Machine-Learning-Based Hazardous Spot Detection Framework by Mobile Sensing and Opportunistic Networks

Watanabe, Yoshito; Liu, Wei; Shoji, Yozo

doi:10.1109/tvt.2020.3021411

Cited by 10 publications

(4 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Watanabe et al [23] identify dangerous road spots using ML on mobile sensing and V2V networking. Their approach employs ML techniques and the Viterbi algorithm.…”

Section: Featuresmentioning

confidence: 99%

“…[18] Low-cost driver profiling ML algorithm investigation Simulation limitations, data validation Lee, J. [19] Assessing driving behaviors Utilization of sensor data Small trial, context validation needed C. C. Chang [20] Collision avoidance using dynamics Real-time object detection Single camera setup, broader validation Moses and Parvathi [21] ML-based traffic flow prediction ML process, stages breakdown Limited to US data, no new methodology Lyu et al [22] Improved collision alerts Novel model, simulation validation Specific scenario, broader context needed Watanabe et al [23] Identifying dangerous roadplaces Sensory data, ML application Real-world validation, generalization Silva et al [24] Various ML methods for collision Prediction overview, evaluation Data precision, lack of real-time data AbouElassad et al [25] Framework for analyzing behavior…”

Section: Algorithms Accuracy Measurementmentioning

confidence: 99%

See 1 more Smart Citation

Advancements in Machine Learning and Data Mining Techniques for Collision Prediction and Hazard Detection in Internet of Vehicles

Manchala,

Kishore

2024

PJBAS

View full text Add to dashboard Cite

The Internet of Vehicles (IoV) has transfigured transportation with connected vehicles, smart infrastructure, and self-driving cars. Road collisions and accidents are still a problem for road safety. This review of the literature discusses the prediction of IoV accidents and collisions as well as the detection of hazards using data mining, deep learning, and machine learning techniques. It describes the most recent developments to these methods and how they enhanced IoV safety. The article starts off by going over data collection, data quality, and the ever-changing nature of IoV traffic scenarios. What follows is a detailed breakdown of the ML, DL, and DM methods used in IoV safety applications. Convolutional neural networks, artificial neural networks, recurrent neural networks, support vector machines, and decision trees. As examples of real-world applications and case studies, intelligent accident prediction models, driver attention forecasting, traffic congestion forecasting, spatiotemporal analysis in autonomous vehicles, scene-graph embedding, and V2P collision risk alerts are discussed. The goal of this review is to give readers a comprehensive overview of the cutting-edge methods enhancing IoV accident prediction, collision avoidance, and hazard detection.

show abstract

“…Watanabe et al [23] identify dangerous road spots using ML on mobile sensing and V2V networking. Their approach employs ML techniques and the Viterbi algorithm.…”

Section: Featuresmentioning

confidence: 99%

Section: Algorithms Accuracy Measurementmentioning

confidence: 99%

Advancements in Machine Learning and Data Mining Techniques for Collision Prediction and Hazard Detection in Internet of Vehicles

Manchala,

Kishore

2024

PJBAS

View full text Add to dashboard Cite

show abstract

“…The most recent review of current technologies and open communication difficulties focused on 5G data transmission between BSs and V2X, as well as challenges of RBS attacks on Internet of Things (IoT) security [21,22]. We report on security issues for V2X scenarios, and synthetic data generation, and delineate the various RBS schemes in the 5G environment [23][24][25][26].…”

Section: Background and Related Workmentioning

confidence: 99%

Synthetic Generation of Realistic Signal Strength Data to Enable 5G Rogue Base Station Investigation in Vehicular Platooning

2022

View full text Add to dashboard Cite

Rogue Base Stations (RBS), also known as 5G Subscription Concealed Identifier (SUCI) catchers, were initially developed to maliciously intercept subscribers’ identities. Since then, further advances have been made, not only in RBSs, but also in communication network security. The identification and prevention of RBSs in Fifth Generation (5G) networks are among the main security challenges for users and network infrastructure. The security architecture group in 3GPP clarified that the radio configuration information received from user equipment could contain fingerprints of the RBS. This information is periodically included in the measurement report generated by the user equipment to report location information and Received Signal Strength (RSS) measurements for the strongest base stations. The motivation in this work, then is to generate 5G measurement reports to provide a large and realistic dataset of radio information and RSS measurements for an autonomous vehicle driving along various sections of a road. These simulated measurement reports can then be used to develop and test new methods for identifying an RBS and taking mitigating actions. The proposed approach can generate 20 min of synthetic drive test data in 15 s, which is 80 times faster than real time.

show abstract

“…Current research on opportunistic networks has mainly focused on the improvement of routing algorithms. Traditional opportunistic network routing algorithms mostly only consider the topological relationship of nodes (link prediction based on existing nodes and their connection attributes) but do not consider the social relationship between people in real social networks [7,15]. Therefore, it is challenging to effectively use the traditional socialist network in real social networks.…”

Section: Introductionmentioning

confidence: 99%

A Data Transmission Algorithm Based on Triangle Link Structure Prediction in Opportunistic Social Networks

et al. 2021

View full text Add to dashboard Cite

With the popularization of 5G communications, the scale of social networks has grown rapidly, and the types of messages have become increasingly complex. The rapid increases in the number of access nodes and the amount of data have put a greater burden on the transmission of information in the networks. However, when transferring data from a large number of users, the performance of traditional opportunistic network routing algorithms is insufficient, which often leads to problems such as high energy consumption, network congestion, and data packet loss. The way in which to improve this transmission environment has become a difficult task. Therefore, in order to ensure the effective transmission of data and reduce network congestion, this paper proposed a link prediction model based on triangular relationships in opportunistic social networks (LPMBT). In the topological structure of the social network, the algorithm scores links based on the frequency of use and selects the optimal relay node based on the score. It can also efficiently track the target node and reconstruct the sub-community. The simulation experimental results showed that the algorithm had excellent performance, effectively reduced overhead, reduced the end-to-end delay, and greatly improved the data transfer rate in the opportunistic network.

show abstract

Machine-Learning-Based Hazardous Spot Detection Framework by Mobile Sensing and Opportunistic Networks

Cited by 10 publications

References 33 publications

Advancements in Machine Learning and Data Mining Techniques for Collision Prediction and Hazard Detection in Internet of Vehicles

Advancements in Machine Learning and Data Mining Techniques for Collision Prediction and Hazard Detection in Internet of Vehicles

Synthetic Generation of Realistic Signal Strength Data to Enable 5G Rogue Base Station Investigation in Vehicular Platooning

A Data Transmission Algorithm Based on Triangle Link Structure Prediction in Opportunistic Social Networks

Contact Info

Product

Resources

About