Privacy-preserving location-based traffic density monitoring

Wu, Lei; Xia, Wei; Meng, Lingzhen; Zhao, Shengnan; Wang, Hao

doi:10.1080/09540091.2021.1993137

Cited by 11 publications

(3 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this experiment we use five metrics to evaluate the algorithm: physical dispersion, false position set generation efficiency, semantic diversity, uncertainty and recognition rate. In addition, three other algorithms are selected for performance comparison: Dummy Location Selection (DLS) algorithm [20], Enhanced-DLS algorithm [21] and Maximum and Minimum Dummy Selection (MMDS) algorithm [22]. The physical dispersion of false location sets can be measured from the minimum distance between false locations and the area of anonymous regions.…”

Section: Experimental Results Of False Location Filtering Algorithm F...mentioning

confidence: 99%

Research on Location Privacy Protection Technology in Wireless Sensor Networks Based on Big Data

Zhang,

2023

JCSANDM

View full text Add to dashboard Cite

The digital footprint in wireless sensor networks can bring great academic and commercial value, but it will also bring the risk of privacy disclosure to users. This study discusses the location privacy protection methods in location based service under multiple scenarios. In the experiment, a false location filtering algorithm for real-time location request scenarios and a false path generation model for offline location release scenarios are proposed. The false position filtering algorithm is implemented based on the idea of a large top heap selection query. The algorithm can construct false position candidate sets and filter false positions. The false track generation model combines the false position technology and the generative adversarial networks model, which mainly protects the user’s track data by synthesizing tracks. In the attacker’s recognition experiment of a real location. The minimum distance between the false locations generated by the algorithm proposed in the study is above 400 m and the generation time does not exceed 5 ms, generating a better set of false locations in terms of both effectiveness and efficiency. Compared with several commonly used privacy-preserving algorithms, the proposed algorithm has the lowest probability of being identified with real locations, with no more than 21% overall, and is almost independent of the k value. the recognition accuracy of the trajectory user link task decreases from over 90% to about 34%, indicating that the proposed fake trajectory generation model can effectively protect users’ data privacy. The experimental results demonstrate that the algorithm and model proposed in the study can quickly generate physically dispersed and semantically diverse sets of fake locations and effectively protect users’ trajectory privacy, which is important for users’ digital footprint privacy protection.

show abstract

Section: Experimental Results Of False Location Filtering Algorithm F...mentioning

confidence: 99%

Research on Location Privacy Protection Technology in Wireless Sensor Networks Based on Big Data

Zhang,

2023

JCSANDM

View full text Add to dashboard Cite

show abstract

“…Wu, Wei, Meng, Zhao, and Wang suggested a traffic density monitoring model, which takes the authentic location information from the vehicles themselves as well as maintains the privacy of the information [31]. To achieve this, the authors used a pseudonym and location anonymization servers, such that vehicles' location and identification were distinctly recorded.…”

Section: Related Workmentioning

confidence: 99%

Identification of Location for Police Headquarters to Deal with Crime Against Women in India Using Clustering Based on K-Means Algorithm

Indu,

Dimri,

Kumar

2023

IJCDS

View full text Add to dashboard Cite

During 2020-21, the rate of crime against women has considerably increased, especially the cases of domestic violence, kidnapping, and rape. Considering the fact that a woman better understands the problem of another woman, and to ensure women's safety in this era of technology, we proposed opening Police Headquarters with lady staff only. For this, we suggest the use of the latitude, longitude, and average weighted crime as clustering parameters for the K-means method to recognize the most probable location at which such headquarters can be opened. The parameter average weighted crime signifies the magnitude or intensity of nine different possible crimes against women and thus identifies the location where the maximum number of crimes against women are reported. Therefore, such locations are ideal for setting up the Lady Police Headquarters to further curtail any such crimes, and deliver swift protection as well as justice to the victim women. It is somehow similar to identifying clusters (locality) from clusters (blocks). Thence, using the suggested approach, the Indian law enforcing agencies including police can prevent and control crimes at the micro level and can also increase their reach.

show abstract

“…For example, a system that can recognize license plates may construct patterns of behavior of the corresponding vehicles' owners and use that for various malicious or illegal purposes. In fact, privacy in traffic monitoring and analysis has been studied for a while, but mostly in the context of localization and location-based services [15,23,22,20,25]. Privacy in vision-based traffic analysis has been much less studied, and privacy related to license plates is a current topic of debate among privacy advocates, legal scholars, and lawmakers [1,11].…”

Section: Introductionmentioning

confidence: 99%

License Plate Privacy in Collaborative Visual Analysis of Traffic Scenes

Alvar¹,

Korcan²,

Bajić³

2022

Preprint

View full text Add to dashboard Cite

Traffic scene analysis is important for emerging technologies such as smart traffic management and autonomous vehicles. However, such analysis also poses potential privacy threats. For example, a system that can recognize license plates may construct patterns of behavior of the corresponding vehicles' owners and use that for various illegal purposes. In this paper we present a system that enables traffic scene analysis while at the same time preserving license plate privacy. The system is based on a multi-task model whose latent space is selectively compressed depending on the amount of information the specific features carry about analysis tasks and private information. Effectiveness of the proposed method is illustrated by experiments on the Cityscapes dataset, for which we also provide license plate annotations.

show abstract

Privacy-preserving location-based traffic density monitoring

Cited by 11 publications

References 31 publications

Research on Location Privacy Protection Technology in Wireless Sensor Networks Based on Big Data

Research on Location Privacy Protection Technology in Wireless Sensor Networks Based on Big Data

Identification of Location for Police Headquarters to Deal with Crime Against Women in India Using Clustering Based on K-Means Algorithm

License Plate Privacy in Collaborative Visual Analysis of Traffic Scenes

Contact Info

Product

Resources

About