Latency-Aware Edge Computing Framework for Secure and Efficient IoT-Driven Smart City Services Mar CornelioOmar Cevallos-TorresLorenzo Center for the Study of Computational Mathematics, University of Computer Sciences, 19370 Havana, Cuba. Brigham Young University–Idaho, USA.

The integration of edge computing with Internet of Things (IoT) technology revolutionizes smart city services by enabling real-time data processing, decision-making, and action close to data sources. This paradigm shift addresses the limitations of traditional cloud-based models, which suffer from high Latency, security risks, and limited bandwidth, especially in densely populated urban areas. Real-time edge computing offers a distributed approach to IoT data management by leveraging localized computing power at the network's edge, reducing the need for data to travel to centralized cloud systems. Edge computing facilitates efficient, responsive services in smart cities where applications like traffic management, environmental monitoring, public safety, and energy optimization demand immediate responses. Processing data locally enables quicker response times, enhances data privacy, and minimizes network congestion. This paper examines the architecture and technologies enabling edge computing in smart city applications and the unique challenges such as interoperability, scalability, and security. Case studies and implementations are explored to illustrate the transformative impact of real-time edge computing on urban infrastructure, contributing to more adaptive, resilient, and intelligent city ecosystems.

 Edge computingInternet of thingsSmart citiesReal-time data processingCloud limitationsData

privacy, Network congestion.

 Introduction

The rapid urbanization and population growth in recent decades have created complex challenges in city management and infrastructure, driving the need for smarter, more efficient urban solutions. The Internet of Things (IoT), a network of interconnected devices and sensors, has emerged as a powerful enabler in addressing these challenges, facilitating real-time monitoring, data collection, and management across various public services in what is now known as "smart cities." From traffic management to energy optimization, IoT-enabled applications in smart cities aim to improve quality of life, reduce costs, and foster sustainable development [1], [2].

However, the massive amount of data generated by IoT devices significantly strains traditional cloud computing frameworks, often resulting in Latency, bandwidth congestion, and security risks. To counter these issues, real-time edge computing has become a vital component in IoT-driven smart city solutions. Edge computing processes data locally—near the source of data generation—thereby reducing Latency, minimizing bandwidth use, and enhancing data privacy and security. By performing analytics at or near the data source, edge computing supports more responsive and resilient smart city applications, such as traffic control systems, smart grids, public safety monitoring, and environmental sensing [3], [4].

This paper examines the integration of real-time edge computing with IoT infrastructures to enable responsive, secure, and efficient smart city services. We will explore the architecture of edge computing systems, their role in enhancing data processing efficiency, and the practical implications of real-time edge applications across various smart city domains. By analyzing case studies and technological advancements, this research aims to shed light on the transformative potential of edge computing in realizing the vision of sustainable and intelligent urban ecosystems [1], [2].

Roles and Benefits of Real-Time Edge Computing in Smart City Services Latency reduction and faster response times

Edge computing significantly reduces Latency by processing data locally rather than sending it to a centralized cloud. This rapid response is critical for applications such as autonomous transportation, emergency response systems, and real-time healthcare monitoring. Minimizing Latency ensures quick decision-making and immediate action, which is essential for critical smart city services [2], [5].

 Network bandwidth optimization

Real-time edge computing helps reduce network bandwidth strain by filtering and processing data at the edge, sending only the necessary information to the cloud for storage or further analysis. For example, smart traffic systems can analyze video streams from CCTV cameras in real time at the edge to identify traffic congestion or accidents, sending only relevant data to the cloud, thereby minimizing network load [6].

 Enhanced data privacy and security

Edge computing reduces the need to transmit sensitive data over the network, enhancing privacy and security. Surveillance, healthcare, and public safety applications particularly benefit from this feature, as data can be processed locally, minimizing exposure to potential cyber threats during transmission [7], [8].

 Scalability and flexibility for city growth

As cities expand, so do the demands on IoT infrastructure. Edge computing offers scalability, as computing nodes can be easily deployed throughout the city as needed. This decentralized architecture allows cities to add new services, expand coverage, and accommodate increased IoT device density without overwhelming centralized cloud systems [9].

 Support for predictive maintenance and resource optimization

IoT-driven smart city services rely on the predictive maintenance of assets, such as streetlights, traffic signals, and public vehicles. Edge computing enables real-time monitoring and analysis of these assets, providing predictive insights that support timely maintenance, reduce downtime, and conserve resources [5].

 Energy efficiency and reduced carbon footprint

By processing data locally, edge computing can reduce energy consumption associated with data transfer to and from centralized cloud systems. This reduction supports smart cities in meeting sustainability goals by lowering energy costs and reducing data center carbon footprints.

Real-time edge computing, a paradigm shift in data processing, is emerging as a critical enabler for IoT-driven smart city services. By bringing computation and storage closer to data sources, edge computing addresses the limitations of traditional cloud-centric approaches, offering significant advantages in terms of Latency, bandwidth, and privacy [2].

Key roles of edge computing in smart cities

 Real-time data processing and analysis

Rapid response: edge computing enables swift processing of data generated by IoT devices, allowing for immediate insights and decisions. This is crucial for applications like traffic management, where real-time analysis of traffic flow data can optimize signal timings and reroute traffic.

Reduced Latency: by processing data locally, edge computing eliminates the Latency of transmitting data to remote cloud servers, leading to faster response times and improved user experiences [10], [11].

 Enhanced privacy and security

Data minimization: edge computing enables local data processing, reducing the amount of sensitive data that must be transmitted to the cloud. This minimizes the risk of data breaches and unauthorized access.

Secure data storage: edge devices can store critical data locally, ensuring its security and integrity. This is particularly important for surveillance and public safety applications, where data privacy is paramount [12].

 Improved scalability and reliability

Distributed computing: edge computing distributes the computational workload across multiple devices, enhancing scalability and resilience. This ensures that services remain operational even in the face of network failures or increased demand.

Reduced cloud dependency: by offloading processing tasks to the edge, smart cities can reduce their reliance on cloud infrastructure, lowering costs and improving overall system reliability [13].

 Enabling innovative applications

Autonomous vehicles: edge computing enables them to make real-time decisions based on sensor data, such as avoiding obstacles and navigating complex traffic.

Smart grids: edge computing enables real-time analysis of energy usage patterns and predictive modeling of future demand, optimizing energy distribution and consumption.

Public safety: edge computing can enhance public safety by enabling rapid response to emergencies, such as fire detection, crime prevention, and disaster recovery.

 Figures and Tables

Imagine a smart city where traffic lights adjust in real time to ease congestion; streetlights dim when no one's around, and emergency services respond instantly to incidents. This is the power of edge computing [2], [11]. How it works:

Smart city edge computing architecture.

This figure is shown in the PDF version of the article.

I. Data generation: IoT devices, such as sensors and cameras, collect data from the city environment.

II. Edge processing: instead of sending all data to a central cloud, it's processed locally at the network's edge. III. Real-time insights: quick analysis leads to immediate actions, such as adjusting traffic signals or alerting authorities [12].

IV. Cloud integration: important data is still sent to the cloud for long-term storage and advanced analytics [13]. Benefits:

I. Reduced Latency: faster response times for critical applications.

II. Improved efficiency: optimized network usage and lower energy consumption.

III. Enhanced privacy: sensitive data is processed locally, minimizing security risks.

IV. Real-time decision making: timely actions based on real-world conditions.

Edge computing revolutionizes how smart cities operate by bringing computing power closer to the data This chart demonstrates the significant reduction in Latency achieved by using edge computing compared to traditional cloud computing. Here's a breakdown:

Latency comparison: cloud vs. edge computing.

This figure is shown in the PDF version of the article.

I. Traffic management: edge computing can process traffic data locally, enabling faster adjustments to traffic signals and real-time rerouting, reducing congestion.

II. Smart lighting: edge-based processing enables dynamic control of streetlights, dimming them when unnecessary and brightening them when pedestrians or vehicles approach [14].

III. Environmental monitoring: by processing sensor data locally, edge computing allows quicker detection of pollution spikes or other environmental issues, enabling timely responses.

Key takeaways:

I. Reduced latency: edge computing significantly reduces the time it takes to process and react to data, leading to more efficient and responsive smart city systems II. Real-time decision-making: edge computing enables it by processing data locally, improving the overall performance of smart city applications.

III. Improved user experience: faster response times and more efficient resource utilization enhance the quality of life for city residents.

Data flow in smart traffic management with edge computing.

This figure is shown in the PDF version of the article.

I. Data capture: a smart camera captures real-time video footage of traffic conditions, including vehicle density, speed, and traffic light status.

II. Edge processing: the video stream is transmitted to a nearby edge node. The edge node analyzes the video using AI algorithms to detect traffic patterns, identify congestion points, and estimate traffic flow.

III. Real-time decision making: based on the analyzed data, the edge node makes real-time decisions, such as adjusting traffic light timings, rerouting traffic, or alerting traffic control centers [15].

IV. Action implementation: the edge node sends commands to the traffic control system to implement the necessary changes, such as adjusting traffic light phases or displaying dynamic traffic information on digital signage.

Benefits of edge computing in traffic management

Reduced Latency: edge computing processes data locally, eliminating the need to send large amounts of video data to a centralized cloud, significantly reducing Latency.

Improved response time: faster decision-making and quicker implementation of traffic control measures improve traffic flow and reduce congestion.

Enhanced efficiency: real-time optimization of traffic signals and rerouting can significantly improve traffic efficiency and reduce fuel consumption.

Enhanced safety: edge computing can help prevent accidents and improve road safety by detecting potential hazards such as traffic jams.

Enhanced security with edge computing.

This figure is shown in the PDF version of the article.

Edge computing significantly enhances the security of smart city systems by:

I. Reduced data transmission: by processing data locally, the amount of sensitive information transmitted over networks is minimized, reducing the potential for data breaches.

II. Local data storage: storing data on edge devices limits the exposure of sensitive information to external threats, as it remains within the local network.

III. Limited network exposure: edge computing reduces the attack surface by minimizing the number of devices connected to the internet, making it harder for hackers to infiltrate the system [16].

Cloud vs. edge computing: a comparison for smart cities.

This table is shown in the PDF version of the article.

I. Latency: edge computing significantly reduces Latency by processing data closer to the source, making it ideal for time-sensitive applications like traffic management and emergency response [10].

II. Bandwidth usage: edge computing minimizes network traffic by processing data locally, reducing bandwidth consumption.

III. Data security: edge computing enhances data security by limiting data transmission and storing sensitive data locally, reducing the risk of data breaches.

IV. Energy consumption: edge computing devices consume less energy than large-scale cloud data centers, making them more energy-efficient.

V. Scalability: While cloud computing offers high scalability, edge computing may face limitations in processing power and storage capacity.

However, advancements in edge technologies are continuously improving scalability.

Latency requirements for various smart city applications.

This table is shown in the PDF version of the article.

Edge computing devices and their specifications for smart cities.

This table is shown in the PDF version of the article.

edge processing Cortex M-series) (Flash/ROM) processing at the source, Raspberry Pi with sensors, capability enabling immediate response Intel NUC with sensors

 Variables and Equations Variables

Latency (C): measured in milliseconds.

Data rate (D): measured in Mbps.

Energy consumption (E): measured in Joules.

Task completion time (T): measured in seconds.

Resource availability (R): representing CPU cycles or memory.

Service level agreement (S): parameters related to service quality.

IoT devices count (I): the number of IoT devices.

Number of edge nodes (N): the count of edge computing nodes.

Network bandwidth (L): measured in Mbps.

Quality of service (Q): metrics defining service quality.

Distance (X): the distance from the IoT device to the edge node, measured in kilometers.

 Parameters

Processing power (a): the power of edge nodes, measured in MIPS.

Power consumption per computation (b): measured in Joules per MIPS.

Data fidelity (f): a quality metric for data.

Traffic load (g): the amount of data transmitted.

Number of concurrent tasks (h): the number of tasks processed simultaneously.

 Potential equations

I. Latency calculation: The Latency can be estimated with the formula

C = (X / L) + (D / R).

This equation calculates Latency based on the distance to the edge node and data transmission rates.

II. Energy consumption: Energy consumption can be modeled as

E = h ∗ b ∗ a ∗ T.

This relates energy consumption to the number of tasks, power consumption per computation, processing power, and task completion time.

III. Task completion time: The time to complete tasks can be estimated with

T = (g / a) + (h / N).

This equation estimates task completion time based on data load and the number of edge nodes.

IV. Quality of service: Quality of service can be modeled using

Q = (f / C) ∗ (S / D).

This shows QoS as a function of data fidelity, Latency, and SLA parameters.

V. Network bandwidth utilization: Network utilization can be calculated as

U = (D / L) ∗ 100.

Here, U represents the percentage utilization of the network bandwidth.

VI. Resource allocation: resource availability can be modeled with

R = (I ∗ a) / N.

This relates resource availability to the number of IoT devices and the processing power of edge nodes.

 Limitations

Despite the promising potential of real-time edge computing in enhancing IoT-driven smart city services, several limitations constrain its large-scale implementation:

Infrastructure cost and deployment complexity: Establishing a reliable edge infrastructure requires significant investment in hardware, communication networks, and maintenance. For developing cities, this can pose serious budgetary and logistical challenges.

Scalability constraints: Although edge nodes improve local processing, they have limited storage and computational capacity compared to centralized cloud systems. Scaling edge networks to handle increasing data volumes and device density remains a technical bottleneck [17].

Data privacy and security challenges: processing sensitive data locally introduces heterogeneous security vulnerabilities across distributed nodes. Ensuring consistent protection levels and compliance with privacy regulations remains difficult.

Interoperability issues: the coexistence of devices from different vendors and platforms leads to protocol incompatibility, making system integration and unified data exchange difficult [18].

Energy and maintenance overhead: while edge computing can reduce network energy consumption, maintaining multiple distributed devices increases local energy use and operational complexity.

 Future Work

Future studies can focus on several directions to overcome current limitations and enhance the efficiency and intelligence of edge-enabled smart city systems:

AI-driven edge intelligence: integrating artificial intelligence and federated learning into edge nodes can enable autonomous decision-making, anomaly detection, and adaptive optimization without compromising privacy.

5G and beyond integration: combining edge computing with 5G and upcoming 6G technologies can drastically improve data transmission speed, reliability, and latency performance in large-scale smart city deployments.

Standardization and interoperability frameworks: developing unified communication standards and open-source middleware will enhance interoperability among heterogeneous IoT and edge systems [19].

Energy-aware edge management: future work should explore dynamic energy management models to balance computational load and reduce power consumption across distributed edge nodes.

Security and privacy reinforcement: implementing blockchain-based authentication and decentralized trust management mechanisms could improve data integrity and resilience against cyber threats.

Real-world pilot implementations: expanding from simulation-based models to large-scale pilot projects in transportation, healthcare, and environmental monitoring will validate the practical scalability and reliability of proposed architectures.

 Conclusion

This study highlights the transformative role of real-time edge computing in advancing IoT-driven smart city ecosystems. By processing data close to the source, edge computing reduces Latency, optimizes bandwidth, and enhances privacy — addressing key limitations of cloud-centric architectures. The research demonstrates how edge-enabled applications in traffic management, public safety, energy efficiency, and environmental monitoring can make urban services more responsive, adaptive, and sustainable.

However, challenges such as interoperability, scalability, and data security continue to hinder widespread adoption. Overcoming these barriers requires a combination of AI-based optimization, robust security frameworks, and collaborative urban infrastructure planning. Ultimately, integrating edge computing as a foundational layer of smart city architecture paves the way for intelligent, resilient, and human-centered urban environments.

 Funding

This research has received no external funding.

 Data Availability

The data used and analyzed during the current study are available from the corresponding author upon reasonable request.

 Conflicts of Interest

The authors declare no conflict of interest regarding the publication of this paper.

 References Trigka, M., & Dritsas, E. (2025). Edge and cloud computing in smart cities. Future internet, 17(3), 118. https://doi.org/10.3390/fi17030118 Khan, L. U., Yaqoob, I., Tran, N. H., Kazmi, S. M. A., Dang, T. N., & Hong, C. S. (2020). Edge-computing-enabled smart cities: A comprehensive survey. IEEE internet of things journal, 7(10), 10200–10232. https://doi.org/10.1109/JIOT.2020.2987070 Patel, C. M. (2024). Edge computing for low-latency loT applications in smart cities. Smart internet of things, 1(4), 282–288. https://doi.org/10.22105/siot.v1i4.251 Sulieman, N. A., Ricciardi Celsi, L., Li, W., Zomaya, A., & Villari, M. (2022). Edge-oriented computing: A survey on research and use cases. Energies, 15(2), 452. https://doi.org/10.3390/en15020452 Kong, L., Tan, J., Huang, J., Chen, G., Wang, S., Jin, X., … & Das, S. K. (2022). Edge-computing-driven internet of things: A survey. ACM computing surveys, 55(8), 1–41. https://doi.org/10.1145/3555308 Andriulo, F. C., Fiore, M., Mongiello, M., Traversa, E., & Zizzo, V. (2024). Edge computing and cloud computing for internet of things: A review. In Informatics (Vol. 11, No. 4, p. 71). MDPI. https://doi.org/10.3390/informatics11040071 Andriulo, F. C., Fiore, M., Mongiello, M., Traversa, E., & Zizzo, V. (2024, September). Edge computing and cloud computing for internet of things: A review. In Informatics (Vol. 11, No. 4, p. 71). MDPI. Jararweh, Y., Otoum, S., & Al Ridhawi, I. (2020). Trustworthy and sustainable smart city services at the edge. Sustainable cities and society, 62, 102394. https://doi.org/10.1016/j.scs.2020.102394 Pacheco, M. (2024). Edge computing and IoT: Key benefits & use cases. https://www.tierpoint.com/blog/edge-computing-and-iot/?utm_source=chatgpt.com Dallaf, A. A. A. (2025). Edge computing in IoT networks: Enhancing efficiency, reducing latency, and improving scalability. International journal of advanced network, monitoring and controls, 10(01), 103–115. https://doi.org/0009-0005-8580-0721 Mao, Y., You, C., Zhang, J., Huang, K., & Letaief, K. B. (2017). A survey on mobile edge computing: The communication perspective. IEEE communications surveys & tutorials, 19(4), 2322–2358. https://doi.org/10.1109/COMST.2017.2745201 Satyanarayanan, M. (2017). The emergence of edge computing. Computer, 50(1), 30–39. https://doi.org/10.1109/MC.2017.9 Yi, S., Li, C., & Li, Q. (2015). A survey of fog computing: Concepts, applications and issues [presentation]. Proceedings of the 2015 workshop on mobile big data (pp. 37–42). https://doi.org/10.1145/2757384.2757397 Chamola, V., Tham, C. K., & Ansari, N. (2020). An optimal delay aware task assignment scheme for wireless SDN networked edge cloudlets. Future generation computer systems, 102, 862–875. https://doi.org/10.1016/j.future.2019.09.003 He, R., Weizhong, M. A., Ma, X., & Liu, Y. (2021). Modeling and optimizing for operation of CO2-EOR project based on machine learning methods and greedy algorithm. Energy reports, 7, 3664–3677. https://doi.org/10.1016/j.egyr.2021.05.067 Tak, A., & Cherkaoui, S. (2020). Federated edge learning: Design issues and challenges. IEEE network, 35(2), 252–258. https://doi.org/10.1109/MNET.011.2000478 Yu, W., Liang, F., He, X., Hatcher, W. G., Lu, C., Lin, J., & Yang, X. (2017). A survey on the edge computing for the internet of things. IEEE access, 6, 6900–6919. https://doi.org/10.1109/ACCESS.2017.2778504 Materwala, H., Ismail, L., Shubair, R. M., & Buyya, R. (2022). Energy-SLA-aware genetic algorithm for edge- -cloud integrated computation offloading in vehicular networks. Future generation computer systems, 135, 205–222. https://doi.org/10.1016/j.future.2022.04.009 Baumgartner, T., Klatt, S., & Donath, L. (2023). Revealing the mutual information between body-worn sensors and metabolic cost in running. Sensors, 23(4), 1756. https://doi.org/10.3390/s23041756 Deb, R., & Roy, S. (2022). A comprehensive survey of vulnerability and information security in SDN. Computer networks, 206, 108802. https://doi.org/10.1016/j.comnet.2022.108802