The 4th International Congress on Technology, Communication and Knowledge (ICTCK 2018)

January 24-25, 2018, Mashhad Branch, Islamic Azad University, Mashhad, Iran

Security Analysis of Big Data on Internet of Things

1st Saeed Banaeian Far 2nd Azadeh Imani Rad
Department of Electrical Engineering, Department of Electrical Engineering,
Yadegar -e- Imam Khomeini (rah), shahr-e-rey Branch Yadegar -e- Imam Khomeini (rah), shahr-e-rey Branch
Islamic Azad University Islamic Azad University
Tehran, Iran Tehran, Iran
[email protected] Azadeh [email protected]

Abstract—The volume of data exchange in this network is sidered the sensors and radio frequency identification
ascending, by expansing of internet of things and its increasing systems (RFID) as the most important resources for the
familiarity in recent years. By increasing of requests for joining generation of BD [7]. In 2016, Jacobik investigated some
to the network and taking advantage of its services, necessity to
maintain the privacy and security is felt more than ever and the security challenges and privacy settings of the BD [8].
whole thing has changed into a challenge. Keeping the security of In the same year Ye et al., investigated and presented a
the users in small networks seems to be simpler and the threats classification of attacks in various sectors of BD. In 2017.
more predictable. As the users of the network which are sensors, Khan presented the utilization of the fourth generation
the volume of data increases as well. As a result the routers of the of industrial revolution of BD [9]. In this scheme, the
network and network servers has become a seriously challenging
task to control and maintain the security for this volume of data. IoT which is one of the generating factors of BD plays
We analyze these challenges in this paper. This paper aims to an important role which has been shortened as Industrial
maintain the security of the users and present procedures to Internet of Thing (IIoT). In the same year, Sivaraja has
dominate the problems and express the strategies to overcome expounded some challenges related to data, processing,
them. Some methods such as using both encryption function and data management, and afterwards has investigated some
secure protocol are proposed. Other method to overcome over
big data challenges is smart design of communication protocol. types of BD [10].
Finally, we point out some future challenges of big-data on the • Our contribution: To overcome on BD challenges, we
internet of thing network. suggest several methods, as follows:
Keyword: Big data, Challenge, Internet of Things, Wireless
– Using encryption functions
Sensor Network
– Network security
I. I NTRODUCTION – Protecting programs
– Using secure protocol
Not only the number of individuals connected to the internet
– Using from lightweight encryption functions
increases day by day, but also the same thing happens to the
– Smart design of communication protocols
things connected to the internet to exchange data and interact
with each other. In modern era, smart devices and things In the section 5, we discuss about the mentioned methods.
could transmit and receive information as well as interact with • Paper organization: This paper, the pre-requisites of
each other. This interaction has reached a level that a major work security and communication in IoT and also BD
volume of data exchange through things is related to Internet would be expounded after presenting some required def-
of Things (IoT). It is forecasted that the number of smart things initions in the section 2 and further scientific familiar-
connected to internet in the year 2020 is going to increase 44 ization with the fundamentals of IoT principles as well
times comparing to 2009 [1]. The sensors could be enumerated as work security and communication in the network of
as the dominant things connected to internet exchanging data, IoT and also, the BD. Afterwards, in the section 4 of
considered as one of the greatest resources of data [2]. IoT the article the existing challenges in BD communication
shall be considered as a smart network consisting of things in IoT would be expounded. In the section 5 of the
interacting with each other [3]. article, some methods to overcome the challenges would
Things in the network produce and exchange data having a be expressed from the viewpoint of network security.
great volume, velocity, and variety. A term titled big data Finally, after the challenges facing to BD communication
(BD) has been coined for the data embodying the above in IoT would be expressed.
mentioned characteristics. BD has no clear boundary and is
usually classified based on its specifications [4], [5]. II. P RELIMINARIES AND D EFINITIONS
• Related work: In 2015, the importance of BD was Before entering detailed discussion of IoT more familiar-
investigated in the industries and got to the conclusion ization is needed. Concepts such as the network we discuss,
that complexity is the most important challenge of BD instruments utilized or cases which faced with. In continuation,
[6]. Another survey carried out in the same year, con- a brief description of IoT is needed to be explained.
 The 4th International Congress on Technology, Communication and Knowledge (ICTCK 2018)
January 24-25, 2018, Mashhad Branch, Islamic Azad University, Mashhad, Iran
TABLE I the powerful servers of companies like Google and Amazon
S OME BIG DATA PRODUCERS [1] supported these servers [11]. The need for cloud servers in
IoT network is strongly felt, since this network involves light
Source Data Type and Size
devices and they can not process and store BD.
YouTube 1. Upload over 100 video/min
2. 1 billion access /mon
D. Big Data Application on Internet of Things
3. Download 6 billion hour.video/mon
Facebook 1. 35 billion likes /min The role of BD in IoT or the role of IoT in BD? In order
2. Upload 100 TB of data /day to reply this question it could somehow be told that the two
3. 1.4 billion of users phrases are interdependent together. The clear sample of BD
Twitter 1. 645 billion users could be observed through BD in IoT network. Also, several
2. 175 billion twits /day years ago, social networks such as Facebook were considered
Google+ 1. 1 billion users as giant producers of BD, and the social media have still a
Google 1. 2 million search /min noticeable role, but predictions show that in a near future, IoT
2. Process 25 PB of data /day gains the highest share of exchanged information. As men-
Apple 1. 47000 application download /min tioned above, future attitude toward IoT focuses on large-scale
smart cities [14]. For example, in 2015, Sun presented a smart
city in four parts: 1)Viability, 2)protection, 3)Rejuvenation,
A. Internet of Things and 4)Stability. In this scheme, the lowest level of architecture
IoT is said to all things which are able to transfer or receive appropriate for rendering services is transmission of informa-
information and could interact with each other. IoT could be tion through sensors. After investigating information in the
termed as an expanded and pervasive in all fields. This network scheme, the time for client services is announced.
is capable of gathering, processing, transmitted data analyzing From the other applications is the transfer of industrial infor-
of all network elements. IoT predicts that in future the world mation. The fourth industrial revolution in the world in 2011
would be controlled by physical things while are connected was the apex of IoT usage. Before that it was through robots,
to each other through a single infra-structure3 [3]. In other computers and chips. It was here that IoT entered to assist
words, IoT is a giant network in which a huge mass of things, industry. The sensors which have the highest share in IoT
sensors, and smart devices are interactive with each other [11]. started to increase [15]. Along with this increase, the volume
Most of this network is made up by sensors. But, there are of exchanged controlled information ascended as well.
other entities like RF ID, smart card, smart phone, and the Another connection between IoT and BD, gathering and envi-
computers which are share the required information. ronmental information, GIS and astronomy through wireless
sensors of IoT. With the increase of IoT, this information.
B. Big Data Ascends as well [16].
No clear and transparent definition could be provided from Philip Chen has expounded another type of BD dependence
BD [5]. But as evident from the term big data which refers to IoT [13]. Navigation, Social media, financial informa-
to data containing high volumes of data, the most important tion, Health information, astronomy information, and smart
point is the volume of data [13]. Many definitions have been transportation are among industries which produce lots of
presented which are dependent upon the objective and research information.
area. At the moment, 2 billion individuals are connected to
internet and 5 billion use cellular phones. It is expected that III. R EQUIREMENTS
the number of devices connected to the internet exceeds 50
With demand increase in social media and information. Ex-
billion devices by 2020. It is predicted that the number of
change in these networks, and also the expansion and pervasive
messages transmitted by this year (2020) exceeds 44 times the
IoT in all fields, the malicious and fraudulent individuals who
number of messages transmitted by 2009, the imagination of
intended to cheat started to increase as well. Therefore, to
this volume of change is thoroughly difficult. The rate of BD is
protect users (individuals/things) whose security is of prime
not clear, but some BD producers are You-tube, Facebook and
importance started to protect their information [8], [17], [18].
Twitters [1]. Table I shows a number of other BD producers.
So, the two important factors are as follows:
Nowadays, wireless sensor networks could be considered as
BD producers as well [2], [7]. Predict that IoT network has
A. Privacy
the largest shared document, in future.
Users’ private security is indicate the level of accessibility
C. Cloud Server and Cloud Computing to other users. People can access to others’ private information
The information systems based on IoT are stored and [19]. Some researchers have divided the privacy information
processed in servers which have ultra-high strength and power. into three sections of: infrastructure security, information secu-
Cloud servers collect their information from sensors, RF IDs rity and information management [9]. Each user (individual or
and other smart devices and store them on their memory device) can give access to its sensitive information to others.
[12]. Cloud computations bore growth from industry and However, it can limit the access and closes its privacy.
 The 4th International Congress on Technology, Communication and Knowledge (ICTCK 2018)
January 24-25, 2018, Mashhad Branch, Islamic Azad University, Mashhad, Iran
B. Security • Value: The value of data shall be kept and maintained.
High volume and velocity makes security difficult. This None of it shall be deleted. It is very important that the
could be investigated from. Users, service rendering centers value of data does not lost/change.
and computer networks. BD security is quite complicated and • Data acquisition: This is a challenging job related to
still not thoroughly known. Also, technology speed and infor- information. Gathering from different sources and keep
mation volume makes it more difficult. BD security is quite them secure.
complicated and still not thoroughly known. Also, technology • Data Mining and Cleansing: One of the greatest chal-
speed and information volume makes it more difficult [8]. lenges, is to cleanse the giant pool of info from additional
Lots of worries exist on information Security and data. As data data and choice of correct data required.
are stored in voluminous spaces, the accesses of fraudulent • Data aggregation and integration: This challenge shall
people is probable. Giant Companies like Google, Microsoft, be explained by one example: Imagine the giant so-
YouTube, Skype, etc. Have tried to devise various information cial media: Twitter. You should answer any tweet by a
Security methods which shows how perilous security protec- retweet. Finding the reply to a tweet in this great volume
tion is in giant servers [18]. of data. The most important is the answer to the right
Security threats consist of several sections. Dennial of service question which is the duty of twitter server.
(DoA) attack is formed in the infrastructure, and attack on • Complexity: As clear from other explanations, as data
encryption function and access control are formed in privacy volume increases, complexity boosts as well. This could
section [9]. Preventing users against these attacks are security be divided into three parts 1)data complexity, 2)compu-
challenges. tational complexity, and 3)system complexity.
• Data analysis and modelling: This challenge points out
IV. T HREATS AND C HALLENGES to some type of sub sets: separation of information, and
This is the most important part of our paper. Since it gathering lost data.
deals with the challenge of protection. In this section could • Data interpretation: This step looks like image building
be divided into gathering of BD, BD analysis and use of step important for received data decision.
BD. These include: subdivisions as BD challenge, processing • Privacy: As mentioned before, privacy is the most chal-
challenge and BD management challenge [21]. Please note lenging task in digital age. In the section 3, we discussed
that: We have pointed out to current challenges. In the section the importance of this feature.
5 focuses on how to overcome on challenges related to security • Security: Keeping users’ privacy and preventing from
and privacy [4], [6], [8], [10], [13], [20]. We describe some spread of malware. According to the importance of pro-
features of BD in the blow: cessing of BD, their security should also be considered. It
• Volume: Volume of data (terra byte and more) is a means that the attacker and malicious user have no ability
great challenge in BD. The diversity of the type of to access the users’ sensitive information.
information environmental information medical info and • Data governance: As BD is increasing, companies and
business information. Facebook, for example produces Organizations shall potentially try to manage the data
500 terra byte data every day. As it is known from the properly, data quality guarantee, data quality improve-
name BD, the volume counts. We do not set boundaries ment and keeping the value of the data is among key
for the BD, but it is not low-scale data. elements in information.
• Variety: The data are different. For example the corre- • Data and information sharing: All Organization, Shall
sponded data by sensors are many variation (e.g. environ- coordinate in sharing information.
ment information, sound, image, data, and even noise). • Cost: Data cost are increasing as demands increases. All
• Velocity: As mentioned before pointing to diversity and throughout the world client data shall be supported. For
complexity of data structure, if high speed and volumi- example, Google supports its clients in 13 centers around
nous data is added, processing becomes a challenging job. the world.
Velocity is another feature of BD and its challenge. The • Data Ownership: In addition to security and privacy,
processing of high-velocity data is a challenge. data ownership is important as well, which shows itself
• Variability: Usually users transmit different data. when sharing.
Google, for instance, receives diverse data from different
users and different sources. in some other resources V. M ETHODS TO OVERCOMING ON C HALLENGES
variability of data is among the four original challenges.
We depict a 4V in the figure I. As discussed, BD has a lot of challenges such as 1)security
• Veracity: The structure of data is quite complicated and and privacy, 2)data management, and 3)processing challenges.
in BD confidence cannot be established. In this paper, we outline the security and privacy implications
• Visualization: Information shall be understandable and of some ways to overcome on these challenges in the Internet
eligible. From every sender key information shall be of Things. The most basic and primitive goal in security
available. For example, eBay has many clients whose info and privacy is to provide all three features of Confidentiality,
have to be received. Antegrity, and Authentication (CIA) [17].
 The 4th International Congress on Technology, Communication and Knowledge (ICTCK 2018)
January 24-25, 2018, Mashhad Branch, Islamic Azad University, Mashhad, Iran

Fig. 1. The 4 original challenges of big data (4v) [16]

a) Using encryption functions: Encountering giant net- of its calculation on the server servers and users (objects)
works like IoT functions and protocols shall be used to comply is powerful. So that poor network elements such as sensors
with all things. Many devices have limited processing power carry less computational burden.
and cannot be encrypted through keys. Many devices have low
power. But today many methods are devised for encrypting VI. C ONCLUSION
which is both speedy and secure and low cost [10], [17]. In this paper, we reviewed and analyzed some of the
b) Network security: Protection techniques are vital [10]. projects on the IoT with the BD mining approach and we have
c) Protecting programs: Program protection is vital, for reviewed the safety and privacy of the users. Eventually, we
there are attacks to steal information. There are also many IoT used several methods, such as the use of encryption functions
programs which are still nascent in ID, digital signature and and protection programs. It is absolutely clear that encryp-
Each has strength and weaknesses. Sometimes through simple tion functions Bring confidentiality. After that, we proposed
changes the program could be practically useful [10]. methods that could provide a safer space for low-power users.
d) Using secure protocol: To protect users’ privacy The most important of our proposed methods is the smart
and protect them on the network, secure protocols (and design of communication protocols, so that the computing
lightweight) should be used. There are many protocols for and processing load will be borne by service providers (cloud
transferring information, user authentication, digital signa- servers). In this method, the light things (e.g. sensors) apply
tures, and so on. Each of them is presented for the purpose light operation. But, on the other side of protocol, the server
of the properties, each of which has strengths and weaknesses uses the same or other cryptographic function. So, the low-
that can be secured by analyzing them in detail. Sometimes, power things can run the protocol and be safe.
with a slight change, the use of a communication protocol
can be changed, or by modifying a protocol with a simple VII. F UTURE W ORKS
technique, it can be used in practice. Given the ever-expanding Internet network of objects
e) Using from lightweight encryption functions: Using in all areas, researchers are thinking of the future of the
cryptographic functions is one of the best ways to protect network. We know that the most important part of the data
information. But since the future targets are clear, the largest is large data aggregation, which should be processed after
share of future information belongs to sensors and RFIDs. processing and other actions on them. We should know
Therefore, the use of light cryptographic functions is very that the most exchanged data of this network is more than
important. Since most of the elements on the Internet are that group in the future. Some scholars will map out the
objects of sensors or smart cards, they can be a symmetric future generations of macroeconomic data into 1) online
cryptographic key in their production plant. data networks, 2) cellphone and Internet data objects, 3)
f) Smart design of communication protocols: geographic information, 4) temporary space data, And 5) flow
Lightweight protocols and the use of cryptographic functions data and many other data [22]. Methods must be considered
of the network security network will not always be used, for the collection, processing and classification of these
and sometimes it is necessary to use asymmetric encryption, categories.
digital signing, and other encryption functions that greatly One of the most important parts of the today world is
increase the computing load of the network. Therefore, it industrialize. The industrialization of societies is on the risk,
is necessary to design a protocol that is the main burden and this trend will continue, and people will be replaced by
 The 4th International Congress on Technology, Communication and Knowledge (ICTCK 2018)
January 24-25, 2018, Mashhad Branch, Islamic Azad University, Mashhad, Iran
cars. The vast Internet of objects in the industry can not [9] Ye, Haina, et al. ”A survey of security and privacy in big data.” Commu-
be ignored, and in the industry, the Internet of Things will nications and Information Technologies (ISCIT), 2016 16th International
Symposium on. IEEE, 2016.
play a significant role. The expansion of the industry, the [10] Sivarajah, Uthayasankar, et al. ”Critical analysis of Big Data challenges
spread of information, as well as the increasing of data sent and analytical methods.” Journal of Business Research 70 (2017): 263-
by the agents in it. Sensors, drives and other components 286.
[11] Hashem, Ibrahim Abaker Targio, et al. ”The rise of “big data” on cloud
can be referred to the factors that are constantly sending computing: Review and open research issues.” Information Systems 47
and receiving data [15]. Securing this volume of equipment (2015): 98-115.
and devices for sending and receiving information is very [12] Cai, Hongming, et al. ”IoT-based big data storage systems in cloud com-
puting: Perspectives and challenges.” IEEE Internet of Things Journal
important. Failure to send and receive one of these data 4.1 (2017): 75-87.
may result in a serious problem with production. Therefore, [13] Chen, CL Philip, and Chun-Yang Zhang. ”Data-intensive applications,
maintaining the security and health of industrial data in the challenges, techniques and technologies: A survey on Big Data.” Infor-
mation Sciences 275 (2014): 314-347.
present and future will be very important. [14] Sun, Yunchuan, et al. ”Internet of things and big data analytics for smart
It is anticipated that the number of objects of the Internet and connected communities.” IEEE Access 4 (2016): 766-773.
service provider will reach a billion by 2030, so companies [15] Khan, Maqbool, et al. ”Big data challenges and opportunities in the
hype of Industry 4.0.” Communications (ICC), 2017 IEEE International
in this field will be researching. According to HP and Intel, Conference on. IEEE, 2017.
large data management and processing should have three [16] Chen, Min, Shiwen Mao, and Yunhao Liu. ”Big data: A survey.” Mobile
features. First of all, there should be a lot of powerful and Networks and Applications 19.2 (2014): 171-209.
[17] Cheng, Chi, et al. ”Securing the Internet of Things in a Quantum World.”
high-capacity terminals for data gathering, second, the data IEEE Communications Magazine 55.2 (2017): 116-120.
produced by the Internet elements of objects is not often [18] Dubey, Arunima, and Satyajee Srivastava. ”A Major Threat to Big Data:
complete, and it should be possible to analyze them properly. Data Security.” Proceedings of the Second International Conference on
Information and Communication Technology for Competitive Strategies.
Finally, the information gathered from the elements of the ACM, 2016.
Internet of objects is only effective when analyzed [16]. [19] A. Adrian TOLE, ”Big Data Challenges,” Database Systems Journal,
The use of electronic devices is increasing day by day, and vol. 4, pp. 31 - 41, 2013.
[20] Bertino, Elisa, and Elena Ferrari. ”Big data security and privacy.” A
as a result, all communities try to make the environment Comprehensive Guide Through the Italian Database Research Over the
more intelligent and use energy efficiently. Increasing the Last 25 Years. Springer International Publishing, 2018. 425-439.
use of electronic devices (sensors, smart devices and even [21] Jung, Jason J. ”Computational Collective Intelligence with Big Data:
Challenges and Opportunities.” (2017): 87-88.
cloud servers) will require more privacy. In Section 5, we [22] Song, Houbing, et al. ”Next-generation big data analytics: State of
used one of the ways to protect information and privacy the art, challenges, and future research topics.” IEEE Transactions on
using cryptographic functions. An important point is using its Industrial Informatics (2017).
[23] N. Ammu and M. Irfanuddin, ”Big Data Challenges,” Special Issue
computational load encryption functions. Designers should be of ICACSE 2013 - Held on 7-8 January, 2013 in Lords Institute of
careful about the privacy of users (information has been sent Engineering and Technology, Hyderabad, vol. 2, pp. 613 - 615, 2013.
and received by objects or individuals) and the optimal use
of energy in designing Internet systems and service servers.

Conflict of Interests
The authors declare that they have no conflict of interests.

R EFERENCES

[1] Khan, Nawsher, et al. ”Big data: survey, technologies, opportunities, and
challenges.” The Scientific World Journal 2014 (2014).
[2] Ding, Xuejun, Yong Tian, and Yan Yu. ”A real-time big data gathering
algorithm based on indoor wireless sensor networks for risk analysis of
industrial operations.” IEEE transactions on industrial informatics 12.3
(2016): 1232-1242.
[3] Rahman, Abdul Fuad Abdul, Maslina Daud, and Madihah Zulfa Mo-
hamad. ”Securing sensor to cloud ecosystem using internet of things
(iot) security framework.” Proceedings of the International Conference
on Internet of things and Cloud Computing. ACM, 2016.
[4] Gani, Abdullah, et al. ”A survey on indexing techniques for big data:
taxonomy and performance evaluation.” Knowledge and Information
Systems 46.2 (2016): 241-284.
[5] C. Perera, R. Ranjan and L. Wang, ”Big Data Privacy in Internet of
Things Era,” Internet of Things (Mag), pp. 32 - 39, 2015.
[6] Jin, Xiaolong, et al. ”Significance and challenges of big data research.”
Big Data Research 2.2 (2015): 59-64.
[7] Kang, Yong-Shin, et al. ”MongoDB-based repository design for IoT-
generated RFID/sensor big data.” IEEE Sensors Journal 16.2 (2016):
485-497.
[8] A. Jakobik, ”Big Data Security,” Computer Communications and Net-
works, vol. 12, pp. 241 - 261, 2016.