Artifical Intelligence for Security of Smart Cities: A

Survey

Abstract — The interest in creating a technologically Biological information analysis has taken a sharp turn in
smarter and safer planet in which to live has been created by the past two decades, contributing to a number of advances in
significant technical, economic, and environmental changes. In genetics and the identification of artificial trends. Advanced
the development of cognitive abilities, brains such as intelligent signal processing and ICA technologies have greatly
reasoning machines with highly responsive sensors and enhanced the Infomax Individual Component Recognition
automatic telecommunications (nerve system), human intellect System [3]. The current climate has stimulated the
coupled with artificial intelligence became incredibly useful. development of the societal, technological, and economic
Smart cities would not only improve people's lives but also
worlds to push quicker than the normal developmental
render civilization safer and more unified developed and
processes. Technology developments in computer to
maintained by artificial intelligence. Smart City is a
metropolitan environment that uses various types of sensors to
computer and data analytics could be the only technologies
collect data and then use artificial intelligence for the effective that will keep pace with such shifts, building a future of
management of assets, infrastructure, and services. This intelligent and safer societies. We are now considered to be
involves knowledge collected from people, equipment, and in the era of global culture, so regional approaches are
resources that are being stored and evaluated for expected at the local level. This is accessible to all the
communication tracking and regulation, power plants, utilities, communities over a free cloud network and helps individuals
waste treatment water delivery networks, crime detection, to invest in modern smart tools [4].
computer systems, colleges, libraries, hospitals, and other
public services. This article offers an overview of how such II. SMART CITIES
concepts are to be implemented. In order to optimize the efficiency of urban
Keywords — Artificial Intelligence, Sensors, & Smart
structures and services and communications with people, the
Cities smart city definition combines knowledge and
communication technology (ICT) with a range of Artificial
I. INTRODUCTION Intelligence semantics. City leaders collaborate directly with
the neighbourhood and social network to see what is
In 1979, the word Artificial Intelligence (AI) was coined changing with the community and how the region is rising.
by John McCarthy as 'Technology and Engineering in the Smart Cities are used to improve the efficiency, output, and
Production of Smart Machines.' Artificial Intelligence has a
engagement of local services, optimize savings and
lot to do with the creation of sophisticated robotics systems
utilization of capital, and maximize public-government
that could address problems explicitly and intelligently [10].
communications. Smart transportation networks are built to
The categorizations of a smart city include: manage traffic flows and have real-time responses. Thus, a
• The infrastructure smart city will be more likely to encounter challenges than
an easy, "transactional" society [14].
• The energy
• The transportation and TECHNOLIGICAL ASPECTS
Important technological, economic, and
• The lifestyle of people environmental advances have given rise to concern in smart
Through better gathering and sharing information all four cities, including climate change, urbanization, online
industries with connected networks and urban culture will be shopping, and entertainment, changing populations,
more efficiently served and the foundation of the community increasing population growth and pressure on public
will be reinforced. Smart Cities and Artificial Intelligence finances. The European Union (EU) has made unwavering
(AI) are thus essential computing tools for the development attempts to establish a strategy to promote 'smart' economic
of smart society. growth in its metropolitan city-regions. The EU has set up a
range of projects under the 'European Digital Strategy.' The
Artificial intelligence is an area in computer science that
seeks to render computers behave like humans, mainly planet is at an alarming stage of urbanization. This high pace
improvise analytical abilities in machines and create of urban population growth is not a fascinating reality but
intelligent agents. These smart machines realize the calls for economic development and a better existence. By
environment around them and take other acts that improve the 18th century, fewer than 5% of the world's people
their likelihood of performance. Several AI approaches are resided by villages [18]. According to a United Nations
used, including search engines, statistical modeling, study on the future of urbanization, 2018 is the year in
reasoning, algorithms, probability, and economics-based which more than 50% of the world's population resides in
processes [13]. Human thinking and the technological towns. This phenomenon is rising increasingly and is
processing of symbols finishes with the development of an projected to hit more than 70% of the world's population by
artificial brain that can be used to create a programmable 2050. Across Germany, 75% of the populace still resides
digital computer. across metropolitan centres. It is projected to hit 80% by
2020 [20]. Nevertheless, the early urbanization of the planet
is an imminent fact that creates several challenges [8]. Rapid • Innovative projects and developments in small-scale
urbanization and attempts to reduce the challenges created communities do not always guarantee effective
by urban population development are also one of the key incorporation at a wider community level.
factors of the creation of a smart city with security prospects • Investment deficiency and lack of municipal priorities and
eliminated by the effective use of Artificial Intelligence. goals, including slow implementation of programs owing to
financial problems and unwillingness to draw people or
ARTIFICAL INTELIGENCE & SMART CITIES services.
The biggest reason for pushing cities toward knowledge • Specific focus on productivity may lead to a limited
resides in their appetite for economic growth. Throughout perception of societal ideals such as social security and the
the global recession of 2008 and 2009, communities found nature of community life and sustainability.
that they were struggling with other communities, albeit in • Repeating technical solutions is a possibility. Not all towns
ways they had not seen before [24]. Not only did they are going to consider the same problems being considered in
interact with their neighbours in the state or globally, but their vicinity.
they interacted for decades (present and future) through • There is a growing need to manage and monitor the
regional supply and demand networks and the Internet with complex ecosystems of people, organizations and
their other peers across the globe. The need for creative and shareholders.
insightful solutions to major economic problems was thus a • Established smart cities may have outdated technology
global necessity [21]. This is the reason why Smart Cities which may delay the development of cities.
powered by Artificial Intelligence have tremendously gained • There are a variety of problems in developing communities
popularity in solving the problems faced by globalization competing for a share of economic resources. It is also not
and the advent of technological innovations. The demands appropriate to solve all the aspects of a smart city.
of the current economic downturn offer a strong impetus to Prioritization should be the foundation of the plan.
address opposition to reform and turn challenges into • The problem of social disparities (financial disparity)
opportunities; in other terms, the most important catalyst for between cultures, knowledge and disproportionate access to
smart cities to grow was the global recession and the need to IT (digital divide, expanding of social divide and increasing
produce capital financing. The world economy is now spatial polarization). Advances of technology and
internationally interconnected, increasingly service-based, cyberspace trends tend to lead to inequality within society.
and smart cities powered by AI are key to this phenomenon • Failure to train and daily updates are expected.
[26]. • Lack of equal access and uncertainty as to the large
quantity of data.
II. CHALLENGES OF SMART CITIES
Awareness comes with both innovative opportunities IV. HOW AI IS USED TO OVEROCME
and obstacles. In a testing center, a smart city is an CHALLENGES OF SMART CITIES
engineering area that inevitably entails threats (untested and Smart city goals are easy to meet, but it is important to
tested). In order to monitor the risks of development, the learn how to implement them. The route selected to enter the
smart city system is not only an innovative stimulus but destination is more important than the destination itself. How
needs to be critically evaluated. quickly AI can solve smart city challenges and how AI
Collapse in risk control culminated in a total breakdown should tackle the future difficulty of smart cities is the
in public sector research programs. Therefore, 85 per cent of question under purview. Worldwide, millions of road deaths
the projects have collapsed in major sections (policy, and injuries have been recorded last year. It is a human
organization and risk management) owing to challenges mistake. So, combining traditional cars with self-driving
raised by the non-technical facets of innovation. Popular automobiles would absolutely eliminate this human error.
explanations include insufficient planning, operational The Stanford University study shows that the amount of road
injuries may be successfully reduced by self-driving vehicles.
weakness, lack of outstanding strategic resources and
Further, It can also inspire us to lead a healthier lifestyle,
leadership, lack of professional competence, incompatibility where we can spend more time working or have fun on our
with organizational goals and project objectives, and trip. "Increased cognitive burden on self-driving cars and
susceptibility to political instability [16]. Excessive shared travel would impact where people want to live," [12].
appreciation of development and strategic pragmatism.
Furthermore, in the public sector, imagination may be a The demand for energy in any community sector is rising
phenomenon. Governments are constructing progressive with the convergence of smart cities with progressively
bureaucracies toward change, resistance or interruption advanced infrastructure. Production is still on the rise,
without competitive strain. Present threats need increasingly leading to environmental concerns related to greenhouse gas
advanced technology to successfully monitor and emissions. At the other side, we do need to tackle
commodities that drain and turn to more secure and
incorporate appropriate management mechanisms to tackle
sustainable sources of livelihood. Therefore AI-driven
community knowledge threats [14]. software may also be really useful in this case [18]. AI will
Following are the main challenges encountered in Smart change our energy resource production, delivery and use in
City Implementation: the future. Indeed, AI has already launched intelligent cities.
• Competitiveness and the inability of cities to perform Internet companies like Google are fitted with huge data
abroad, given municipal economic cohesion. centres with 24/7 servers. These servers need huge power
• Failure to add financial support and dynamism to the plan supplies to stay cool without heating up at high temperatures.
with a sustainable city project. Google is using the newly established DeepMind platform to
fix this problem. Through tracking when the cooling device
is working, the framework reduces excess energy usage.
Through this AI, Google saved nearly 40% of the electricity they assimilate what we make them assimilate. This might
costs of its data centres [12]. also be that we have got them to follow a path in the
labyrinth and out of the superior ability of the AI to drive the
AI driven supply networks would monitor and control the self-created with adverse implications.
usage of customers and touch points. This means that energy
knowledge is received concurrently. The United States has
been implementing this since 2010 The Energy Department
has spent $4.5 billion in smart grid infrastructure including
15 million smart meters in urban cities [28]. These V. Conclusion
measurements help to monitor energy consumption and alert
the authorities of blackouts. The usage of such AI-integrated Smart City planning requires systematic action at various
meters is projected to raise the grid's peak electricity governmental and social rates. Smart City is a holistic term
consumption by just 1% by 2050, whereas US power demand that tackles current issues and takes advantage of the new
is predicted to grow to 25% by 2050. There are now developments in digital infrastructure, connectivity and
significant prospects for reductions and rewards. In fact, urbanization. There is still no clear agreement on this topic,
according to Bloomberg Reports, General Electric is despite the comprehensive literature on the concept of smart
exploring ways to optimize battery movements and the point cities. Scientists from various research backgrounds have
of sale. Steven Martin, Chief Financial Officer at GE, said suggested specific outcomes. Some new innovations are
that this program might eventually save $200 billion globally viewed as the only or least important aspect of smart cities,
[19]. others have suggested ideas that go beyond technology and
The issue of smart cities is mainly the increasing usage of believe that technical adoption is not the end of the job.
information and communication technology (ICT) that Technologies can be used in societies to empower citizens by
depends on the Internet, computation and processing implementing these technologies to their needs instead of
capability to function. If the number of smart grids and smart adjusting their lives to the needs of technology. The review
sensors grows, more data has to be managed, more machines of broad arrays in literature indicates that the concept of a
and more computing power are required. Eventually, if the smart city is multifaceted and thus requires to be applied after
usage of AI-powered machines rises, the adverse a strict examination and evaluation of its viability and
environmental consequences of greenhouse gas emissions applicability.
will also rise [14]. The solution to this question is now being
figured out in smart cities. Georgia Tech University in the
U.S. has reported that the shortage of help & support is one REFERENCES
of the key causes for graduates to leave school halfway
[1] Kylili Angeliki and Fokaides Paris, A. European smart
through. In order to solve this issue, the university has set up
cities: The role of zero energy buildings. Sustainable Cities
a robotic teaching assistant [14]. The artificially intelligent
computer can address the student's questions with 97% and Society, 15:86–95, 2015.
precision. And the students are very happy to be interacting [2] Michael Batty, Kay W Axhausen, Fosca Giannotti,
with their newest multimedia tutor. Alexei Pozdnoukhov, Armando Bazzani, Monica
Wachowicz, Georgios Ouzounis, and Yuval Portugali.
On the contrary, it must be noted that every student has Smart cities of the future. The European Physical Journal
his or her own style of learning and comprehension at Special Topics, 214(1):481–518, 2012.
various stages – some are fast learners, and some are sluggish [3] Fatih Birol and al. Energy and climate change, 2015.
learners. It is also important to tackle this challenge of one- [4] Noelie Bonjean, Wafa Mefteh, Marie-Pierre Gleizes,
to-one personalized training sessions, which cannot be done
Christine Maurel, and Fred´ eric Migeon. Adelfe 2.0. In ´
through our conventional method of teaching. That is where
Handbook on Agent-Oriented Design Processes, pages 19–
artificial intelligence is going to be a huge help. AI will prove
to be a smart teacher and will solve all learning problems of 63. Springer, 2014.
students through its thorough knowledge of every subject. It [5] Eduardo de Oliveira Fernandes Fernandez, Leonardo
is going to teach as much as humans sustain daily chores Meeus, Vitor Leal, Isabel Azevedo, Erik Delarue, and Jean-
[25]. Michel Glachant. How to foster a quick transition towards
local sustainable energy systems, 2011.
Through incorporating AI in the education sector, smart [6] Valerian Guivarch, Val ´ erie Camps, Andr ´ e P ´
cities would have to reconsider our whole education system eninou, and Pierre Glize. ´ Self-adaptation of a learnt
and potentially build a scalable network that fits current behaviour by detecting and by managing user’s implicit
needs. In comparison, completely digital one-to-one session
contradictions. In Proceedings of the 2014 IEEE/WIC/ACM
teachers are a costly practice for any person. In contrast, for
International Joint Conferences on Web Intelligence (WI)
the specific sessions given by the AI professor, there are
possibilities for the AI to use an unethical method of drawing and Intelligent Agent Technologies (IAT)-Volume 03, pages
conclusions. Plus, he cannot be as affectionate as a human 24– 31. IEEE Computer Society, 2014.
teacher, so he can't understand the feelings of a pupil as [7] Rob Kitchin. The real-time city? big data and smart
much as a human teacher should have. In fact, all these urbanism. GeoJournal, 79(1):1–14, 2014.
questions need to be thoroughly answered before AI teachers [8] Alan J Miller. Algorithm as 274: Least squares routines
are deployed [26]. to supplement those of gentleman. Applied Statistics, pages
458–478, 1992.
Since we've designed this unpredictably efficient tool, it's us [9] Mehryar Mohri, Afshin Rostamizadeh, and Ameet
who need to find out how AI can be used to solve the
Talwalkar. Foundations of machine learning. MIT press,
problems not just of smart cities, but of every city in the nook
2012.
and cranny of the globe. Of course, AI is good, and it will
prove excellent in resolving the problems of the region. Not [10] Taewoo Nam and Theresa A Pardo. Conceptualizing
to suggest that this is an issue for AI, but for humans, that smart city with dimensions of technology, people, and
institutions. In Proceedings of the 12th Annual International networking, systems, and applications, 2012, Ambleside,
Digital Government Research Conference: Digital United Kingdom, pp. 135-138.
Government Innovation in Challenging Times, pages 282– [26] D. B. Rawat, D. C. Popescu, G. Yan, and S. Olariu,
291. ACM, 2011. ”Enhancing VANET Performance by Joint Adaptation of
[11] Rocha Paula, Siddiqui Afzal, and Stadler Michael. Transmission Power and Contention Window Size,” IEEE
Improving energy efficiency via smart building energy Transactions on Parallel and Distributed Systems, vol. 22,
management systems: A comparison with policy measures. no. 9, pp. 1528-1535, Sep. 2011.
Energy and Buildings, 88:203–213, 2015. [27] S. Djahel and Y. Ghamri-Doudane, ”A Robust
[12] K.R. Rao. Radioactive waste: The problem and its Congestion Control Scheme for Fast and Reliable
management. Current Science, 81(12):1534–1539, 2001. Dissemination of Safety Messages in VANETs”, IEEE
[13] Janet L. Sawin and al. Renewables 2015 global status WCNC 2012, Paris, France, pp. 2264-2269, April 1-4, 2012.
report. Technical report, Renewable Energy Policy Network [28] S. Djahel, M. Salehie, I. Tal and P. Jamshidi, ”
for the 21st century, 2015. Adaptive Traffic Management for Secure and Efficient
[14] Xindong Wu, Xingquan Zhu, Gong-Qing Wu, and Wei Emergency Services in Smart Cities”, IEEE Pervasive
Ding. Data mining with big data. Knowledge and Data Computing and Communication (PerCom) conference, San
Engineering, IEEE Transactions on, 26(1):97–107, 2014. Diego, California, USA, pp. 240-243, March 18-22, 2013.
[15] U. Nagaraj and P. Dhamal, ”Broadcasting Routing [29] I. H. Brahmi, S. Djahel and Y. Ghamri-Doudane. ”A
Protocols in VANET,” Network and Complex Systems, vol. Hidden Markov Model based Scheme for Efficient and Fast
1, no. 2, pp. 13-19, 2012. Dissemination of safety Messages in VANETs”. IEEE
[16] M. T. Sun, W. C. Feng, T. H. Lai, K. Yamada, H. GLOBECOM 2012, Anaheim, California, USA, pp. 177-
Okada, and K. Fujimura, ”GPS-based message broadcasting 182, December 3-7, 2012.
for inter-vehicle communication,” in Parallel Processing, [30] V. T. Ngoc Nha, S. Djahel and J. Murphy. ”A
2000. Proceedings. 2000 International Conference on, 2000, Comparative Study of Vehicles’ Routing Algorithms for
Toronto, Canada, pp. 279-286. Route Planning in Smart Cities”. VTM 2012, Satellite
[17] B. Karp and H. T. kung, ”GPSR: Greedy perimeter Workshop of IFIP Wireless Days 2012, Dublin, Ireland, pp.
Stateless Routing for Wireless Networks”, ACM MobiCom 1-6, November 20, 2012.
2000, August 6-11, 2000, Boston, MA, USA, pp. 243-254.
[18] Z. Mo, H. Zhu, K. Makki, and N. Pissinou, ”MURU: A
multi-hop routing protocol for urban vehicular ad hoc
networks,” in Mobile and Ubiquitous Systems: Networking
& Services, 2006 Third Annual International Conference on,
2006, London, Great Britain, pp. 1-8.
[19] B. C. Seet, G. Liu, B. S. Lee, C. H. Foh, K. J. Wong,
and K. K. Lee, ”A-STAR: A mobile ad hoc routing strategy
for metropolis vehicular communications,” NETWORKING
2004. Networking Technologies, Services, and Protocols;
Performance of Computer and Communication Networks;
Mobile and Wireless Communications, pp. 989-999, 2004.
[20] J. Bernsen and D. Manivannan, ”Unicast routing
protocols for vehicular ad hoc networks: A critical
comparison and classification,” Pervasive and Mobile
Computing, vol. 5, no. 1, pp. 1-18, 2009.
[21] A. Bachir and A. Benslimane, ”A multicast protocol in
ad hoc networks inter-vehicle geocast,” in Vehicular
Technology Conference, 2003. VTC 2003-Spring. The 57th
IEEE Semiannual, 2003, vol. 4, Jeju, Korea, pp. 2456-2460.
[22] M. Jerbi, R. Meraihi, S. M. Senouci, and Y. Ghamri-
Doudane, ”GyTAR: improved greedy traffic aware routing
protocol for vehicular ad hoc networks in city
environments,” in Proceedings of the 3rd international
workshop on Vehicular ad hoc networks, 2006, Los
Angeles, CA, USA, pp. 88-89.
[23] W. Sun, H. Yamaguchi, K. Yukimasa, and S.
Kusumoto, ”GVGrid: A QoS routing protocol for vehicular
ad hoc networks,” in Quality of Service, 2006. IWQoS
2006. 14th IEEE International Workshop on, 2006, New
Haven, CT, USA, pp. 130-139.
[24] H. Lu and C. Poellabauer, ”Balancing broadcast
reliability and transmission range in VANETs,” in Vehicular
Networking Conference (VNC), 2010 IEEE, 2010, Jersey
City, New Jersey, USA, pp. 247-254.
[25] M. Feiri, J. Petit, and F. Kargl, ”Congestion-based
certificate omission in VANETs,” in Proceedings of the
ninth ACM international workshop on Vehicular inter-