Protection of Location Privacy Based on Distributed Collaborative Recommendations

In the existing centralized location services system structure, the server is easily attracted and be the communication bottleneck. It caused the disclosure of users’ location. For this, we presented a new distributed collaborative recommendation strategy that is based on the distributed system. In this strategy, each node establishes profiles of their own location information. When requests for location services appear, the user can obtain the corresponding location services according to the recommendation of the neighboring users’ location information profiles. If no suitable recommended location service results are obtained, then the user can send a service request to the server according to the construction of a k-anonymous data set with a centroid position of the neighbors. In this strategy, we designed a new model of distributed collaborative recommendation location service based on the users’ location information profiles and used generalization and encryption to ensure the safety of the user’s location information privacy. Finally, we used the real location data set to make theoretical and experimental analysis. And the results show that the strategy proposed in this paper is capable of reducing the frequency of access to the location server, providing better location services and protecting better the user’s location privacy.在现有集中式位置服务系统架构中，服务器极易成为受攻击目标并沦为通信瓶颈，进而引发用户位置信息泄露问题。针对该问题，本文提出一种基于分布式系统的新型分布式协同推荐策略：各节点可构建自身的位置信息简档（location information profiles）。当存在位置服务请求时，用户可通过邻近用户的位置信息简档推荐获取对应位置服务；若未获得适配的推荐结果，用户可基于以邻近节点质心构建的k匿名（k-anonymous）数据集，向服务器发起服务请求。本策略中，我们基于用户位置信息简档设计了新型分布式协同推荐位置服务模型，并通过泛化与加密技术保障用户位置信息隐私安全。最后，本文采用真实位置数据集开展理论与实验分析，结果显示：本文所提策略可有效降低位置服务器的访问频次，提供更优质的位置服务，同时更好地保护用户位置隐私。