Інтелектуалізація граничних обчислень інтернету речей.

Коваленко О.Є.

doi:10.34121/1028-9763-2024-3-4-50-68

Authors

Kovalenko O.Ye. https://orcid.org/0000-0002-9639-3544 , Національний університет біоресурсів і природокористування України, м. Київ, Україна

DOI:

https://doi.org/10.34121/1028-9763-2024-3-4-50-68

Keywords:

edge computing, Internet of Things, artificial intelligence, security of edge computing, граничні обчислення, інтернет речей, штучний інтелект, безпека граничних обчислень

Abstract

The combination of edge computing technologies and the Internet of Things creates the edge Internet of Things. Internet of Things edge computing represents the architecture of edge computing integrated into the Internet of Things system with the help of an edge (computing) gateway. This gateway, which integrates network and computing storage and application capabilities, is deployed at the network edge near devices or data sources to provide device management and Internet of Things system management services at network edge nodes. Edge artificial intelligence or «artificial intelligence at the edge» refers to the combination of edge computing and artificial intelligence to perform machine-learning tasks directly on interconnected edge devices. One of the trends in the evolution of software related to edge computing is the use of artificial intelligence methods and algorithms both directly for information processing and for adaptive software-controlled deployment of network infrastructure. The advantage of edge computing architecture is security. Many companies are reluctant to send sensitive company data to the cloud and prefer to keep it on-premises, thereby reducing cybersecurity risks. The convergence of information technology and operational technology increases the potential attack area. Therefore, it is important for enterprises to improve their edge computing nodes to protect their data. The rapid development and spread of artificial intelligence technologies is spreading towards edge computing, since artificial intelligence can potentially provide the means to achieve the properties of the «intelligent environment» of the edge-cloud continuum. The article analyzes the use of artificial intelligence in the frontier area of the Internet of Things. As a result of the analysis, the main advantages and problems related to the frontier area of the Internet of Things and approaches to their solution based on the use of artificial intelligence methods were determined. Edge computing and edge artificial intelligence have different applications, but the most important factor is improving the quality of service for Internet of Things devices.

References

1. Kovalenko O., Vishnevsky V., Kosolapov V. Towards Creating the Network of Situatonal Governance Centers and Decision Making Technologies in Distributed Environments. 2020 IEEE 15th International Conference on Advanced Trends in Radioelectronics, Telecommunications and Computer Engineering (TCSET). Lviv-Slavske, Ukraine, 2020. P. 540–545. DOI: https://doi.org/10.1109/TCSET49122.2020.235491.

2. EdgeAI Consortium. URL: https://edge-ai-tech.eu/consortium/.

3. Iorga M., Goren N., Feldman L., Barton R., Martin M. Charif Mahmoudi NIST SP 500-325. Fog Computing Conceptual Model. 2018. 14 March. DOI: https://doi.org/10.6028/NIST.SP.500-325.

4. The Edge-to-Cloud Continuum. URL: https://www.computer.org/csdl/magazine/co/2020/11/09237349/1o8m4Fl5nUc.

5. The European Cloud, Edge and IoT Continuum Initiative. URL: https://eucloudedgeiot.eu/about/.

6. Khalyeyev D., Bureš T., Hnětynka P. Towards Characterization of Edge-Cloud Continuum. Proc. of ECSA 2022 Tracks and Workshops. 2023. P. 215–230. URL: https://doi.org/10.1007/978-3-031-36889-9_16.

7. Petersen A.M., Ahmed M.E., Pavlidis I. Grand challenges and emergent modes of convergence science. Humanit. Soc. Sci. Commun. 2021. Vol. 8. Article number 194. DOI: https://doi.org/10.1057/s41599-021-00869-9.

8. Roco M. Convergence-Divergence Process. Handbook of Science and Technology Convergence / W. Bainbridge, M. Roco (eds.). Cham: Springer, 2016. P. 79–93.

9. Kovalenko O. Knowledge Driven Cyber-Convergent Systems Based on Situational Agents. 2022 IEEE 17th International Conference on Computer Sciences and Information Technologies (CSIT). 2022. P. 243–246. DOI: https://doi.org/10.1109/CSIT56902.2022.10000762.

10. Kovalenko O. Systems Convergence for Situational Control and Decision Making in Distributed Environments. 2022 IEEE 16th International Conference on Advanced Trends in Radioelectronics, Telecommunications and Computer Engineering (TCSET). 2022. P. 344–347. DOI: https://doi.org/10.1109/TCSET55632.2022.9767006.

11. Athanasios Tziouvaras and Fotis Foukalas. Edge AI for Industry 4.0: An Internet of Things Approach. Proc. of the 24th Pan-Hellenic Conference on Informatics (PCI '20). Association for Computing Machinery. New York, NY, USA, 2021. P. 121–126. DOI: https://doi.org/10.1145/3437120.3437289.

12. Chen Zhang, Yu Xie, Hang Bai, Bin Yu, Weihong Li, Yuan Gao A survey on federated learning, Knowledge-Based Systems. 2021. Vol. 216. P. 106775. DOI: https://doi.org/10.1016/j.knosys.2021.106775.

13. Edge AI Benefits, Use Cases & Deployment Models. URL: https://www.run.ai/guides/machine-learning-operations/edge-ai.

14. Introduction to Edge Computing in IIoT An Industrial Internet Consortium White Paper IIC:WHT:IN24:V1.0:PB:20180618 Edge Computing Task Group, ed. Stephen Mellor. URL: https://www.iiconsortium.org/pdf/Introduction_to_Edge_Computing_in_IIoT_2018-06-18.pdf.

15. The Essentials of SD-WAN Architecture: Advantages and Options. URL: https://www.sdxcentral.com/networking/sd-wan/definitions/software-defined-sdn-wan/essentials-sd-wan-architecture/.

16. What is Universal Customer Premises Equipment (uCPE)? URL: https://www.sdxcentral.com/networking/sd-wan/definitions/software-defined-sdn-wan/essentials-sd-wan-architecture/what-is-ucpe/.

17. vRan: How It Works & Why It Matters. URL: https://www.celona.io/network-architecture/vran.

18. Intel® Movidius™ Vision Processing Units (VPUs). URL: https://www.intel.com/content/www/us/en/products/details/processors/movidius-vpu.html.

19. Accelerate AI development with Google Cloud TPUs. URL: https://cloud.google.com/tpu/?hl=uk.

20. NPU Ethos-U55. URL: https://www.arm.com/products/silicon-ip-cpu/ethos/ethos-u55.

21. Li W., Hacid H., Almazrouei E., Debbah M. A Review and a Taxonomy of Edge Machine Learning: Re-quirements, Paradigms, and Techniques. arXiv 2023, arXiv:2302.0857.

22. Edge AI Inference Platform. URL: https://www.aetina.com/products-features.php?t=337.

23. Is Secure Access Service Edge (SASE)? URL: https://www.zscaler.com/resources/security-terms-glossary/what-is-saseb.

24. How is Edge Security Helping Secure Devices that Use Edge AI? URL: https://xailient.com/blog/how-is-edge-security-helping-secure-devices-that-use-edge-ai/.