The rise of Internet of Things (IoT) technologies significantly transforms urban traffic management, offering cities advanced tools for real-time traffic monitoring, flow optimization, and effective congestion reduction [1–3]. As IoT-enabled sensors and devices become more widely deployed across urban spaces, cities gain access to a constant stream of critical data on vehicle speeds, densities, and movement patterns. However, managing and analyzing such high data volumes in real-time requires a powerful computing infrastructure. Integrating cloud and edge computing allows cities to develop scalable and efficient systems capable of processing and analyzing traffic data at unprecedented speeds and accuracies [4], [5].

Fig. 1 shows a bar chart or line graph comparing latency times between cloud-only and hybrid cloud-edge setups in traffic management, emphasizing the lower latency of edge computing.

Edge computing enables localized data processing close to the source, often at the data collection site, allowing immediate responses to sudden traffic changes or incidents [6], [7]. This localized processing minimizes latency, ensuring critical adjustments, such as rerouting traffic or adjusting signal timings at high-congestion points, can happen in near real-time. By reducing the need to send all data to a central server, edge computing also optimizes bandwidth usage, vital in large-scale IoT networks where consistent, real-time data flow is necessary for smooth network operations.

Fig. 2 shows intelligent transportation systems for sustainable smart cities [8], [9].

On the other hand, cloud computing offers centralized data storage and powerful analytics, allowing cities to analyze data across broader timeframes and geographic scales. This centralized approach helps uncover larger traffic trends and patterns that might go unnoticed through edge computing, enabling urban planners and traffic management authorities to make informed, data-driven decisions. For instance, cloud-based analysis of historical traffic data can highlight recurring congestion patterns, informing targeted infrastructure improvements and strategic planning.

Fig. 3 depicts a simplified view of time traffic information.

Together, the hybrid architecture of cloud-edge computing overcomes the latency and bandwidth constraints that have historically limited centralized IoT systems. By assigning immediate data processing to edge devices and high-level analytics to the cloud, cities can effectively balance computing resources, enhance traffic flow, and proactively address bottlenecks. Consequently, IoT-enabled systems powered by cloud-edge collaboration are fostering smarter, more adaptable urban environments that can better respond to the increasing complexities and demands of urban traffic, ultimately contributing to safer, more efficient transportation networks.

Fig. 4 shows the comparison of traditional intelligence and edge intelligence from the perspective of implementation. In traditional intelligence, all data must be uploaded to a central cloud server, whilst in edge intelligence, intelligent application tasks are done at the edge with locally generated data in a distributed manner.

Integrating cloud and edge computing has gained prominence as a solution to the challenges faced in IoT-enabled traffic monitoring, particularly data volume, latency, and resource management. With the advent of IoT devices in urban settings, real-time traffic monitoring requires a balance between high-capacity data analysis and low-latency response times, both enabled by cloud and edge computing. This section reviews key studies on cloud-edge collaboration, highlighting its applications, advantages, and ongoing challenges in IoT-enabled traffic monitoring systems.

The proposed system architecture integrates IoT-enabled traffic sensors, edge processing units, and a central cloud server to create a streamlined solution for urban traffic management. This setup allows cities to gather, process, and analyze traffic data efficiently, helping to reduce congestion and improve overall mobility. The system starts with IoT devices deployed across urban areas—on traffic lights, roadways, and intersections— that continuously capture real-time raw data on variables like vehicle speed, traffic density, and flow patterns. This data, rich with potential insights, is first sent to nearby edge nodes for immediate processing. Data filtering, aggregation, and preprocessing occur at the edge, allowing for efficient handling of large data volumes before transmission.

Processing data at the edge—close to the source—offers several benefits. Reducing the data volume that needs to be sent to the cloud optimizes network bandwidth and significantly minimizes latency. This low-latency processing allows for near-instantaneous traffic insights and quicker decision-making, enabling a responsive and adaptive traffic management system to address real-time issues, such as rerouting traffic away from congestion points.

Once the edge nodes process the data, it is transmitted to the central cloud server, where data from multiple sources across the city is aggregated and stored for long-term analysis. The cloud server acts as a high-capacity repository that can perform complex analysis across extensive datasets, identifying trends, optimizing traffic flow, and predicting congestion patterns. This hybrid architecture of edge and cloud computing maximizes resource efficiency, enabling a scalable, data-driven approach to traffic management that balances immediate needs with long-term planning capabilities for sustainable urban mobility.

Fig. 5 illustrates the cloud-edge IoT traffic monitoring framework.

Fig. 6 illustrates edge computing.

This study implements a multi-layered network architecture that strategically positions edge nodes at critical traffic hotspots throughout the city, ensuring efficient data processing and management. These nodes can perform essential, localized processing tasks on real-time traffic data by situating edge devices close to data-generating sources such as intersections, highways, and high-traffic areas. This includes initial data filtering, aggregation, and basic analytics, which enable the system to offload a significant portion of data handling from the cloud. This setup reduces latency by allowing data to be processed closer to its point of origin. It minimizes the need for constant data transmission to the cloud, optimizing network efficiency and system response time in a manner suitable for real-time applications.

Lightweight communication protocols such as Message Queuing Telemetry Transport (MQTT) are employed to further streamline data transmission between edge nodes and the cloud server. MQTT is particularly optimized for IoT applications, offering reliable and low-overhead data transfer, crucial for managing the high-frequency data flow that IoT sensors generate. This protocol minimizes bandwidth usage while ensuring robust and stable data transfer across the network, addressing the unique challenges of large-scale urban IoT networks. By employing such protocols, the architecture can handle extensive data volumes without overwhelming the network infrastructure, enabling a sustainable and scalable solution for city-wide traffic management. This optimized data flow from the edge to the cloud enhances real-time traffic monitoring, facilitating rapid, data-driven decision-making and paving the way for smarter, more responsive urban infrastructure. Table 1 provides an overview of the specifications and roles of each primary component within this layered system. The cloud server is the central hub for large-scale data storage and advanced analytics, including traffic prediction, integral to long-term data management and strategic insights. Edge devices operate as localized processors that handle data filtering and preprocessing, transmitting only critical, summarized data to the cloud. Finally, IoT devices—a network of traffic sensors across urban areas— continuously capture real-time data on parameters like vehicle flow, speed, and traffic density. This multi-layered architecture enables efficient and scalable data processing, balancing workloads between the edge and cloud and ensuring streamlined, real-time urban traffic management.

Cloud Centralized servers for analysis and storage Traffic prediction and data storage Edge Local processors for initial data filtering Data filtering and preprocessing IoT devices Traffic sensors in urban areas Collect real-time traffic data

A comprehensive simulated analysis was conducted to evaluate and compare cloud-only architectures' latency and data transfer rates versus hybrid cloud-edge architectures in an urban traffic management context. The results demonstrated that the hybrid model significantly outperformed the traditional cloud-only approach, exhibiting a remarkable 30% decrease in latency. This improvement in response time is crucial for real-time traffic management applications, where timely data processing and decision-making are essential for effective congestion management and overall traffic flow optimization. Additionally, the hybrid architecture achieved a 25% reduction in data transfer requirements, enhancing network efficiency and minimizing bandwidth consumption, making it a more sustainable solution for managing the high volumes of data generated by IoT devices.

Fig. 7 provides a visual representation of the comparative latency results, illustrating the clear advantages of the hybrid model over the cloud-only setup. The improved latency metrics highlight the benefits of local edge processing, allowing faster data handling and quicker responses to changing traffic conditions. Meanwhile, Table 2 summarizes key efficiency metrics derived from the analysis, offering insights into the performance differences between the two architectures. These findings underscore the effectiveness of adopting a hybrid cloud-edge approach for urban traffic management systems, paving the way for smarter, more responsive urban environments that can effectively address the challenges of modern mobility.

Fig. 7 illustrates the comparative latency results, with Table 2 summarizing the efficiency metrics.

In this paper, we presented a comprehensive cloud-edge computing framework specifically designed for IoT-enabled traffic monitoring, illustrating how a hybrid model can effectively manage large volumes of data generated by numerous traffic sensors deployed across urban areas. The hybrid architecture reduces latency, allowing for faster real-time decision-making, and conserves bandwidth, crucial for maintaining efficient network operations under heavy data loads. This approach is particularly advantageous in scenarios where real-time data processing is essential, such as during peak traffic hours or in traffic incidents, as it enables immediate responses and dynamic traffic management.

This cloud-edge framework's scalability and efficiency make it an ideal solution for smart city applications, where the integration of IoT technologies can significantly enhance urban mobility and infrastructure management. By processing data locally at the edge and utilizing the cloud for long-term storage and advanced analytics, cities can balance immediate operational needs and strategic planning requirements.

Future research could enhance edge intelligence, enabling edge nodes to independently perform more sophisticated analytics and decision-making. Additionally, exploring dynamic resource allocation between cloud and edge layers could optimize performance further, allowing for adaptive resource management based on real-time demands. These advancements will be critical in developing more resilient and intelligent traffic management systems, ultimately contributing to the evolution of smart cities that prioritize efficiency, sustainability, and improved quality of life for residents.

This research received no external funding.

The author declares no conflicts of interest regarding the publication of this paper.

I thank our institutions for their support in conducting this research.