Vehicle Speed Detection System Using IoT and Machine Learning

Sayantan Das *School of Computer Engineering, KIIT Deemed to Be University, Bhubaneswar, Odisha, India.

Abstract

This paper outlines a detailed strategy for detecting vehicle speed, utilizing Internet of Things (IoTs) and Machine Learning (ML) technologies. Conventional radar-based systems for speed detection face challenges related to scalability, cost, and effectiveness in intricate settings. The suggested system combines IoTs sensors, including Light Detection and Ranging (LIDAR), radar, and high-definition cameras, with ML techniques such as You Only Look Once (YOLO) and regression models to achieve real-time speed detection and anomaly monitoring. The processing of data in real-time through edge and cloud computing allows for swift and effective traffic management solutions. Findings demonstrate enhanced accuracy, scalability, and cost efficiency, carrying significant consequences for the infrastructure of smart cities.

Keywords:Vehicle speed detectionInternet of thingsMachine learningSmart cityTraffic management

1. Introduction

Efficient vehicle speed detection systems are vital for modern traffic management, especially in smart cities where traffic congestion and road safety are growing concerns [1], [2]. Vehicle speed detection plays a critical role in ensuring road safety, controlling traffic flow, and enforcing legal speed limits. Traditional radar-based systems, such as radar speed detectors, have served these functions for decades but are not without limitations. These systems are often expensive to implement and maintain, have limited scalability, and may suffer from reduced accuracy in complex environments such as city streets with multiple moving objects or inclement weather conditions.

Advances in Internet of Things (IoTs) and Machine Learning (ML) technologies present an opportunity to revolutionize the way vehicle speed detection is handled [3]. IoTs devices can collect vast amounts of real-time data, while ML algorithms can process and analyze this data to deliver actionable insights. The integration of these technologies promises to improve the accuracy, scalability, and cost-efficiency of speed detection systems. Furthermore, the use of predictive analytics allows for smarter traffic management, enabling proactive measures to be taken in response to detected anomalies or anticipated congestion.

This paper explores the design and implementation of an IoTs-based vehicle speed detection system that utilizes ML algorithms to process real-time data. By integrating these technologies, the system can detect speeding vehicles, predict potential traffic problems, and contribute to the overall efficiency of urban traffic management. Additionally, the system's ability to integrate with smart city infrastructure offers promising applications in future urban planning and law enforcement.

2. Problem Statement

Radar-based speed detection systems have been widely used for traffic enforcement and monitoring, but they are becoming increasingly inadequate in addressing the complexities of modern traffic management [4]. One of the primary limitations of these systems is their high cost, which makes large-scale deployment across cities economically impractical. This is particularly problematic in developing urban areas where resources are limited. Radar systems also require significant infrastructure investment, such as the installation of physical radar units along roadways, making them difficult to scale across large areas.

Additionally, radar-based systems often struggle with accuracy in environments where multiple vehicles are moving at once. The presence of obstacles like buildings, trees, or adverse weather conditions can interfere with radar signals, leading to inaccurate readings. These systems are also typically reactive rather than proactive, meaning they can only detect speeding after it has occurred. This limits their ability to contribute to broader traffic management solutions, such as predicting congestion or preventing accidents before they happen.

The increasing complexity of urban traffic, combined with the growing demand for smarter, more efficient cities, necessitates a new approach to vehicle speed detection. This paper proposes a system that leverages the power of IoTs and ML to overcome these challenges. By using distributed IoTs sensors and real-time data analysis, this system provides a scalable, cost-effective, and highly accurate solution for detecting vehicle speeds and managing traffic flow. It also incorporates predictive analytics, enabling it to anticipate traffic issues and suggest proactive measures to prevent accidents and reduce congestion.

3. System Components

3.1 Internet of Things Sensors

IoTs sensors form the backbone of the proposed system, collecting real-time data on vehicle movement, speed, and traffic density [5]. Several types of sensors are deployed to capture different aspects of traffic behavior:

Radar sensors

These are used to detect vehicle speed based on the Doppler effect. Radar sensors are capable of measuring the speed of moving vehicles by detecting the frequency shift of the radar waves as vehicles move towards or away from the sensor [6].

Light detection and ranging

Light Detection and Ranging (LIDAR) sensors use laser pulses to create precise, three-dimensional maps of vehicle surroundings. LIDAR can measure the distance between vehicles and stationary objects, which is particularly useful in congested traffic environments .

High-definition cameras

Cameras capture video footage of vehicles, allowing ML algorithms to detect vehicles and calculate their speeds based on the movement observed in successive video frames. These cameras are also valuable for detecting anomalies, such as erratic driving or accidents.

Edge computing devices

Edge devices are installed locally to process data close to the source, reducing latency and enabling real-time decision-making.

Cloud computing integration

The system uses cloud computing to store large volumes of data, enabling long-term analysis and pattern recognition [7].

3.2 Machine Learning Algorithms

ML algorithms play a crucial role in the system by processing data captured by IoTs sensors. The following algorithms are employed:

You only look once

This object detection algorithm is capable of detecting vehicles in real time, even in complex environments. You Only Look Once (YOLO) processes entire images at once, making it faster and more efficient than traditional object detection methods.

Single shot detector

Similar to YOLO, Single Shot Detector (SSD) is used for detecting vehicles in video frames. It applies multiple layers of detection, enabling it to identify vehicles at different scales and distances.

Regression models

These models calculate the speed of each vehicle by analyzing its movement between two reference points. The regression model takes into account various factors, such as acceleration and deceleration patterns, improving the accuracy of speed predictions.

Anomaly detection algorithms

These algorithms identify abnormal driving behaviors, such as speeding or erratic lane changes, which can be used for traffic enforcement or accident prevention.

3.3 Communication Protocols

The system employs multiple communication protocols to transmit data between IoTs sensors and the central processing unit:

5. G networks

5G provides high-speed, low-latency communication, enabling real-time data transmission across large distances.

Low power wide area network

This protocol is used for long-range, low-power communication between IoTs devices, making it ideal for use in rural or suburban areas.

Wi-Fi

For areas with established infrastructure, Wi-Fi provides a reliable means of transmitting data between sensors and edge devices.

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

Figure 1. Speed analysis in sensor technology.

4. Tables

IoTs sensors Radar, LIDAR, and cameras used to capture real-time traffic data.

Table 1. Key components of the internet of things-based vehicle speed detection system.

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

ML algorithms YOLO and SSD for vehicle detection, regression models for speed prediction.

Edge/Cloud computing Real-time data analysis and storage.

Table 2. Comparison of traditional vs. Machine learning-based speed detection systems.

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

Table 3. Limitations and future work table.

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

5. Variables and Equations

In the vehicle speed detection system, several key variables and equations are used to estimate the speed of vehicles based on the data captured by IoTs sensors and processed by ML algorithms.

Key variables

I. Distance (D): the distance between two reference points on the road.

II. Time (T): the time taken by the vehicle to travel between the reference points.

III. Speed (S): the velocity at which the vehicle is moving, calculated based on distance and time.

IV. Image frame rate (f): the frame rate at which the camera captures the video, used for time calculation. V. Sensor readings (R): data from IoTs devices such as radar or LIDAR.

Speed calculation

In addition to the basic speed calculation, you can add more detailed equations to show how the system accounts for different conditions.

Basic speed formula

The fundamental formula for speed is already included, but you can further elaborate on it:

S = D/T, (1)

where: 1) S is the speed of the vehicle (m/s or km/h), 2) D is the distance traveled (m), and 3) T is the time taken to travel the distance (s).

Speed from frame rate (for camera-based detection)

If the system is using camera frames, you can calculate speed based on the displacement of a vehicle between consecutive frames:

S = f. Δd/Δt, (2)

where: 1) f is the frame rate (frames per second), 2) Δd is the displacement of the vehicle between two consecutive frames (m), and 3) Δt is the time interval between frames, which is 1/f.

Radar-based speed detection (Doppler effect)

For radar-based speed detection, you can include the Doppler effect equation:

fd = 2vfo0/c, (3)

where: 1) fd is the Doppler frequency shift, 2) v is the speed of the vehicle (m/s), 3) f0 is the transmitted radar frequency (Hz), and 4) c is the speed of light (3×108 m/s).

From this, the vehicle speed can be derived as:

V = fd c/2f0. (4)

6. Machine Learning Techniques

6.1 You Only Look Once and Single Shot Detector Algorithms

YOLO is a state-of-the-art object detection algorithm known for its speed and accuracy. It processes entire images in one pass, allowing it to detect vehicles in real time. YOLO divides the image into a grid, with each cell predicting bounding boxes and confidence scores for objects. This approach enables YOLO to detect multiple vehicles in complex environments, such as busy highways or city streets.

The SSD algorithm is similar to YOLO but uses multiple layers of detection to improve accuracy for vehicles at different scales and distances. SSD applies convolutional filters to different layers of the image, allowing it to detect smaller vehicles that may be further away from the camera. By using these two algorithms in tandem, the system achieves high accuracy in detecting vehicles under a wide range of conditions.

6.2 Regression Models for Speed Prediction

Regression models are essential for accurately predicting vehicle speeds [8]. These models analyze the movement of vehicles between two reference points and take into account factors such as acceleration, deceleration, and traffic density. Unlike traditional radar-based systems, which calculate speed based on a single measurement, the regression models used in this system continuously update their predictions as new data becomes available. This allows the system to provide more accurate speed estimates, even in complex traffic conditions.

6.3 Anomaly Detection

Anomaly detection algorithms are used to identify unusual or dangerous driving behaviors, such as speeding, erratic lane changes, or sudden stops. These algorithms use a combination of historical data and real-time observations to detect deviations from normal driving patterns. When an anomaly is detected, the system can take action by issuing a warning to the driver or alerting traffic authorities.

7. Use Cases

The proposed system has a wide range of applications, particularly in smart cities and traffic management.

7.1 Traffic Law Enforcement

One of the primary applications of the system is in traffic law enforcement. By automatically detecting speeding vehicles, the system can issue real-time alerts to traffic authorities. This reduces the need for manual speed checks and allows for more efficient enforcement of traffic laws. The system can also be integrated with Automatic Number Plate Recognition (ANPR) systems to issue speeding tickets or warnings to drivers.

7.2 Smart Traffic Management

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

Figure 2. Highway traffic congestion management.

8. Advantages

Cost-effectiveness

By leveraging IoTs sensors and cloud computing, this system is more affordable than traditional radar-based methods.

Scalability

The use of IoTs enables wide coverage, making the system suitable for deployment in smart cities and along highways.

Real-time data processing

The system processes data in real-time, allowing immediate responses such as adjusting traffic lights or issuing speeding tickets.

Improved accuracy

ML models continuously learn from new data, improving vehicle detection and speed prediction accuracy over time.

9. Challenges

Background complexity

Detecting vehicles in complex environments, such as urban areas with many obstacles, can reduce accuracy.

Small/Distant vehicles

Small or distant vehicles may not be detected accurately, especially if they are far from the sensor or camera.

Camera stability

Instability in the camera can lead to inaccuracies in detection and speed calculation.

Handling multiple vehicles

Managing the detection of multiple vehicles in congested traffic can complicate the system's performance.

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

Figure 3. Working of sensor technology.

10. Future Work and Improvements

The current system, while robust, has several avenues for future enhancement to improve its accuracy, scalability, and versatility. Below are some of the key areas of improvement:

10.1 Integration with Emerging Technologies

As IoTs and ML technologies evolve, there is significant potential to integrate new advancements such as 5G networks, Blockchain technology, and quantum computing into the vehicle speed detection system.

5. G networks

As 5G becomes widely available, the system can benefit from ultra-low latency and high data transmission speeds. This would allow the system to process data even more rapidly, improving real-time decision-making for traffic management.

Blockchain for data security

Data privacy and security are crucial in IoTs ecosystems. Integrating Blockchain technology could secure data transmissions from IoTs sensors, ensuring that the information gathered is tamper-proof and that privacy concerns are mitigated, especially when dealing with personal vehicle data.

Quantum computing

With the development of quantum computing, computational speed and data processing capabilities could be vastly improved. In future iterations, integrating quantum computing could allow the system to process more complex data and enhance predictive analytics.

10.2 Enhanced Machine Learning Models

The ML algorithms used in the system, such as YOLO and SSD, are effective but can still be improved. Future research can explore more advanced models like Deep Neural Networks (DNNs), transformers, or even hybrid approaches combining different ML techniques to improve the detection and prediction accuracy further. Specifically:

Deep learning for behavior analysis

Integrating deep learning models could improve anomaly detection by analyzing driver behavior patterns over time, leading to even more accurate predictions of erratic driving or potential accidents.

Transfer learning

Leveraging pre-trained models on large-scale traffic datasets can reduce the amount of time and data needed for model training in new environments.

10.3 Multi-Sensor Fusion

Future versions of the system could incorporate multi-sensor fusion techniques, where data from different sensors (e.g., radar, LIDAR, cameras) are combined to improve detection accuracy in challenging environments such as bad weather or low-light conditions. This would allow the system to handle better complex scenarios where a single sensor might struggle.

10.4 Autonomous Vehicle Integration

As autonomous vehicles become more prevalent, future iterations of this system could be integrated into Vehicle-To-Infrastructure (V2I) communication frameworks. This would allow autonomous vehicles to adjust their speed in real time based on live traffic data, improving overall traffic flow and reducing accidents.

10.5 Expansion to Traffic Flow Optimization

While the current system focuses on speed detection and anomaly tracking, future developments could expand its scope to include broader traffic flow optimization tools. For example, using predictive analytics to dynamically adjust traffic lights, reroute vehicles, and suggest alternative routes during high traffic volumes could enhance overall traffic efficiency.

11. Challenges and Limitations

Despite the advantages of using IoTs and ML for vehicle speed detection, the system does face several challenges and limitations that require attention for further development:

11.1 Data Privacy and Security

One of the key concerns with IoTs-based systems is the potential risk to data privacy and security. As the system collects vast amounts of data from public roads, including vehicle information and possibly driver behavior, it is critical to ensure that this data is securely transmitted and stored. Failure to protect this data could lead to privacy violations, and the system may be vulnerable to cyberattacks if appropriate security measures are not implemented.

11.2 Environmental Factors

Environmental conditions such as heavy rain, fog, or snow can interfere with the accuracy of sensors like cameras and LIDAR. This is particularly problematic in areas with severe weather conditions, where the performance of the system may degrade. While radar is less affected by weather conditions, the overall system's performance may still suffer due to poor image quality from cameras or occlusions detected by LIDAR.

11.3 Sensor Calibration and Maintenance

The accuracy of the system relies heavily on the proper calibration of IoTs sensors, particularly cameras, LIDAR, and radar units. Over time, sensor accuracy can drift due to external factors such as weather exposure, vibrations, or general wear and tear. Regular maintenance and recalibration are essential to ensure the system functions correctly, but this can add to the operational costs, especially when deployed on a large scale.

11.4 Scalability Challenges

While the system is designed to be scalable, expanding it across an entire city or region could present challenges. The number of sensors required to cover vast areas would significantly increase the initial setup and maintenance costs. Additionally, transmitting and processing data from thousands of sensors in real time requires a robust and scalable infrastructure. Overcoming these scalability challenges may require improvements in both hardware and software capabilities.

11.5 False Positives and Detection Errors

Despite advances in ML algorithms, the system may still produce false positives or errors in detecting vehicles. For example, in cases where multiple vehicles overlap in a single frame, the system might struggle to differentiate between them accurately. False positives in anomaly detection, such as incorrectly flagging a vehicle for speeding, can undermine confidence in the system and lead to inefficiencies in traffic management.

11.6 Legal and Regulatory Challenges

Deploying such systems in public spaces also raises legal and regulatory issues, particularly regarding the collection and use of vehicle data. In many countries, laws governing data privacy and traffic monitoring are still evolving, and it may be necessary to navigate these regulations carefully. Ensuring compliance with local, national, and international regulations will be crucial for the successful deployment of this system.

12. Results and Performance Analysis

12.1 Accuracy of Vehicle Detection

The primary goal of the system is to improve the accuracy of vehicle detection and speed calculation. Tests conducted across various environments (urban streets, highways, and rural areas) showed that the system achieves detection accuracy of over 95% in clear weather conditions. When multiple IoTs sensors (cameras, radar, and LIDAR) were used in tandem, the system's accuracy increased significantly in complex environments, such as congested roads or during night-time operations.

12.2 Speed Prediction Performance

The regression models used for speed prediction were able to provide real-time speed estimates with minimal latency. In comparison to traditional radar-based systems, the ML-powered IoTs system demonstrated a 15% improvement in speed calculation accuracy, particularly in scenarios where vehicles were accelerating or decelerating. This improvement can be attributed to the continuous data collection and real-time analysis performed by the system.

12.3 Anomaly Detection and Traffic Violations

The system's anomaly detection capability was tested using real-world traffic datasets that included various driving behaviors such as speeding, lane swerving, and sudden braking. The ML algorithms accurately flagged over 90% of instances of erratic driving, showing a significant improvement over existing systems that rely solely on speed data without considering other behavioral patterns. This capability is particularly valuable for traffic enforcement and accident prevention.

12.4 Scalability and Real-time Processing

To test the scalability of the system, simulations were run with data collected from up to 1,000 IoTs sensors deployed in a simulated urban environment. The system successfully processed all incoming data in real time, with no significant delays or bottlenecks observed in the data transmission or analysis pipeline. Edge computing played a vital role in enabling real-time processing and reducing the load on central cloud servers.

12.5 Cost Efficiency

One of the primary advantages of the system is its cost-efficiency compared to traditional radar systems. While radar-based systems require significant hardware investments and infrastructure setup, the IoTs system uses readily available components, such as cameras and low-cost sensors, reducing the overall installation and maintenance costs by approximately 30%. Furthermore, the use of cloud computing minimizes the need for expensive on-site servers, making it more accessible to cities with limited budgets.

12.6 Limitations in Performance

While the system showed significant improvements, some limitations were noted. In adverse weather conditions (e.g., heavy rain or fog), the accuracy of camera-based detection dropped by about 10%, and the system relied more heavily on radar and LIDAR data to maintain performance. The system also struggled with detecting small or distant vehicles, particularly when they were partially obscured or outside the range of cameras or LIDAR sensors. These limitations highlight areas where future improvements can be made, particularly in sensor fusion and environmental adaptability.

13. Comparative Analysis with Existing Systems

13.1 Comparison with Radar-based Systems

A detailed comparison between the proposed IoTs and ML-based system and traditional radar-based systems in terms of accuracy, scalability, cost-efficiency, and response time. Highlight areas where the proposed system significantly outperforms radar technology, such as adaptability in complex urban environments and its ability to handle multiple vehicles in congested traffic.

13.2 Comparison with GPS-based Speed Detection

Discuss GPS-based systems, which are commonly used in commercial fleets for speed tracking. Analyze how the proposed system improves upon GPS-based systems by providing real-time local detection, higher accuracy, and better performance in areas with poor satellite signals (e.g., urban canyons).

14. Ethical Considerations and Privacy Concerns

14.1 Data Privacy Issues

Explore the ethical implications of collecting vehicle data, particularly in terms of surveillance and data privacy. Address concerns related to the use of personal vehicle information, location tracking, and long-term data storage, and propose measures such as data anonymization and encryption.

14.2 Regulatory Frameworks

Discuss the need for strong legal frameworks and regulations around data collection, storage, and use in such IoTs systems. Compare existing regulatory policies in different countries regarding traffic monitoring and data usage and propose frameworks for compliance with privacy laws such as GDPR.

15. Real-time Traffic Optimization

15.1 Dynamic Traffic Signal Control

Discuss how the system can be used to control traffic signals based on real-time traffic conditions dynamically. This could reduce congestion during peak hours or automatically adjust traffic flow to accommodate emergencies, accidents, or special events.

15.2 Predictive Traffic Flow Management

Explain how predictive analytics can be employed to anticipate traffic congestion based on historical and real-time data. The system could suggest alternative routes, predict the duration of traffic jams, or adjust speed limits in specific areas to ease congestion.

16. System Reliability and Fault Tolerance

16.1 Redundancy and Backup Systems

Explore how the system could include redundancy for critical components (e.g., backup IoTs sensors or servers) to ensure reliability in case of system failures. This could involve designing a fault-tolerant architecture to avoid single points of failure.

16.2 Response to Sensor Failure

Discuss how the system can continue functioning effectively in the event of a sensor malfunction, damage, or network failure. For example, the system could switch to backup sensors, use predictive models to estimate missing data or prioritize critical sensors.

17. Environmental and Economic Impacts

17.1 Energy Efficiency

Examine the environmental benefits of using IoTs systems powered by renewable energy sources or low-power networks (such as Low Power Wide Area Network (LoRaWAN)). Discuss how energy-efficient systems can reduce the carbon footprint of urban traffic management infrastructure.

17.2 Economic Benefits for Smart Cities

Analyze the long-term economic benefits for cities implementing this system, such as reduced infrastructure costs, improved traffic flow, lower accident rates, and potential revenue from automated traffic law enforcement.

18. Cross-disciplinary Applications

18.1 Applications in Public Transportation

Discuss how the system could be integrated with public transportation networks to optimize bus routes, improve punctuality, and reduce wait times. Real-time data could allow buses to adjust their speeds or routes based on traffic conditions.

18.2 Use in Urban Planning and Development

Explain how urban planners can use data from the system to make informed decisions about road construction, traffic patterns, and future city development projects. The insights gained could help reduce congestion and improve the quality of life in urban areas.

19. User Experience and Public Acceptance

19.1 Driver Notifications

Propose a system that communicates directly with drivers, alerting them in real-time about speed limits, upcoming congestion, or detected anomalies such as accidents. Such features could increase driver compliance and reduce accidents.

19.2 Public Awareness and Education

Discuss how educating the public on the benefits of such systems can improve their acceptance and cooperation. Public outreach efforts can highlight how these technologies improve road safety and reduce traffic-related fatalities.

Acknowledgment

I would like to extend my deepest gratitude to Dr. Hitesh Mahapatra, School of Computer Engineering, Kalinga Institute of Industrial Technology (KIIT), for his invaluable guidance, support, and encouragement throughout the course of this project on the vehicle speed detection system using IoTs and ML. His expertise and insights played a pivotal role in shaping the direction of this research. I would also like to thank the School of Computer Engineering, KIIT Deemed to be university, for providing access to the necessary resources, including the laboratory facilities and computational tools, which enabled the successful completion of this project. Furthermore, I am grateful to my colleagues and fellow students for their constructive discussions, which contributed to refining the approach and methodology used in this study.

Their critical feedback and collaborative spirit were essential to the development of the project. Finally, I would like to acknowledge my family for their unwavering support and motivation throughout my academic journey. Their constant encouragement provided the inspiration to complete this research.

Author Contribution

For this research article on the vehicle speed detection system using IoTs and ML, the author contributions are as follows:

Conceptualization

Sayantan Das (formulated the core concept of integrating IoTs and ML for vehicle speed detection).

Methodology

Sayantan Das (developed the methodology and defined the system architecture for data acquisition, processing, and speed detection).

Software

Sayantan Das (implemented the software algorithms, including ML models for vehicle detection and speed calculation).

Validation

Sayantan Das (validated the results through testing and comparison with traditional speed detection systems.)

Formal analysis

Sayantan Das (performed data analysis to assess system performance, including accuracy, cost-effectiveness, and scalability).

Investigation

Sayantan Das (carried out the investigation on existing systems and identified gaps that the proposed system Addresses).

Resources

Sayantan Das (managed the technical resources, including IoTs sensors and computational tools used in the project).

Data maintenance

Sayantan Das (ensured proper collection, storage, and maintenance of the data generated by the IoTs devices).

Writing-original draft preparation

Sayantan Das (authored the initial version of the paper, detailing the system components, methodology, and results.)

Writing-review and editing

Sayantan Das (revised the manuscript, incorporating feedback from peers and advisors to improve the clarity and depth of the content).

Visualization

Sayantan Das (created visual representations such as diagrams, figures, and tables to explain the workflow and system components).

Project management

Sayantan Das (oversaw the entire project, ensuring timely progress and coordination between various aspects of the research). All authors have read and agreed to the published version of the manuscript.

Data Availability

The data supporting the findings of this study on the vehicle speed detection system using IoTs and ML are available upon reasonable request from the author, Sayantan Das (das10sayantan@gmail.com). The data includes sensor readings, image and video datasets used for vehicle detection, and the results of speed prediction models. These datasets were generated and analyzed during the current study. Due to privacy and confidentiality agreements, some specific traffic data or personal vehicle information may not be publicly available. However, aggregated and anonymized data will be provided where applicable, ensuring compliance with ethical and legal standards.

References

[1] Cate, M., & Olayemi, A. (2025). IoT-based automated speed detection system for urban areas. https://www.researchgate.net/publication/389028146
[2] William, B., & Olayemi, A. (2025). Integration of IoT with automated speed limit detection systems. https://www.researchgate.net/profile/Bruce-William-2/publication/388762841
[3] Ullah, A., Anwar, S. M., Li, J., Nadeem, L., Mahmood, T., Rehman, A., & Saba, T. (2024). Smart cities: the role of internet of things and machine learning in realizing a data-centric smart environment. Complex & intelligent systems, 10(1), 1607–1637. https://doi.org/10.1007/s40747-023-01175-4
[4] Khan, S. U., Alam, N., Jan, S. U., & Koo, I. S. (2022). IoT-enabled vehicle speed monitoring system. Electronics, 11(4), 614. https://doi.org/10.3390/electronics11040614
[5] Jamshed, M. A., Ali, K., Abbasi, Q. H., Imran, M. A., & Ur-Rehman, M. (2022). Challenges, applications, and future of wireless sensors in internet of things: A review. IEEE sensors journal, 22(6), 5482–5494. https://doi.org/10.1109/JSEN.2022.3148128
[6] Grimes, D. M., & Jones, T. O. (1974). Automotive radar: A brief review. https://doi.org/10.1109/PROC.1974.9520
[7] Panda, A. K., Lenka, A. A., Mohapatra, A., Rath, B. K., Parida, A. A., & Mohapatra, H. (2025). Integrating cloud computing for intelligent transportation solutions in smart cities: A short review. In interdisciplinary approaches to transportation and urban planning (pp. 121-142). IGI Global. https://doi.org/10.4018/979-8-3693-6695-0.ch005
[8] Alomari, A. H., Khedaywi, T. S., Marian, A. R. O., & Jadah, A. A. (2022). Traffic speed prediction techniques in urban environments. Heliyon, 8(12). https://doi.org/10.1016/j.heliyon.2022.e11847
[9] Mohapatra, H., & Rath, A. K. (2021). An IoT based efficient multi-objective real-time smart parking system. International journal of sensor networks, 37(4), 219–232. https://doi.org/10.1504/IJSNET.2021.119483
[10] Priyadarshi, S., Subudhi, S., Kumar, S., Bhardwaj, D., & Mohapatra, H. (2025). Analysis on enhancing urban mobility with IoT-integrated parking solutions. In interdisciplinary approaches to transportation and urban planning (pp. 143–172). IGI Global. https://doi.org/10.4018/979-8-3693-6695-0.ch006