Mobility-Aware Privacy-Preserving Mobile Crowdsourcing
The increasing popularity of smartphones and location-based service (LBS) has brought us a new experience of mobile crowdsourcing marked by the characteristics of network-interconnection and information-sharing. However, these mobile crowdsourcing applications suffer from various inferential attacks...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-04-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/21/7/2474 |
| _version_ | 1797538951029850112 |
|---|---|
| author | Guoying Qiu Yulong Shen Ke Cheng Lingtong Liu Shuiguang Zeng |
| author_facet | Guoying Qiu Yulong Shen Ke Cheng Lingtong Liu Shuiguang Zeng |
| author_sort | Guoying Qiu |
| collection | DOAJ |
| description | The increasing popularity of smartphones and location-based service (LBS) has brought us a new experience of mobile crowdsourcing marked by the characteristics of network-interconnection and information-sharing. However, these mobile crowdsourcing applications suffer from various inferential attacks based on mobile behavioral factors, such as location semantic, spatiotemporal correlation, etc. Unfortunately, most of the existing techniques protect the participant’s location-privacy according to actual trajectories. Once the protection fails, data leakage will directly threaten the participant’s location-related private information. It open the issue of participating in mobile crowdsourcing service without actual locations. In this paper, we propose a mobility-aware trajectory-prediction solution, TMarkov, for achieving privacy-preserving mobile crowdsourcing. Specifically, we introduce a time-partitioning concept into the Markov model to overcome its traditional limitations. A new transfer model is constructed to record the mobile user’s time-varying behavioral patterns. Then, an unbiased estimation is conducted according to Gibbs Sampling method, because of the data incompleteness. Finally, we have the TMarkov model which characterizes the participant’s dynamic mobile behaviors. With TMarkov in place, a mobility-aware spatiotemporal trajectory is predicted for the mobile user to participate in the crowdsourcing application. Extensive experiments with real-world dataset demonstrate that TMarkov well balances the trade-off between privacy preservation and data usability. |
| first_indexed | 2024-03-10T12:38:27Z |
| format | Article |
| id | doaj.art-34aad927d84e46c59f46f180eebdff24 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-10T12:38:27Z |
| publishDate | 2021-04-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-34aad927d84e46c59f46f180eebdff242023-11-21T14:01:38ZengMDPI AGSensors1424-82202021-04-01217247410.3390/s21072474Mobility-Aware Privacy-Preserving Mobile CrowdsourcingGuoying Qiu0Yulong Shen1Ke Cheng2Lingtong Liu3Shuiguang Zeng4Shaanxi Key Laboratory of Network and System Security, School of Computer Science and Technology, Xidian University, Xi’an 710071, ChinaShaanxi Key Laboratory of Network and System Security, School of Computer Science and Technology, Xidian University, Xi’an 710071, ChinaShaanxi Key Laboratory of Network and System Security, School of Computer Science and Technology, Xidian University, Xi’an 710071, ChinaShaanxi Key Laboratory of Network and System Security, School of Computer Science and Technology, Xidian University, Xi’an 710071, ChinaShaanxi Key Laboratory of Network and System Security, School of Computer Science and Technology, Xidian University, Xi’an 710071, ChinaThe increasing popularity of smartphones and location-based service (LBS) has brought us a new experience of mobile crowdsourcing marked by the characteristics of network-interconnection and information-sharing. However, these mobile crowdsourcing applications suffer from various inferential attacks based on mobile behavioral factors, such as location semantic, spatiotemporal correlation, etc. Unfortunately, most of the existing techniques protect the participant’s location-privacy according to actual trajectories. Once the protection fails, data leakage will directly threaten the participant’s location-related private information. It open the issue of participating in mobile crowdsourcing service without actual locations. In this paper, we propose a mobility-aware trajectory-prediction solution, TMarkov, for achieving privacy-preserving mobile crowdsourcing. Specifically, we introduce a time-partitioning concept into the Markov model to overcome its traditional limitations. A new transfer model is constructed to record the mobile user’s time-varying behavioral patterns. Then, an unbiased estimation is conducted according to Gibbs Sampling method, because of the data incompleteness. Finally, we have the TMarkov model which characterizes the participant’s dynamic mobile behaviors. With TMarkov in place, a mobility-aware spatiotemporal trajectory is predicted for the mobile user to participate in the crowdsourcing application. Extensive experiments with real-world dataset demonstrate that TMarkov well balances the trade-off between privacy preservation and data usability.https://www.mdpi.com/1424-8220/21/7/2474location-based servicemobile crowdsourcing applicationprivacy preservationtrajectory predictionspatiotemporal markov |
| spellingShingle | Guoying Qiu Yulong Shen Ke Cheng Lingtong Liu Shuiguang Zeng Mobility-Aware Privacy-Preserving Mobile Crowdsourcing Sensors location-based service mobile crowdsourcing application privacy preservation trajectory prediction spatiotemporal markov |
| title | Mobility-Aware Privacy-Preserving Mobile Crowdsourcing |
| title_full | Mobility-Aware Privacy-Preserving Mobile Crowdsourcing |
| title_fullStr | Mobility-Aware Privacy-Preserving Mobile Crowdsourcing |
| title_full_unstemmed | Mobility-Aware Privacy-Preserving Mobile Crowdsourcing |
| title_short | Mobility-Aware Privacy-Preserving Mobile Crowdsourcing |
| title_sort | mobility aware privacy preserving mobile crowdsourcing |
| topic | location-based service mobile crowdsourcing application privacy preservation trajectory prediction spatiotemporal markov |
| url | https://www.mdpi.com/1424-8220/21/7/2474 |
| work_keys_str_mv | AT guoyingqiu mobilityawareprivacypreservingmobilecrowdsourcing AT yulongshen mobilityawareprivacypreservingmobilecrowdsourcing AT kecheng mobilityawareprivacypreservingmobilecrowdsourcing AT lingtongliu mobilityawareprivacypreservingmobilecrowdsourcing AT shuiguangzeng mobilityawareprivacypreservingmobilecrowdsourcing |