Cryptography Arbitration: Security Complexities of Cloud Enabled
IoT in Europe and Beyond
Morgan Eldred, Hassan Alnoon and Sultan AlTamimi
School of Computing, University of Portsmouth, Portsmouth, U.K.
Keywords: Arbitration, Cloud, Cryptography, Data Protection, Internet of Things (IoT).
Abstract: The global nature of the Internet of Things and cloud has and will result in emerging challenges, such as
whom is liable for data protection and security breaches of personal data. This paper puts forward the concept
of cryptography arbitration’ and the need to design and architect legally compliant solutions. As the world
becomes more interconnected we are likely to see more example of technology practices sweeping the globe
and raising further data protection challenges; much like the fault lines between tectonic plates. This paper
provides contribution by capturing some emerging impacts and challenges and how they relate to the internet
of things and the need for solutions to have the appropriate cryptography safeguards.
1 INTERNET OF THINGS
There are many definitions of the internet of things
(IoT) such as:
A network of items-each embedded with sensors-
which are connected to the Internet (IEEE, 2014),
“the network of physical objects that contain
embedded technology to communicate and sense or
interact with their internal states or the external
environment” (Gartner). “The IoT links smart objects
to the Internet. It can enable an exchange of data
never available before, and bring users information in
a more secure way” (Cisco).
For the purposes of this paper the definition of
cloud enabled IoT taken is that it is it is a network of
physical objects that contain embedded technologies
to interact via the cloud with an external environment
and uses the below layered three-tiered architecture,
which is a modified lightweight version of IEEE
P2413 standard for an Architectural Framework for
the Internet of Things.
Figure 1: Cloud Enabled IoT.
This architecture consists of three critical layers:
Applications - are the software that controls,
monitors and provides the user interface for the IoT
application.
Cloud - is the network connectivity between the
applications and physical equipment, identified as
things.
Things - are the physical objects/device that
contain embedded technology to communicate, sense
and interact with the external format.
An example of the cloud internet of things, would
be a drone that is used by a retailer to deliver
purchased items. If the drone was to take a
photograph of a facial images or car registration plate
and inadvertently delivered that image by accident to
a third party, then data protection acts could be
breached, whom then would be liable and for what
extended would they be liable. Considering that it has
been indicated that the security of user’s data has the
highest priority and concern from users (Chang and
Ramachandran, 2016). In such an example, if legal
proceedings occurred, it could be beneficial to have a
security framework that is verified and vetted by a
security/cryptographic specialist. Lessons learned
from the cloud computing adoption framework
(CCAF) that has security suitable for business clouds,
based upon a three major security technologies:
firewall, identity management, and encryption based
on the development of enterprise file sync and share
356
Eldred, M., Alnoon, H. and AlTamimi, S.
Cryptography Arbitration: Security Complexities of Cloud Enabled IoT in Europe and Beyond.
DOI: 10.5220/0005922103560361
In Proceedings of the International Conference on Internet of Things and Big Data (IoTBD 2016), pages 356-361
ISBN: 978-989-758-183-0
Copyright
c
2016 by SCITEPRESS Science and Technology Publications, Lda. All rights reserved
technologies (Chang, Kuo, and Ramachandran,
2016).
2 CRYPTOGRAPHY
Cryptology is the art and science of making and
breaking secret codes. Most people are interested in
Cryptography which is the art of making secret codes.
Cryptography is required in securing communication
lines, whether it is wireless or wired communication
lines against leakage of private and confidential
information to undesirable individuals or parties
while ensuring the basic pillars of information
security are met which are confidentiality, integrity
and availability of data; and also the authentication
and non-repudiation of the individuals involved in the
communication are protected.
While using cryptography in IoT the information
security basics are important, it ensures that the
devices and signals sent over the network are not
compromised and ensures the safety and privacy of
the user (Ning and Liu, 2012). The confidentiality of
a signal being sent to a connected device over the
cloud is important and is enforced by making sure that
the signal is encrypted. If the signal sent is
compromised an adversary can easily understand the
type of signals being sent to the connected device, and
therefore be able to take control of the device in
question. Sending an encrypted signal can’t be useful
by itself, if an adversary can also change certain
values in the encrypted signal without the device
having a way of checking the integrity of the signal,
an adversary can try random changes until a device
reacts different and therefore understands more about
how the device mechanism works. However having
high security and integrity, with powerful
cryptography and protocols used to communicate
with the device would certainly cause delay in
transmitting the signal to the device and therefore
cause an issue with availability of the device
mechanism. It is important when building an IoT
communication between the different things in the
cloud to ensure synergy between the three security
pillars (confidentiality, integrity and accessibility)
(CIA) (Ning, Liu and Yang, 2013).
Authentication is also another important aspect of
information security that needs to be preserved in a
cloud IoT environment. Being able to authenticate a
device or a user is crucial to the communication that
happens in IoT, if the authentication processes is not
done in a proper cryptographic fashion using a strong
secure protocols then both the devices and the user
can be vulnerable. Communication protocols are
complicated to build, most of the sophisticated
attacks rely on weakness in the mathematical aspect
of the protocols, formal verification methods are
required to ensure such protocols are secure.
Weakness in authentication and integrity of the
communication can cause a repudiation of the
messages sent, this can cause legal aspects, such as
the leaking of personal data and that is why it is
important that a protocol is capable of preserving the
non-repudiation aspect of the communication. As
provided before in the cloud IoT use case, if the image
that was taken was not properly secured, what would
be the liability of the company that did not properly
secure the personal data of customers.
Cryptography Arbitration can be tricky due to the
complexity that arises from establishing and
implementing a cryptographic algorithm or protocols.
To fully understand the underlying aspect of the
cryptography a good combination of expertise in the
fields of mathematics, computer science, and
electrical engineering is required. To understand the
complexity of cryptography, it is classified into the
following categories:
Symmetric Key Cryptography, involves using
the same key for both the encryption and decryption
of data. Symmetric key crypto is typically used to
create fast stream or block cipher algorithms. Popular
symmetric ciphers are AES, 3-DES, TwoFish and
RC4 etc.
P
Figure 2: Symmetric Key Cryptography.
Asymmetric Key Cryptography, involves creating
two separate keys, one for encryption and one for
decryption of data. Asymmetric key crypto is
typically used for transferring keys and digital
signatures. Popular asymmetric ciphers are RSA,
ECC etc.
Figure 3: Asymmetric Key Cryptography.
Cryptography Arbitration: Security Complexities of Cloud Enabled IoT in Europe and Beyond
357
Hashing Algorithm, involves creating techniques
where we are able to map data of any size to data of
fixed size (Digest). This process should be
irreversible Hashing algorithms are typically used for
data integrity. Popular hash ciphers are SHA-256,
SHA3, MD5 etc.
Figure 4: Hashing Algorithm Cryptography.
Message Authentication Code, is similar to a hash
algorithm but provides the capability to authenticate
the digest. MACs are typically used for Data Integrity
and Authenticity. Popular MAC ciphers are HMAC,
CMAC, CBC-MAC etc.
Key Management and Key Exchange
Protocols, are an extremely important aspect of
cryptography (Martin, 2006). Key management
basically involves the generation, exchange, storage,
use and replacement of keys used in the
cryptosystems. The initial suspect in any
cryptographic failure of a system would be due to
weaknesses in the key management or the key
exchange protocols. Most of the algorithms
mentioned above have their source code and logic
published online, that is mainly because these
algorithm have been proven mathematically to be
secure with today’s available computational power.
Though the keys used in any cryptosystem has to be
well guarded and never be published publically or to
any individual who shouldn’t be part of that specific
system. Key management even becomes a bigger
challenge in IoT and cloud based solutions (NIST,
2013).
To do cryptography arbitration a person has to
understand how these different cryptographic
classification work together and the proper
implementation procedures. Since most of the
algorithms are published online, a crypto arbitrator
has to understand the different mechanism to
implement an algorithm, as a wrong implementation
can cause a serious risk to the system. For example,
the most used symmetric key cipher is AES, even
though AES is a very powerful cipher, its
implementation in both software and hardware can
cause the overall security of the system to fail. While
implementing AES you have to choose a mode-of-
operation which is a technique that helps you
implement the algorithm and deal with large amount
of data to be encrypted. There are several modes-of-
operation for example an ECB (Electronic Codebook)
mode-of-operation can be very vulnerable whereas
Cipher Feedback (CFB) can provide a better
implementation security (Eng, 2008). A popular
example that is used to generally help understand
implementation difference is the following diagram:
Figure 5: Cryptography differences.
Another attack on AES implementation which would
also relate to IoT is a hardware based attack, where
even a proper mode-of-operation usage can cause a
side-channel attack on the algorithm. A very simple
example of such an attack is power analysis attack,
where an adversary monitors the power consumption
of the device to understand more about the key or the
algorithm being used (Gurkaynak, Oswald and
Preneel, 2004). In the figure below a person doing a
power analysis attack can understand more about the
algorithm being used in the device by counting the
number sets and how they relate to the inner works of
the algorithm (algorithm rounds).
Figure 6: AES Power Consumption.
3 DATA COLLECTION AND
ANALYSIS
A short survey was conducted on a sample of 39 IT
managing professionals coming from a host of
industries such as retail, finance, manufacturing and
petrochemicals. The question asked the importance of
cryptography in cloud IoT on a rating scale of 1 to 5,
with 1 being very low and 5 being very high. The
participants selection criteria, was that they needed to
be involved with setting up the organisations cloud
strategy, substantial management scope (over $5
IoTBD 2016 - International Conference on Internet of Things and Big Data
358
million budget, and 50 people). The hypothesis was
that cryptography would be an importance aspect.
Table 1: Importance of Crytography in Cloud IoT.
Mean Standard
Deviation
P Value
3.33 1.108183277 0.002565532
The mean resulted in an average over neutral in terms
of the importance of cryptography, slightly moving
towards the importance of cryptography in cloud IoT.
The standard deviation was a little over one place,
while the p value indicated a null hypothesis.
The participants then answered if they had
engaged in a cloud iot project, of which the majority
(74.358%) said no and then if they would go to a
cryptography arbitrator of which a slight majority
(53.846%) indicated they would. The data collection
calculated yes as a one and no as a zero.
Table 2: Yes or No Survey Questions.
Question Response
MEAN STANDARD
DEVIATION
Yes No
Have you
engaged in
a cloud iot
project
10
29 0.25641 0.538
Is a crypto-
arbitrator
relevant
21 18 0.538461 0.505035
4 EUROPEAN DATA
PROTECTION
The current guiding European directive on the
protection of individuals with regard to the processing
of personal data and its free movement is the
Directive 95/46/EC dated 24th of October 1995. A
basic protection level of personal data is explicitly
included in the EU’s Charter of Fundamental Rights
(Article 8), but also in the Treaty on the Functioning
of the European Union (Article 16).
According to the Directive 95/46/EC the meaning
of personal data is “any information relating to an
identified natural person (data subject).” When it
comes to the processing of personal data, the
Directive 95/46/EC points to “any operation or set of
operations which is performed upon personal data,
whether or not by automatic means, such as
collection, recording, organization, storage, adaption
or alteration, retrieval, consultation, use, disclosure
by transmission, dissemination or otherwise making
available, alignment or combination, blocking,
erasure or destruction.”
The Directive identifies the subjects involved in
the processing of personal data as the controller,
processor, third party and the recipient. All subjects
refer to a natural or legal person, public authority,
agency or any other body, with the controller
determining the purpose and means of the processing,
the processor being the one that processes the
personal data on behalf of the controller, while the
recipient is the one to whom data is disclosed. The
third party refers to any other subject that is
authorized to process the personal data.
Data protection is enforced within the European
Union through the principle of applicable national
law. Under Article 4, the Directive 95/46/EC
specifies that each Member State shall apply the
national provisions defined pursuant to the Directive
for any controller that has activities of an
establishment set on the territory of the Member
State. If multiple European establishments exist, the
controller must comply with local requirements put
forth by each Member State. In the example used for
this paper, if the company was European it would be
liable under Directive 95/46/EC, and if based in the
UK, then it would also be liable under the Data
Protection Act, 1998.
5 ARBITRATION & CORPORATE
RULES
The current European Directive 95/46/EC on data
protection is outdated and does not take the impact of
cloud or IoT into consideration as it leaves a gap in
the data protection area that cannot easily be covered
through legislative efforts. Indeed, the European
Commission is actively engaged in reformation
processes that target to clarify the missing pieces,
while modernizing the legal framework, but this is
viewed as a long-term result.
Besides the extended duration of legislative
reforms, analyzed proposals could be implemented in
a very distinct format than the one when they were
initially proposed. Taking all these aspects into
consideration, it is imperative that at least a partial
clarification and solution be introduced while the
more advanced legislative process unfolds. “Binding
corporate rules (BCR)” is a term that is increasingly
used when it comes to international data transfers that
imply third countries. However it would not be
difficult to adjust binding corporate rules for external
organisations that will have physical devices that
Cryptography Arbitration: Security Complexities of Cloud Enabled IoT in Europe and Beyond
359
interact with each other via the cloud. In an adjusted
manner the approval of the binding corporate rules
given by one organization is also enriched with
expansion coverage power over all national
authorities in the light of the Directive proposal.
After a successful application of binding
corporate rules at the level of controllers, the Article
29 Data Protection Working Party advanced to
another level by adopting in June 2012 a working
document on binding corporate rules for processors,
both companies and data protection authorities. BCR
are viewed as “internal rules applicable to entities of
a multinational company and contain key principles
legally covering the transfers of personal data coming
from the European Union”. They are regarded as an
alternative to the Safe Harbor Principles and the
European Commission’s Standard Contractual
Clauses. When transcended at the level of processors,
binding corporate rules should be able to provide
clients with the security and privacy of their data
under European Union data protection regulations.
The Article 29 Working Party’s working document
provides processors with a conditions checklist that
must be fulfilled for being granted their adequacy.
The A29 DPWP working paper also came as a
response to the industry’s numerous requests to move
the usage of binding corporate rules at the level of
processors, as well. There are also voices that demand
BCRs to be included for community cloud,
considering that there might be cases when
community members that belong to different
corporate groups might own similar interests. Even
though improvements at the level of binding
corporate rules are definitely a step forward, their
approval process remains a long and expensive
procedure under the current regulations. While
Member States grant approval based on diverse
conditions, there is still a range of Member States that
tends to remain on the safe side requesting an
individual approval for each transfer under an already
approved BCR.
If applied on a large scale, binding corporate rules
could solve one of the main issues implied by both
adequacy findings and Safe Harbor compliance –
their restrictive geographical reach. In a July 2012
paper on cloud computing adopted by the Article 29
Data Protection Working Party, the organization
states that companies that export data should act with
increased diligence and question the statement of the
data importer that it owns a Safe Harbor certification.
Also, cloud clients should verify that standard
contractual terms comply with national requirements
regarding contractual data processing. Within cloud
IoT the same policies could work to include
cryptography arbitration. Whereby a third party is
engaged to determine if the appropriate technical
considerations where conducted in the design and
architecture of the system. This would need to be
agreed up front and drafted in end user license
agreements, to provide the appropriate level of
protection. Further challenges would come in the
form of determining what type of organization would
be certified as a cryptography arbitrator and what
standards would be legally sound from a quality
perspective, rather than just technical forensic
solutions.
6 CONCLUSIONS
This paper has put forward the concept of
cryptography arbitration and the need for its inclusion
within designing cloud iot solutions. The research
itself requires much more analysis to bring the
concept out, such as increasing the survey question
and sample frame, looking at other legal and
cryptography aspects and other use cases. However
the paper has put forward the position of the need for
cryptography arbitration, and has looked at the
security and legal challenges, providing
recommendations to learn from lessons learned from
existing cloud security frameworks, and in using
existing legal frameworks, such as corporate binding
rules.
REFERENCES
Chang, V., Kuo, Y.-H. and Ramachandran, M. (2016)
‘Cloud computing adoption framework: A security
framework for business clouds’, Future Generation
Computer Systems, 57, pp. 24–41. doi:
10.1016/j.future.2015.09.031.
Chang, V. and Ramachandran, M. (2016) Towards
achieving data security with the cloud computing
adoption framework. Available at:
http://ieeexplore.ieee.org/xpl/abstractMetrics.jsp?arnu
mber=7299312 (Accessed: 17 March 2016).
Ning, H. and Liu, H. (2012). Cyber-Physical-Social Based
Security Architecture for Future Internet of Things.
Advances in Internet of Things, 02(01), pp.1-7.
Ning, H., Liu, H. and Yang, L. (2013). Cyberentity Security
in the Internet of Things. Computer, 46(4), pp.46-53.
NIST, (2013). Cryptographic Key Management Issues &
Challenges in Cloud Services (NISTIR 7956).
Martin, K. (2006). Cryptographic Key Management. Eng,
C. (2008). Cryptography for Penetration Testers.
Gurkaynak, F., Oswald, E. and Preneel, B. (2004).
IoTBD 2016 - International Conference on Internet of Things and Big Data
360
Power-Analysis Attack on an ASIC AES
Implementation. In: Information Technology: Coding
and Computing. Leuven, Belgium: IEEE.
IEEE, Retrieved from http://iot.ieee.org/about.html on
05/01/2016.
Gartner, Retreived from http://www.gartner.com/it-
glossary/internet-of-things/) on 05/01/2016.
Cisco, Retreived from http://www.cisco.com/c/en/us/
solutions/internet-of-things/overview.html on 05/01/
2016.
Cryptography Arbitration: Security Complexities of Cloud Enabled IoT in Europe and Beyond
361