引用本文
  • 于海宁,张宏莉,余翔湛,曲家兴.隐私保护的网约出行的研究综述[J].信息安全学报,已采用    [点击复制]
  • Yu Haining,Zhang Hongli,Yu Xiangzhan,Qu Jiaxing.A Survey on Privacy-Preserving Ridesharing for Online Ride Hailing Services[J].Journal of Cyber Security,Accept   [点击复制]
【打印本页】 【在线阅读全文】【下载PDF全文】 查看/发表评论下载PDF阅读器关闭

过刊浏览    高级检索

本文已被:浏览 369次   下载 0  
隐私保护的网约出行的研究综述
于海宁1, 张宏莉1, 余翔湛1, 曲家兴2
0
(1.哈尔滨工业大学网络空间安全学院;2.黑龙江省网络空间研究中心)
摘要:
为高效利用交通资源,在线网约出行(ORH)服务整合车辆供给和乘客请求信息,派遣符合条件的车辆提供非巡游的出行服务。人们在享受ORH服务带来的便利时,也面临着严重的隐私泄露风险。为此,许多研究利用密码学技术设计隐私保护的ORH服务。首先,本文介绍了隐私保护的ORH服务主要面临的城市动态场景下高效计算密态行程开销、实时动态规划密态行程、安全共享不同ORH服务的运力资源等挑战。然后,回顾了欧式距离、路网距离和行驶时间三类行程开销的安全计算方法,其中,欧式距离计算效率高,但误差大,现有路网距离和行驶时长的安全计算方法多数面向静态路网场景,针对城市动态路网场景的安全计算方法有待进一步研究。分析了面向司机、乘客、ORH平台的行程规划问题的求解方法,现有研究往往仅针对司机、乘客或ORH平台的单一目标进行行程规划,事实上行程规划不但要考虑ORH平台自身收益,更要同时兼顾乘客和司机的用户体验。综述了隐私感知的行程预处理方法,单车单客模式、单车多客模式的行程安全共享方法,并总结了其不足与启示。多车单客、多车多客动态模式的行程安全共享有待进一步研究。最后,从城市动态路网下高效的密态行程开销的安全计算与比较、多方隐私保护的大规模密态行程动态规划与安全保障、跨服务域的去中心化密态行程协作共享、ORH服务的法律法规合规保证四方面展望了隐私保护的ORH服务的未来研究方向。本文旨在保护多方隐私的前提下,提高ORH服务质量、促进多ORH服务合作,使得网约出行更加智慧、更加安全。
关键词:  位置隐私  多方安全计算  网约出行服务  行程动态共享
DOI:
投稿时间:2022-04-02修订日期:2022-05-13
基金项目:国家自然科学基金项目(面上项目,重点项目,重大项目)
A Survey on Privacy-Preserving Ridesharing for Online Ride Hailing Services
Yu Haining1, Zhang Hongli1, Yu Xiangzhan1, Qu Jiaxing2
(1.School of Cyberspace Science, Harbin Institute of Technology;2.Heilongjiang Province Cyberspace Research Center)
Abstract:
For efficient utilization of transportation resources, Online Ride Hailing (ORH) services enable riders to hail available vehicles by matching vehicle supplies and rider requests. Along with the advantage of ORH services raises serious privacy concerns. Thus, many studies focus on privacy-preserving ORH services by using some well-established cryptographic primitives. Firstly, we introduce main challenges for privacy-enhanced ORH services under in dynamic city scenarios, including encrypted travel cost computation, encrypted trips dynamic planning and secure ridesharing between different ORH services; Then, we review secure travel cost computation about Euclidean distance, road distance and travel time. Euclidean distance secure computation is very efficient but not accurate. Most road distance and travel time secure computation methods are designed for static road networks, but it is desirable to have more methods for dynamic city road networks. We analyze ridesharing dynamic scheduling oriented from riders, drivers and ORH platforms. Existing works solve ridesharing dynamic scheduling with single optimal objective from riders, drivers or ORH platforms. Actually, comprehensive multiple objectives of riders, drivers and ORH platforms should be considered, such as income of ORH platforms, user experience of riders and drivers. We further summarize privacy-aware trip preprocessing and privacy-preserving ridesharing over encrypted trips, including single driver - single rider mode and single driver - multiple riders mode, and then further point out disadvantages and inspiration of existing studies. However, it is desirable to have secure ridesharing of multiple drivers - single rider dynamic mode and multiple drivers - multiple riders dynamic mode. Finally, we summary further works in privacy-preserving ORH services, including secure travel cost computation and comparison over encrypted trips, multiparty privacy-preserving dynamic ridesharing and safety guarantee over large scale encrypted trips, decentralized secure ridesharing cross ORH services, legal and regulatory compliance guarantee. The paper aims to improve the quality of an ORH service and enhance cooperation cross ORH services, while protecting the privacy of all parties. It can make ORH services more intelligent and more secure.
Key words:  location privacy  secure multi-party computation  ride hailing service  dynamic ridesharing