Оцінювання обчислювальних моделей функціонування кіберфізичних систем
This paper reviews the use of computational models to support the functioning of cyber-physical systems (CPS) in the parallel world of the Internet of Things (IoT). Existing models, methods, techniques and their implementation in this direction are studied. The necessity of using machine learning me...
Збережено в:
| Дата: | 2020 |
|---|---|
| Автори: | , |
| Формат: | Стаття |
| Мова: | Англійська |
| Опубліковано: |
The National Technical University of Ukraine "Igor Sikorsky Kyiv Polytechnic Institute"
2020
|
| Теми: | |
| Онлайн доступ: | https://journal.iasa.kpi.ua/article/view/209082 |
| Теги: |
Додати тег
Немає тегів, Будьте першим, хто поставить тег для цього запису!
|
| Назва журналу: | System research and information technologies |
| Завантажити файл: |  |
Репозитарії
System research and information technologies| _version_ | 1866391920450732032 |
|---|---|
| author | Pankratova, Nataliya D. Ptukha, Y. A. |
| author_facet | Pankratova, Nataliya D. Ptukha, Y. A. |
| author_sort | Pankratova, Nataliya D. |
| baseUrl_str | http://journal.iasa.kpi.ua/oai |
| collection | OJS |
| datestamp_date | 2020-08-11T08:50:57Z |
| description | This paper reviews the use of computational models to support the functioning of cyber-physical systems (CPS) in the parallel world of the Internet of Things (IoT). Existing models, methods, techniques and their implementation in this direction are studied. The necessity of using machine learning methods due to inaccuracy, fuzziness, incompleteness of the transmitted data from sensors of physical systems is substantiated. The task is to make informed decisions in a timely manner to support the functioning of real objects of a particular cyber-physical system in real time conditions. |
| doi_str_mv | 10.20535/SRIT.2308-8893.2020.1.03 |
| first_indexed | 2025-07-17T10:26:42Z |
| format | Article |
| fulltext |
N.D. Pankratova, Y.A. Ptukha, 2020
28 ISSN 1681–6048 System Research & Information Technologies, 2020, № 1
TIДC
ПРОБЛЕМИ ПРИЙНЯТТЯ РІШЕНЬ ТА
УПРАВЛІННЯ В ЕКОНОМІЧНИХ, ТЕХНІЧНИХ,
ЕКОЛОГІЧНИХ І СОЦІАЛЬНИХ СИСТЕМАХ
UDC 303.732.4, 519.816
DOI: 10.20535/SRIT.2308-8893.2020.1.03
ESTIMATION COMPUTATIONAL MODELS
OF THE CYBER-PHYSICAL SYSTEMS FUNCTIONING
N.D. PANKRATOVA, Y.A. PTUKHA
Abstract. This paper reviews the use of computational models to support the func-
tioning of cyber-physical systems (CPS) in the parallel world of the Internet of
Things (IoT). Existing models, methods, techniques and their implementation in this
direction are studied. The necessity of using machine learning methods due to inac-
curacy, fuzziness, incompleteness of the transmitted data from sensors of physical
systems is substantiated. The task is to make informed decisions in a timely manner
to support the functioning of real objects of a particular cyber-physical system in re-
al time conditions.
Keywords: cyber-physical systems, the Internet of Things, system methodology,
machine learning, computer systems, sensors.
INTRODUCTION
A cyber-physical system (CPS) is a complex distributed system, managed and
controlled by computer systems, tightly integrated with the Internet and its users.
The main principle of the CPS is a deep relationship between its physical and
computational elements to make decisions regarding the maintenance of the func-
tioning of real objects.
The cyber-physical system was proposed for the study of complex systems
consisting of various heterogeneous natural objects, artificial systems, controllers
and distributed computing systems, including embedded real-time systems,
automated control systems for technical processes and objects, wireless sensor
networks, combined into a single whole [1]. The technological basis of CPS is the
Internet of Things (IoT), which is the “brain” of a system in the form of artificial
intelligence and other technologies for analysis, processing of data received from
sensors in the real world [2].
The Internet of things becomes a modern tool that includes several stages of
interaction with physical systems: collecting data from a specific physical system,
bringing this information to the required format, performing calculations based on
models, methods and techniques that allow you to make decisions based on in-
formation, obtained from physical models. In CPS, it becomes a fundamentally
new fact that not only close communication and coordination between computa-
tional and physical resources must be ensured, but also the ability to effectively
Estimation computational models of the cyber-physical systems functioning
Системні дослідження та інформаційні технології, 2020, № 1 29
respond to emerging cyber-physical effects due to the interaction of physical ob-
jects and computational processes, and the ability to make adjustments to ensure
the survivability of the functioning of physical systems.
Recently, the Internet of things has become quite popular due to its potential
ability to be integrated into any complex system. So, on the market you can
already find many ready-made software products, such as Blynk [3], Fracttal [4],
PRG Network Monitor [5], IoT Analytics [6] etc. Even Google offers its Google
Cloud IoT [7] solution on the market. Also, there are companies that offer their
services to create products for IoT of the customer systems [8]. But most of them
are only ready-made tools and algorithms for the field of study in question for
collecting and processing information, and can offer customers forecasting and
decision making support in the subject area. However, IoT involves the
autonomous functioning and management of the system in cases of detecting
threats of emergency situations. For example, according to the article [9] IoT is
described as a network of connected embedded objects or devices with identifiers
where control can occur without human intervention.
REVIEW OF IOT APPLICATION
The future of IoT is being determined. IoT provides decision-making options in
all sectors, including manufacturing, fashion, restaurant, healthcare, education, etc.
IoT applications have already appeared in many aspects of the smart city. In
the process of developing such applications, conflicting goals specific to the se-
lected city are determined and their performance indicators are taken into account.
For example, the government of Amsterdam is investing its financial resources in
reducing transport, energy efficiency and improving the city safety [10], sensor
technologies are implemented into new bus network in Barcelona [11], and Santa
Cruz police use IoT technology to maximize their presence in the most criminal
areas [12]. The Government of India has announced 100 cities that can be devel-
oped as “smart cities”, and also allocated 7,060 crore in the budget for 2015 [13].
Another equally common example of the use of the Internet of Things is the
creation of a smart home. Smart home systems have gained great popularity in
recent decades because they increase comfort and quality of life by making home
appliances more intelligent, remotely controlled and interconnected. But it is also
important that one of the basic functions of the system – the creation of thief
warning technology, is developed in accordance with the environment and culture
of the country [14, 15, 16].
The versatility of IoT makes it an attractive option for so many businesses,
organizations and government agencies so there is no doubt about using this.
However, organizations face many challenges when integrating IoT devices into
outdated ecosystems. So, when the IoT is implemented, all devices work simulta-
neously and the question of data collection and its potential problems arises. IoT
works through remote sensors that can make enterprise privacy public [17]. And a
separate issue is the big risk of hacking company data on IoT devices. Therefore
the Internet of Things is part of Industry 4.0’s modern strategy, which connects
information systems and big data to form a single digital device [18]. As part of
this strategy, modern researchers are developing various frameworks for IoT sys-
tems management solutions [19, 20, 21].
N.D. Pankratova, Y.A. Ptukha
ISSN 1681–6048 System Research & Information Technologies, 2020, № 1 30
For example, in article [22], the authors propose an IoT infrastructure that
focuses on security and reducing complexity, regardless of industry. While
researchers of the Department of Technology Management developed an IoT
framework to monitor all the conditions and results of smart agriculture in
Thailand, which also accompanies developers in improving various processes in
the industry [23].
Other scientists from Lancaster University are exploring the challenges of
the widespread service-oriented architecture (SOA) of IoT software, such as
integration, scaling and sustainability in IoT systems [24]. The development of
frameworks in IoT has become so popular that companies even rank the highest
quality solutions [25, 26] .
Smart city, smart homes, pollution control, energy saving, smart
transportation, smart industries — modern development due to the Internet of
things. Many important studies have been conducted to improve technology using
IoT. However, there are still many problems that need to be solved in order to
fully realize the potential of IoT. The main ones that most scientists put emphasis
on can be determined by the following list:
security and privacy issues related to threats, cyber-attacks, risks and vul-
nerabilities;
compatibility issues that arise due to the heterogeneous nature of the vari-
ous technologies and solutions used to develop IoT;
ethics, law and regulatory rights to comply with quality standards and
prevent illegal use of people;
scalability, availability and reliability to support a large number of de-
vices with different memory, processing, storage capacity and bandwidth;
compliance with quality of service (QoS) standards.
It means that improving existing solutions in the above areas can improve
the quality and safety of human life at the highest possible level for IoT systems [27].
APPLICATION OF MACHINE LEARNING METHODS
Despite great progress, developing IoT applications is still a complex and time-
consuming task. Basically, there are 5 main stages of developing such solutions
[28–31] (Figure).
Architecture of IoT
Estimation computational models of the cyber-physical systems functioning
Системні дослідження та інформаційні технології, 2020, № 1 31
1. Perception stage is the physical layer that is responsible for collecting en-
vironmental information. It defines some physical parameters or identifies other
intelligent objects in the environment.
2. The transport phase transfers sensor data from the perception level to the
processing level and vice versa through networks such as wireless networks, 3G,
LAN, Bluetooth, RFID and NFC.
3. The processing layer stores, analyzes and processes huge volumes of data
(Big Data) coming from the transport layer, using technologies such as databases
and cloud computing.
4. The application layer is responsible for providing the user with applica-
tion-specific services. It identifies various applications where the Internet of
Things can be deployed, such as smart homes, smart cities, and smart health.
5. The business layer manages the entire IoT system, including applications,
business models and profit models, as well as user privacy.
In this study, we consider 3–4 steps, which involve processing the data re-
ceived and supporting timely decision making.
The existing IoT network architecture allows you to extract, convert, delete
and consolidate structured data from existing databases and unstructured data
from sensors, various transmitting devices. Such data is analyzed using software
services that run on a virtual machine for advanced analysis using various ma-
chine learning (ML) methods. Most aspects such as managing a smart city, home
or business, forecasting water demand, or detecting anomalies are solved directly
using ML methods [32, 33].
It is also necessary to consider the issue of processing a large amount of data
generated by the IoT system, especially possible approaches to creating state con-
trol algorithms that can operate on client devices or nodes with low computational
performance [34]. However, in each specific case, in order to create real systems,
it is necessary to analyze the most effective algorithms for forming a fuzzy
knowledge base and draw fuzzy inference, as well as pre-train systems during the
formation of a fuzzy knowledge base and when choosing algorithms for fuzzy
inference [35].
An additional problem that we face in the process of studying existing IoT
applications is the lack of formalization of the methods described in materials
available for research. The main part of the works mentioned above offers only
descriptive materials of architecture, selected methods and obtained application
results.
CONCLUSION
Based on the above review of the development and application of IoT, in this pa-
per, it is proposed to use the IoT concept where, in real-time, data obtained from
physical systems will come to a parallel world and a decision will be formed in a
timely manner to ensure the survivability of their functioning based on the analy-
sis and processing of these data. A feature of the application of ML methods in
IoT systems is that these methods must control the operation of devices or form
decisions on the behavior of the system in emergency and critical situations. In
other words, they must accompany the work of physical systems in real time.
N.D. Pankratova, Y.A. Ptukha
ISSN 1681–6048 System Research & Information Technologies, 2020, № 1 32
As an example of the implementation of the system strategy of guaranteed
survivability of the functioning of the system, it is proposed to consider a real
closed deep water supply system [36]. The main purpose of the system is to
provide the indicated level of water consumption for consumers, provided that the
priority is the cooling process of an environmentally hazardous technological
installation.
In order to achieve the goal of timely identification of the causes of
potentially possible emergency situations and to ensure the survivability of its
operation in real time, monitoring is carried out in the form of technical
diagnostics and indicators are taken at key points in the water supply system,
which are transmitted to the IoT to develop a solution to support the functioning
of the system. It is assumed that the implementation of a water management
system can be accomplished by using hybrid neural networks, which is planned to
be presented in further works.
REFERENCES
1. Vyatkin V. Cyber-physical engineering as a way of applying the theory of cyber-
physical systems / Valeriy Vyatkin, Dmitry Drozdov, Yulia A. Golub // SAEC.
— 2019. — No. 1.
2. Xia F. Internet of things / F. Xia // International Journal of Communication Systems.
— 2012. — Vol. 25. — No. 9. — P. 1101.
3. https://blynk.io/
4. https://www.fracttal.com/en/fracttal-box
5. https://www.paessler.com/prtg
6. https://iot-analytics.com/
7. https://cloud.google.com/solutions/iot
8. https://www.analyticsinsight.net/top-iot-companies/
9. Gubbi J. Internet of things (IoT): A vision, architectural elements, and future direc-
tions / J. Gubbi, R. Buyya, S. Marusic, M. Palaniswami // Future Generation
Computer Systems. — 2013. — Vol. 29. — No. 7. — P. 1645–1660.
10. Mora L. How to Become a Smart City: Learning from Amsterdam / L. Mora,
R. Bolici // In: Bisello A., Vettorato D., Stephens R., Elisei P. (eds) Smart and
Sustainable Planning for Cities and Regions. SSPCR 2015. Green Energy and
Technology. — Springer, Cham. — 10.1007/978-3-319-44899-2_15
11. Madakam S. Barcelona Smart City: The Heaven on Earth (Internet of Things: Tech-
nological God) / S. Madakam, R. Ramaswamy. — An International ICT R&D
Journal Sponsored by ZTE Corporation. — 2015. — Vol. 13. — N 4.
12. Baxter S. Modest gains in first six months of Santa Cruz's predictive police program
/ S. Baxter // Santa Cruz Sentinel. — 2012. — Retrieved 2015-05-26.
13. Chatterjee S. Success of IoT in Smart Cities of India: An empirical analysis /
S. Chatterjee, A. Kar, MP. Gupta // Government Information Quarterly. — 2018.
— 10.1016/j.giq.2018.05.002.
14. Tran A. Thuan. SmartHome IoT Application Development / Tran A. Thuan, T. Huy,
D. Hoang et al. — 2019.
15. Alaa M.. A Review of Smart Home Applications based on Internet of Things /
M. Alaa, A. Zaidan, B. Bahaa et al. // Journal of Network and Computer Applica-
tions. — 2017. — 97. 10.1016/j.jnca. 2017.08.017.
16. Yang H. IoT Smart Home Adoption: The Importance of Proper Level Automation /
H. Yang , L. Wonji, L.Hwansoo // Journal of Sensors. — 2018. — P. 1–11. —
10.1155/2018/6464036.
17. Khan A. The Role of IoT in Digital Governance / A. Khan, N. Zaman, M. Humayun.
— 2020. — 10.4018/978-1-7998-1851-9.ch007.
18. Koman G. Advantages of IoT implementation in enterprise process management /
G. Koman, M. Kubina, D. Šulyová. — 2020.
Estimation computational models of the cyber-physical systems functioning
Системні дослідження та інформаційні технології, 2020, № 1 33
19. Mohapatra S. A Solution Framework for Managing Internet of Things (IoT) /
S. Mohapatra, J. Bhuyan // International journal of Computer Networks & Com-
munications. — 2016. — 8. — P. 73–87. — 10.5121/ijcnc.2016.8606.
20. Ando B. Editorial: Hot I&M topics in the IoT framework / B. Ando // IEEE In-
strumentation & Measurement Magazine. — 2019. — 22.2-2.
10.1109/MIM.2019.8917895.
21. Vranken C. Combination of IoT framework with liquid software. — 2018.
22. Cheruvu S. IoT Frameworks and Complexity / S. Cheruvu, A. Kumar, N. Smith,
D. Wheeler— 2020. — 10.1007/978-1-4842-2896-8_2.
23. Krongthong T. Modeling and Simulink of Smart Agriculture Using IoT Framework /
T. Krongthong, B. Muangmeesri. — 2019. — P. 185–188. —
10.1109/ICORIS.2019. 8874914.
24. Onoriode U. IoT Architectural Framework: Connection and Integration Framework
for IoT Systems / U. Onoriode, G. Kotonya // Electronic Proceedings in Theo-
retical Computer Science. — 2018. — 264. — P. 1–17. —
10.4204/EPTCS.264.1.
25. https://www.byteant.com/blog/5-best-open-source-iot-frameworks/
26. https://www.techtic.com/blog/top-10-open-source-iot-frameworks/
27. Kumar S. Internet of Things is a revolutionary approach for future technology en-
hancement: a review / S. Kumar et al. // Journal of Big Data. — 2019. — 6. —
P. 1–21.
28. Mashal I. Choices for interaction with things on Internet and underlying issues /
I. Mashal, O. Alsaryrah, T.-Y. Chung et al. — Ad Hoc Networks. — 2015. —
Vol. 28. — P. 68–90.
29. Wu M. Research on the architecture of internet of things / M. Wu, T.-J. Lu,
F.-Y. Ling et al. // in Proceedings of the 3rd International Conference on
Advanced Computer Theory and Engineering (ICACTE ’10). — Vol. 5. —
P. 484–487. — IEEE, Chengdu, China, August 2010.
30. Khan R. Future internet: the internet of things architecture, possible applications and
key challenges / R. Khan, S.U. Khan, R. Zaheer, S. Khan // in Proceedings of the
10th International Conference on Frontiers of Information Technology (FIT ’12).
— 2012. — P. 257–260.
31. Sethi P. Internet of Things: Architectures, Protocols, and Applications / P. Sethi,
S. Sarangi // Journal of Electrical and Computer Engineering. — 2017. —
P. 1–25. — 10.1155/2017/9324035.
32. Vijai P. Design of IoT Systems and Analytics in the Context of Smart City Initiatives
/ P. Vijai, P. Sivakumar // India. Procedia Computer Science. — 2016. — 92. —
P. 583–588. — 10.1016/j.procs.2016.07.386.
33. Zantalis F. A Review of Machine Learning and IoT in Smart Transportation /
F. Zantalis, G. Koulouras, S. Karabetsos // Future Internet. — 2019. — 11.
10.3390/fi11040094.
34. Luntovskyy A. Performance, Reliability and Scalability for IoT / A. Luntovskyy,
L. Globa. — 2019. — P. 316–321. — 10.1109/DT.2019.8813679.
35. Globa L. Fuzzy logic usage for the data processing in the Internet of Things net-
works / L. Globa et al. // Open Semantic Technologies for Intelligent Systems
(OSTIS-2018): International scientific and technical conference proceedings
(Minsk, February 15–17, 2018) / V. V. Golenkov, Ed. — Minsk: BGUIR, 2018.
36. Pankratova N.D. System analysis: Theory and applications: textbook / N.D. Pankra-
tova; NASU, NTUU “KPI”, IASA NASU. — K.: Naukova Dumka, 2018. — 347 p.
Received 20.03.2020
____________________________________
From the Editorial Board: the article corresponds completely to submitted manuscript.
|
| id | journaliasakpiua-article-209082 |
| institution | System research and information technologies |
| keywords_txt_mv | keywords |
| language | English |
| last_indexed | 2025-07-17T10:26:42Z |
| publishDate | 2020 |
| publisher | The National Technical University of Ukraine "Igor Sikorsky Kyiv Polytechnic Institute" |
| record_format | ojs |
| resource_txt_mv | journaliasakpiua/67/fe5dc9a2221bfc6ba9bb3c5481866367.pdf |
| spelling | journaliasakpiua-article-2090822020-08-11T08:50:57Z Estimation computational models of the cyber-physical systems functioning Оценивание вычислительных моделей функционирования киберфизических систем Оцінювання обчислювальних моделей функціонування кіберфізичних систем Pankratova, Nataliya D. Ptukha, Y. A. кіберфізичні системи інтернет речей системна методологія машинне навчання комп'ютерні системи датчики киберфизические системы интернет вещей системная методология машинное обучение компьютерные системы датчики cyber-physical systems the Internet of Things system methodology machine learning computer systems sensors This paper reviews the use of computational models to support the functioning of cyber-physical systems (CPS) in the parallel world of the Internet of Things (IoT). Existing models, methods, techniques and their implementation in this direction are studied. The necessity of using machine learning methods due to inaccuracy, fuzziness, incompleteness of the transmitted data from sensors of physical systems is substantiated. The task is to make informed decisions in a timely manner to support the functioning of real objects of a particular cyber-physical system in real time conditions. Рассмотрено использование вычислительных моделей для поддержки функционирования киберфизических систем (CPS) в параллельном мире интернета вещей (IoT). Изучены существующие модели, методы, методики и их реализация в этом направлении. Обоснована необходимость использования методов машинного обучения из-за неточности, нечеткости, неполноты передаваемых данных с датчиков физических систем с целью своевременного обоснования решения по поддержанию функционирования реальных объектов конкретной киберфизической системы в режиме реального времени. Розглянуто використання обчислювальних моделей для підтримання функціонування кіберфізичних систем (CPS) у паралельному світі інтернету речей (IoT). Вивчено існуючі моделі, методи, прийоми та їх реалізація в цьому напрямі. Обґрунтовано необхідність використання методів машинного навчання через неточність, нечіткість, незавершеність переданих даних від датчиків фізичних систем з метою своєчасного обґрунтовання рішення щодо підтримання функціонування реальних об’єктів певної кіберфізичної системи в умовах реального часу. The National Technical University of Ukraine "Igor Sikorsky Kyiv Polytechnic Institute" 2020-06-23 Article Article application/pdf https://journal.iasa.kpi.ua/article/view/209082 10.20535/SRIT.2308-8893.2020.1.03 System research and information technologies; No. 1 (2020); 28-33 Системные исследования и информационные технологии; № 1 (2020); 28-33 Системні дослідження та інформаційні технології; № 1 (2020); 28-33 2308-8893 1681-6048 en https://journal.iasa.kpi.ua/article/view/209082/209528 Copyright (c) 2021 System research and information technologies |
| spellingShingle | кіберфізичні системи інтернет речей системна методологія машинне навчання комп'ютерні системи датчики Pankratova, Nataliya D. Ptukha, Y. A. Оцінювання обчислювальних моделей функціонування кіберфізичних систем |
| title | Оцінювання обчислювальних моделей функціонування кіберфізичних систем |
| title_alt | Estimation computational models of the cyber-physical systems functioning Оценивание вычислительных моделей функционирования киберфизических систем |
| title_full | Оцінювання обчислювальних моделей функціонування кіберфізичних систем |
| title_fullStr | Оцінювання обчислювальних моделей функціонування кіберфізичних систем |
| title_full_unstemmed | Оцінювання обчислювальних моделей функціонування кіберфізичних систем |
| title_short | Оцінювання обчислювальних моделей функціонування кіберфізичних систем |
| title_sort | оцінювання обчислювальних моделей функціонування кіберфізичних систем |
| topic | кіберфізичні системи інтернет речей системна методологія машинне навчання комп'ютерні системи датчики |
| topic_facet | кіберфізичні системи інтернет речей системна методологія машинне навчання комп'ютерні системи датчики киберфизические системы интернет вещей системная методология машинное обучение компьютерные системы датчики cyber-physical systems the Internet of Things system methodology machine learning computer systems sensors |
| url | https://journal.iasa.kpi.ua/article/view/209082 |
| work_keys_str_mv | AT pankratovanataliyad estimationcomputationalmodelsofthecyberphysicalsystemsfunctioning AT ptukhaya estimationcomputationalmodelsofthecyberphysicalsystemsfunctioning AT pankratovanataliyad ocenivanievyčislitelʹnyhmodelejfunkcionirovaniâkiberfizičeskihsistem AT ptukhaya ocenivanievyčislitelʹnyhmodelejfunkcionirovaniâkiberfizičeskihsistem AT pankratovanataliyad ocínûvannâobčislûvalʹnihmodelejfunkcíonuvannâkíberfízičnihsistem AT ptukhaya ocínûvannâobčislûvalʹnihmodelejfunkcíonuvannâkíberfízičnihsistem |