AI at the Edge of Urban Intelligence: Real-Time Awareness and Precision Diagnostics for Resilient Smart Cities
DOI:
https://doi.org/10.56472/WCAI25-104Keywords:
AI at the Edge, Smart Cities, Edge Computing, 5G, Digital Twin, Real-Time Awareness, Precision Diagnostics, Federated Learning, NGSI-LD, Adversarial AI DefenseAbstract
Artificial intelligence (AI), the Internet of Things (IoT), and fifth-generation (5G) access-network technologies have introduced potential traits for building resilient smart-city infrastructures. However, dependence on cloud-focused analytics introduces latency, wasted bandwidth, and vulnerability to the inherent risks in cloud connectivity, which can hinder real-time responses in mission-critical operations. This paper proposes a system model for airborne urban intelligence at the edge of AI, where machine learning models and federated IoT interacting digital twins operate at the edge and close to data sources, providing real-time situational awareness and precision diagnostics for urban systems. The proposed framework of CityEdge-Rx combines multi-access edge computing (MEC), container-native orchestration, NGSI–LD–based context management, and TinyML for low-cost inference while ensuring resiliency against adversarial AI attacks, such as sensor spoofing, data poisoning, and large language model (LLM) prompt injection. Governance is based on NIST AI RMF, Zero Trust concepts, and the EU AI Act, with compliance requirements for high-risk implementations. Testing in the context of traffic control, power grid monitoring, and water leak detection results in lower detection latency (35–50% reduction), bandwidth savings (>70%), and improved operational resilience. Through the integration of edge intelligence, digital-twin–in-the-loop diagnostics, and adversarial robustness, the framework provides a pragmatic roadmap for municipalities and technology providers to scale smart-city systems from pilots to production-grade installations
Downloads
References
[1] ETSI GS MEC 003 V3.2.1, Multi-access Edge Computing (MEC); Framework and Reference Architecture, European Telecommunications Standards Institute, 2022–2024.
[2] M. Satyanarayanan, P. Bahl, R. Caceres, and N. Davies, “The case for VM-based cloudlets in mobile computing,” IEEE Pervasive Computing, vol. 8, no. 4, pp. 14–23, 2009.
[3] National Institute of Standards and Technology (NIST), Artificial Intelligence Risk Management Framework (AI RMF 1.0), NIST AI 100-1, Jan. 2023.
[4] European Union, Regulation (EU) 2024/1689, Artificial Intelligence Act, Official Journal of the European Union, July 12, 2024.
[5] Microsoft Azure, “Azure Digital Twins Documentation,” Microsoft, 2023. [Online]. Available: https://learn.microsoft.com/en-us/azure/digital-twins/
[6] Google Cloud, “Vertex AI Documentation,” Google, 2023. [Online]. Available: https://cloud.google.com/vertex-ai/docs
[7] ETSI GS CIM 009 V1.8.1, Context Information Management (CIM); NGSI-LD API, European Telecommunications Standards Institute, Mar. 2024.
[8] IBM, “Fog Computing and the Internet of Things: Extend the Cloud to Where the Things Are,” White Paper, IBM, 2015.
[9] N. H. Tran, D. T. Hoang, W. Bao, D. Niyato, P. Wang, and Z. Han, “Federated Learning over Wireless Networks: Optimization Model Design and Analysis,” in Proc. IEEE INFOCOM, pp. 1387–1395, 2019.
[10] HashiCorp, “Vault: Identity-based Security for Secrets and Data,” HashiCorp, 2023. [Online]. Available: https://www.vaultproject.io/
[11] Red Hat, “An Introduction to Service Mesh with Istio,” Red Hat, 2023. [Online]. Available: https://www.redhat.com/en/topics/microservices/what-is-a-service-mesh
[12] J. A. Sánchez-Vaquerizo, M. Á. Redondo, and F. J. García-Peñalvo, “Urban Digital Twins and Metaverses Towards City Operations,” Humanities and Social Sciences Communications, vol. 11, no. 1, pp. 1–12, 2024.
[13] National Institute of Standards and Technology (NIST), SP 800-82 Rev. 3: Guide to Operational Technology (OT) Security, NIST, Sept. 2023.
[14] National Institute of Standards and Technology (NIST), SP 800-207: Zero Trust Architecture, NIST, Aug. 2020.
[15] KubeEdge Project, “KubeEdge: Kubernetes Native Edge Computing Framework,” CNCF, 2023–2025. [Online]. Available: https://kubeedge.io/en/
[16] The OWASP Foundation, “Top 10 for Large Language Model Applications v1.1,” OWASP, 2025. [Online]. Available: https://owasp.org/www-project-top-10-for-llm/
[17] MITRE, “Adversarial Threat Landscape for Artificial-Intelligence Systems (ATLAS),” MITRE Corporation, 2024. [Online]. Available: https://atlas.mitre.org/
[18] EdgeX Foundry, “EdgeX Foundry: Open, Vendor-Neutral Edge Middleware,” LF Edge, 2023. [Online]. Available: https://www.edgexfoundry.org/
[19] Open Policy Agent, “OPA: Policy-Based Control for Cloud-Native Environments,” CNCF, 2023. [Online]. Available: https://www.openpolicyagent.org/
[20] Google Cloud, “Anthos: Managed Application Platform,” Google, 2023. [Online]. Available: https://cloud.google.com/anthos
[21] Kothuru, S. K., & Sehrawat, S. K. (2024, April). Impact of Artificial Intelligence and Machine Learning in the Sustainable Transformation of the Pharma Industry. In International Conference on Sustainable Development through Machine Learning, AI and IoT (pp. 60-69). Cham: Springer Nature Switzerland.
[22] Sandeep Rangineni Latha Thamma reddi Sudheer Kumar Kothuru , Venkata Surendra Kumar, Anil Kumar Vadlamudi. Analysis on Data Engineering: Solving Data preparation tasks with ChatGPT to finish Data Preparation. Journal of Emerging Technologies and Innovative Research. 2023/12. (10)12, PP 11, https://www.jetir.org/view?paper=JETIR2312580
[23] L. N. R. Mudunuri and V. Attaluri, “Urban development challenges and the role of cloud AI-powered blue-green solutions,” In Advances in Public Policy and Administration, IGI Global, USA, pp. 507–522, 2024.
[24] V. M. Aragani and P. K. Maroju, "Future of blue-green cities emerging trends and innovations in iCloud infrastructure," in Advances in Public Policy and Administration, pp. 223–244, IGI Global, USA, 2024.
[25] Thirunagalingam, A. (2024). Transforming real-time data processing: the impact of AutoML on dynamic data pipelines. Available at SSRN 5047601.
