might depend on
application fields and on protection objectives.
Figure 3: Communication technologies in the IoT.
For example, in smart home systems, privacy
objectives could be concealing presence or absence of
persons, consumption habits, and appliances installed
inside houses. In pay-as-you-drive insurance, black-
box car insurance, and car-sharing services, because
of routes and driver’s guide style monitoring, users
could be exposed to linkability, identifiabiliy and
disclosure of information threats (to give just some
examples). For the listed cases, avoiding fine-grained
information communication (also via secure media)
might reduce the risk of private information
disclosure. When a device communicate sensed data
to a remote service (see Table 1), linkability,
identifiability, detectability, and disclosure of
information threats might violate the system. Issues
might derive from devices settings and from
identifiability of remote services to which they
connect. Problems might become more serious when
the aforementioned settings and service are set by the
manufacturers and cannot be altered by end users. As
an instance, it could be possible, through traffic
analysis, identifying installed smart appliances within
a home. Hence, adversaries, by exploiting known
vulnerabilities of them, could steal or infer private
users’ information.
The just discussed problems may be reported to
the more specific conceptual perspective. Both wired
and wireless communications can be considered,
though we only deepen the latter because more
exposed to eavesdropping. The communication
infrastructure taken into consideration is multi-tier,
ultra-dense and heterogeneous. Several wireless
communication technologies and protocols can be
analyzed, such as IEEE 802.11x, IEEE 802.15.x,
WiMax, ZigBee, and LTE/LTE-A/LTE-U.
Attackers, to launch an attack against users’ privacy,
IoTBDS 2018 - 3rd International Conference on Internet of Things, Big Data and Security
128
Table 2: In this table, we report the definitions of some entities, protocols, and procedures involved during the UE-to-Core
network communications (LTE).
# Symbol Description
Paging
Refers to the process in which the mobility management entity (MME) needs to locate an UE in a
particular area and to deliver network services, such as incoming calls.
Radio Resource
Control
(RRC)
Includes a set of functions to manage connectivity between UE and eNB, that is broadcasted
information (sent by eNBs over a broadcast channel) and UE measurement reports or radio link
failure (RLF) sent by UEs
Access Stratum (AS)
Is a functional layer within LTE protocol stack, responsible for radio resource management and
data transportation over the wireless channel
Access Stratum
Security Context
The purpose of AS security context is to delive
r
RRC messages between an UE and an access
point (eNB) through the control plane, and IP packets through the user plane using AS security
keys.
Radio Link Failure
report
(RLF)
It allow detecting connection failures caused by intra-LTE mobility and intersyste
m
handovers
between LTE, GSM, and 3G networks.
Measurement report
It includes throughput measurements, latency, reference signal received power (RSRP), received
signal strength indicator (RSSI), as well as information about dropped calls and, sometimes,
latitude and longitude.
could exploit weaknesses and known vulnerabilities
in wireless technologies and in devices’ interfaces to
wireless networks. For instance, in LTE networks the
subscribers' unique identifier is masked by using
pseudonyms. Since during paging processes (as
described in the next section) such information can be
broadcasted in clear text, adversaries can link
subscriber identity to a specific geographical area. In
order to protect them from being identified, such
associations should neither be publicly available nor
it should be possible infer any correlation among
them. Even if generally useful for protecting security
and privacy of information, as the cellular network
evolves toward the ultra-dense paradigm, current
cryptographic approaches (as described before, in
Section 2.2) might not be still suitable to satisfy
protection requirements or could not be implemented
due to devices’ limited characteristics. Cryptographic
operations should be lightened, leading, however, to
a reduction of effectiveness protection. Indeed,
simplifying current cryptographic techniques might
allow even non-powerful adversaries to succeed in
breaking cryptography.
4.2 Logical Perspective
The Logical is further specialized and require more
effort to be analyzed with respect to the previous
viewpoints. For this reason, in this paper, only LTE
communications are considered. Exploiting LTE
network as a part of the IoT communication
infrastructure, besides of producing economic
benefits and providing pervasive connectivity, also
offers security of communications since it integrates
various authentication and encryption algorithms
(e.g., EPS AKA, SNOW 3G, MILENAGE). As
asserted in (Shariatmadari, et al., 2015), some
security arrangements to the IoT might include
embedding a SIM card into devices. Anyway, but
despite this, devices’ security and privacy might be at
risk. Indeed this means sending signal measurements
to a central unit (i.e. a server), thus making them
radio-frequency finger-printable. In this section, we
provide some background about LTE
communications. Let us briefly introduce the concept
of access stratum (AS) security. The AS security keys
are generated every time a new radio link is
established (that is when a mobile device moves from
IDLE state to CONNECTED state). When the AS
security setup is completed, the mobile device (UE)
and the eNodeB (eNB) share an RRC integrity key,
an RRC encryption key, and a user plane encryption
key. In order to locate an UE and serve him/her with
network services, the network can trigger Paging
Messages (see Table 2 for further details). The
Mobility Management Entity (MME) generates a
paging message and forwards it to several eNBs
within a tracking area (TA). Thus, all eNBs within
the paged TA broadcast a radio resource control
(RRC) paging message to locate the UE (3GPP,
2016). Paging messages contain identities of UEs
such as serving temporary mobile subscriber
identities (S-TMSI(s)). S-TMSI is a temporary
identifier and it is part of a global unique temporary
identifier (GUTI). When it is in the IDLE state, the
UE decodes RRC paging messages and searches for
its IMSI in it. If its IMSI matches, it initiates a new
Attach procedure to receive a GUTI. RRC messages
IoT Privacy in 5G Networks
129
also indicate UEs which information it should be
returned in response (either Measurement report or
RLF report). Narrowband IoT (NB-IoT) functionality
is specified in the LTE technical specifications
(3GPP, 2016). In (Ratasuk, et al., 2016) authors
describe two optimizations introduced for small data
transmission, namely the RRC connection
suspend/resume procedure and data transmission
using control plane signaling. As reported by NB-IoT
specifications (3GPP, 2017) M2M communications
are not provided with measurements reporting and
handover management. Until serving eNB does not
release the connection or a link failure happens, UEs
stay in the connected mode. When the connection is
interrupted, they go to the idle state and then trigger
RRC connection reestablishment procedure. Paging
processes, if triggered when users are in IDLE state,
could allow relating IMSIs and GUTIs to TAs (Kune,
et al., 2012) (3GPP, 2017). In fact, in its first phase,
RRC paging lacks encryption protection (Shaik, et al.,
2015). Moreover, correlations among TAs and eNBs
can be disclosed.
In summary, our work has presented a
methodology of analysis for identifying privacy
threats in the IoT with a view to 5G networks
implementation. The paradigm shift of wireless
networks toward the 5G evolution will result in
employment of ultra-dense networks as to provide,
among all benefits, high data rate and low
communication latencies. Anyway, this network
transformation may seriously undermine, to some
extent, the privacy of devices and users. The proposed
methodology extends and the LINDDUN
frameworks by introducing temporal and location
information to the threats identification process.
Moreover, taking a cue from the popular Zachman
framework, it also addresses the privacy weaknesses
identification by investigating the entangled IoT from
four different points of view, namely contextual,
conceptual, logical, and physical.
The current paper lacks a comparative evaluation
and validation. Anyway, we planned to provide these
enhancements in the future.
5 CONCLUSION
We have presented a privacy assessment
methodology, which aims at discovering privacy
threats in the IoT through a systematic approach. Our
technique extends the LINDUNN framework by
introducing temporal and location information to the
threats identification processes. Moreover, it draws
on from the Zachman framework, thus observing
privacy issues from various viewpoints.
An application example of our methodology has
been discussed. However, in order to be brief, it has
not been conducted a thorough investigation. We
foresee to provide supplementary information in
future works. Further studies, which consider
different IoT architectures, will need to be
undertaken. Although our approach has been thought
to a specific case of the IoT, hopefully, it could be
also applied to other technological systems in which
privacy is critical. The prospect of being able to do
deliver secure and privacy-preserving services in
many contexts, serves as a continuous stimulus for
future research.
REFERENCES
3GPP, 2016. Evolved Universal Terrestrial Radio Access
(E-UTRA) and Evolved Universal Terrestrial Radio
Access Network (E-UTRAN); Overall description.
Issue TS 36.300.
3GPP, 2017. Evolved Universal Terrestrial Radio Access
(E-UTRA); Radio Resource Control (RRC); Protocol
specification. s.l.:s.n.
Bailey, D. A., 2012. Moving 2 mishap: M2M’s impact on
privacy and safety. IEEE Security Privacy, 10(1), pp.
84-87.
Chen, S. et al., 2017. Machine-to-Machine Communica-
tions in Ultra-Dense Networks-A Survey. IEEE
Communications Surveys & Tutorials.
CISCO, 2016. Cisco Visual Networking Index: Global
Mobile Data Traffic Forecast Update, 2015--2020
White Paper. Cisco Public Information.
Kende, M., 2014. Internet society global internet report
2014. Internet Society, 15 July.
Kune, D. F., Koelndorfer, J., Hopper, N. & Kim, Y., 2012.
Location leaks on the GSM Air Interface. s.l., s.n.
Malina, L., Hajny, J., Fujdiak, R. & Hosek, J., 2016. On
perspective of security and privacy-preserving
solutions in the internet of things. Computer Networks,
Volume 102, pp. 83 - 95.
Mannilthodi, N. & Kannimoola, J. M., 2017. Secure IoT:
An Improbable Reality. s.l., SciTePress, pp. 338-343.
Mylonas, A. et al., 2012. Smartphone forensics: A proactive
investigation scheme for evidence acquisition. s.l.,
Springer.
Porambage, P. et al., 2016. The Quest for Privacy in the
Internet of Things. IEEE Cloud Computing, 3(2), pp.
36-45.
Ratasuk, R. et al., 2016. Overview of narrowband IoT in
LTE Rel-13. 2016 IEEE Conference on Standards for
Communications and Networking (CSCN), pp. 1-7.
Sfar, A. R., Natalizio, E., Challal, Y. & Chtourou, Z., 2017.
A Roadmap for Security Challenges in Internet of
Things. Digital Communications and Networks.
IoTBDS 2018 - 3rd International Conference on Internet of Things, Big Data and Security
130
Shaik, A. a. B. R., Asokan, N., Niemi, V. & Seifert, J.-P.,
2015. Practical attacks against privacy and availability
in 4G/LTE mobile communication systems. arXiv
preprint arXiv:1510.07563.
Shariatmadari, H. et al., 2015. Machine-type
communications: current status and future perspectives
toward 5G systems. IEEE Communications Magazine,
53(9), pp. 10-17.
Swetina, J. et al., 2014. Toward a standardized common
M2M service layer platform: Introduction to oneM2M.
IEEE Wireless Communications, 21(3), pp. 20-26.
Wuyts, K., 2015. Privacy Threats in Software
Architectures. s.l.:Ph.D. dissertation, KU Leuven.
Yang, N. et al., 2015. Safeguarding 5G wireless
communication networks using physical layer security.
IEEE Communications Magazine, 53(4), pp. 20-27.
Yaqoob, I. et al., 2017. Internet of Things Architecture:
Recent Advances, Taxonomy, Requirements, and Open
Challenges. IEEE Wireless Communications, 24(3), pp.
10-16.
Zachman, J. A., 1987. A framework for information
systems architecture. IBM systems journal, 26(3), pp.
276--292.
Zhou, J., Cao, Z., Dong, X. & Vasilakos, A. V., 2017.
Security and Privacy for Cloud-Based IoT: Challenges.
IEEE Communications Magazine, 55(1), pp. 26-33.
IoT Privacy in 5G Networks
131