基于位置预测和缓存的边缘计算轨迹隐私保护方案  

Trajectory Privacy Protection Scheme for Edge Computing Based on Location Prediction and Cache

作　　者：程思亮 刘静 CHENG Si-liang;LIU Jing(School of Computer Science&Technology,Wuhan University of Science and Technology,Wuhan 430065,China;Key Laboratory of Intelligent Information Processing and Real-time Industrial Systems in Hubei Province,Wuhan University of Science and Technology,Wuhan 430065,China)

机构地区：[1]武汉科技大学计算机科学与技术学院,湖北武汉430065 [2]武汉科技大学湖北省智能信息处理与实时工业系统重点实验室,湖北武汉430065

出　　处：《计算机技术与发展》2025年第4期45-52,共8页Computer Technology and Development

基　　金：国家自然科学基金资助项目(61602350);湖北省教育厅资助项目(D20201102)。

摘　　要：在连续LBS的应用场景中,用户的位置数据经常需要被传输到位置服务提供商(LSP)以获取各种服务,然而这会带来用户位置和轨迹隐私的泄露。大多数现有的保护方案使用无第三方或匿名器的系统架构,由于硬件限制这种频繁的数据传输会导致较大的网络延迟,影响了用户体验。为了缓解这一问题,现有的技术方案往往采用边缘计算和缓存策略来处理LBS查询。边缘计算通过在网络的边缘节点处理数据,从而降低了延迟。而缓存策略则允许将查询结果存储在边缘节点上,以便于后续的查询可以直接从缓存中获取结果,无需再次与LBS服务器交互。然而现有许多方案无法充分利用边缘服务器的响应以回答后续的查询。为了应对这一挑战,该文提出了一种基于位置预测和缓存的边缘计算轨迹隐私保护方案。在移动用户和边缘服务器都使用k匿名以及对LBS查询数据的缓存,移动用户使用预测模型根据连续的位置数据对下一次位置进行预测,并对未来位置的LBS数据缓存,增强用户的位置隐私,提高缓存命中率。与CDKA方案相比,该方案具有更高的缓存命中率和更低的延迟。In the application scenario of continuous LBS,users’location data often needs to be transmitted to location service providers(LSPs)to obtain various services,which can lead to the leakage of users'location and trajectory privacy.Most existing solutions use system architectures without third-party or anonymizers,and due to hardware limitations,frequent data transmission can result in significant network latency,affecting user experience.In order to alleviate this problem,existing technical solutions often use edge computing and caching strategies to process LBS queries.Edge computing reduces latency by processing data at the edge nodes of the network.The caching strategy allows the query results to be stored on edge nodes,so that subsequent queries can directly retrieve the results from the cache without the need to interact with LSP again.However,many existing solutions cannot fully utilize the response of edge servers,which can be cached to answer subsequent queries.To meet this challenge,we propose a trajectory privacy protection scheme based on location prediction and caching for edge computing.Both mobile users and edge servers use k-anonymity and cache LBS query data.Mobile users use prediction model to predict their next location based on continuous location data and cache LBS data for future locations,enhancing user location privacy and improving cache hit rates.The proposed scheme has a higher cache hit rate and lower latency compared to the CDKA scheme.

关 键 词：轨迹隐私 k匿名 位置预测 缓存 边缘计算 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]